Please check the description if you want links to videos and the Ukikipedia pages featured in this video. i highly recommend checking those videos if you liked seeing those weird anomalies. EDIT: Here's a link to the SM64 TASing and ABC Discord Server for those interested. Many SM64 experts are active there: discord.gg/ECskvyF Also I want to make it clear that cosmic rays have likely caused bitflips in computers like that one election in Belgium. I'm just saying it probably didn't happen to DOTA_Teabag in this specific scenario.
Hey Lunatic! Just want to say that it’s been great watching your channel grow, especially since you’ve had such high production quality right out of the gate. Can’t wait for that 100k!
@@glidershowerI literally wouldn’t go to sleep for like 2 weeks if this happened to me (even though I was born in 07). I had a deathly fear of boos when I was little.
As an aside, cosmic ray interactions are in fact very common when working with some equipment! They're a regular nuisance for Raman spectrometry, for example, as they interrupt your spectra by creating 'spikes' at random wavelengths. Speedrunners, please take care while running Mario 64 on your CCD detectors.
Whoa, I just made a comment talking exactly about raman spectroscopy! The first time I saw that huge spike I got excited! Then after having seen it over and over again it lost some magic.
I'm not very far into this video but just wanted to mention that cosmic rays flipping bits in RAM is not an "out there" concept. Server-grade RAM is literally designed to combat this for data integrity.
Cosmic rays being the cause of this is probably what is being questioned. To your point though, there is also something called "functional safety" in ISO and IEC standards. There are dictations to the way code must be written that assumes a bit could randomly be toggled. An example is that there cannot be any "hanging" or dormant code in which it has the potential of being executed even though it would never execute in any expected scenario. Even if that piece of code is commented out, it cannot be present at all if it doesn't belong there.
@@seinfan9 Nah the "cosmic rays are questioned" is just an excuse, do not follow the video creator blindly, the video clearly showed ways to neglect even the existence of cosmic rays and failed at it, the guy talks about it like it was something impossible and magical while it's actually not that uncommon. I can't believe this video isn't mass disliked due to providing missinformation and confusion just for views and likes, everytime I come back I remember how stupid it is where he shows multiple recreations or possibilities that are already debunked and just says "well cosmic rays don't exist so it didn't happend". Actual moron he is
@@seinfan9 When you compile code, comments are stripped out, so any commented-out code is not only not able to be triggered; it literally is not in the compiled program, and never would have been even in the past. Modern linkers are also good at cutting out dead code, but ISO 26262 might predate that behavior. In either case, that requirement sounds less like it's striving for resilience against bit flips and more like it's striving for the code being easier for a human to understand. That requirement is basically saying, "Keep your workspace tidy and throw out things that aren't useful anymore so that it's easier to organize the things that matter."
I'm an electronics engineer. It's definitely unlikely to be a cosmic ray bitflip, but it could be as simple as environmental radiation or contamination in the packaging of the semiconductor chips, or a variety of other semiconductor device physics stuff. Fun fact: did you know that trace contamination of certain radioactive metals in the solder used to put chips down can cause bitflips?
Interesting. There's a 1996 scientific paper called "Single event upset at ground level" that states that (quoting Wikipedia): "Electrical or magnetic interference inside a computer system can cause a single bit of dynamic random-access memory (DRAM) to spontaneously flip to the opposite state. It was initially thought that this was mainly due to alpha particles emitted by contaminants in chip packaging material, but research has shown that the majority of one-off soft errors in DRAM chips occur as a result of background radiation, chiefly neutrons from cosmic ray secondaries, which may change the contents of one or more memory cells or interfere with the circuitry used to read or write to them."
I've talked to astrophysicists about this before and they disagree, cosmic rays (especially neutrinos) are very capable of flipping bits and should actually do it fairly often (over years duh) statistically speaking. You and an astrophysicist should get together to come out with a truly informed opinion.
Where do you think the classic static noise pattern comes from? Radiation, yea, but a lot of that radiation is in fact cosmic in nature (fun fact, a not insignificant portion of that noise comes from the CMB.)
And yes, before you go ''ackchually'', yes, i am aware modern electronic devices including chips are built to self correct enough to make that noise negligible, but that design does not cover all energy levels of particles (and even types of particles themselves) that the chip will realistically be exposed to.
@@thelelanatorlol3978 About 100 trillion neutrinos pass through your body every second. With 8 billion people and many many many many more electronic devices, a cosmic particle hitting a chip is not rare at all. Just happens continiously.
hey, weighing in with this is a programmer and systems administrator, bit flips happen constantly, be it from heat, radiation or cosmic rays and are usually caught by inbuilt error correction, the n64 used 4mb of RDRAM running at 400mhz on a 9 bit bus, using it's built in error correction bit for the GPU instead. early versions of this RAM standard had no error checking but the standard used a pretty big transistor which would require more energy comparative to the phone in your pocket or your laptop's modern hardware, so a high energy particle is more likely than you may think. changing this bit at the right time seems to me like the most likely cause of the error. weather its ionising radiation from space or because the passively cooled memory was overheating or a solar flare is completely academic and unknowable, but IS more likely to be the case because of the energy requirement to flip a bit in that kind of memory module and it is a compelling story, and one I will continue to tell to comfort people when they can't open their word documents due to data corruption tl;dr we live in a world where ECC memory has made us forget that this isn't even unlikely anymore, it just goes unnoticed because Richard Hamming is an underappreciated genius edit: removed erroneous assumption about hexadecimal conversion because it was incorrect and not really relevant, I couldn't even remember what I meant by it when it was pointed out. I also corrected myself from "a bit flip is more likely" to "A high energy particle is more likely than you might think" because its what I meant and it didn't really make a lot of sense before people have also pointed out that I say in the TL;DR it's because of ECC memory and when they google ECC memory they find the Hamming module and find out that isn't applied to consumer hardware, then assume consumer grade memory has *no* protection from errors which isn't the case, every stick of memory since DDR4 has had a Cyclic redundancy check (CRC) and parity bits on different parts of the memory module, and DDR5 has a sort of on die ECC even in it's "non ecc" variants, the ecc variants send additional parity bits to the cpu for a second check (unless you're on intel, since I don't think their consumer CPU's even support true ECC), which will catch any errors in transport, because the density of a ddr5 chip is kind of insane now, it bit flips constantly. there's also all of this on the non-volatile storage side which as I understand it wouldn't be a problem for this error, if there was an unhandled error in the non volatile storage it would be consistently reproduced, and I don't think it could happen at all in a variable because the CPU creates and references these in memory
also wtf are you talking about when you call those clips too different, they match perfectly, except one is running on native hardware with shitty 90s frame generation
@@gunslingerspartan I think his point is that they don't overlay 1 to 1 over each other, but the outcome shown is completely identical. A rather silly gripe to claim "Not identical"
Fking thank you man. Frustrated me watching this video him claiming "this definitely wasn't caused by a cosmic ray" while citing 0 evidence that it wasn't. When the TAS practically lines up 1 to 1, the bit flip is pretty much what caused it, and again, whether it was caused by a cosmic ray or not is unknowable. But to make an entire video essentially making the claim that "No, it was 100% not a cosmic ray" is infuriating when it's essentially unknowable. ECC memory and error correction is so commonplace nowadays in all software that this essentially doesn't happen anymore, but this did use to happen quite a bit and caused data corruption all the time. It seems like we've forgotten that, goes to show how unbelievably effective these algorithms are at protecting our data from random corruption.
Hi, Cosmic-Radiation-Bit-Flip researcher here (I am not making that up, that is actually my job). So to keep a long story short: computers are weird and trying to model anything happening on hardware exactly is practically impossible. I disagree with the statement that a cosmic bit flip cannot be the cause, simply as it is very hard to prove that it isn't and it is within the realm of (at least theoretical) possibilities. However, there are two other things we should consider that make the cosmic ray theory very unlikely. The first is the very low amount of radiation we observe on earth. A satellite within the Low Earth Orbit will still only get a small number of radiation events in the mission time, for a device on earth the possibility is negligible. On satellites we still need to consider the effects of radiation if hardware is safety critical, we have a very bad setup (high amounts of radiation expected), or we have long missions and aging effects, but that is satellite mission stuff. Secondly, we shouldn't forget how much hardware just doesn't work. Modern computer chips have a huge percentage of the chip that just doesn't work due to manufacturing problems. If you produce a CPU with 12 cores but only 10 work: sell it as a 10 core CPU, that's how it's done. Thus hardware malfunctions are definitely not unlikely and again really really hard to identify, even if you have the hardware in your hands. TLDR: is the cosmic ray theory possible? Yes, it is possible. Is the theory likely: No, it's not likely. Even though it's not impossible, the odds are very low. So we shouldn't say it's impossible, but also not act as if it was definitely the solution.
Well said. this mirrors my thoughts, but it bothers me that people are thinking i don't believe cosmic rays can affect computers when all I'm saying is it probably didn't happen in this specific instance
@@LunaticJI do agree with both of these comments I just think maybe next time use your words more carefully if that makes sense this is just feedback and not a way to make you look bad, just saying it's a myth and it's rare is not best things to say when making an statement, next time maybe not put that people opinions as much and make people look dumb, im not sure it was your intentions and I know everyone has different views just this is what I think. Over all even if I disagreed and thought sometimes you was a bit dramatic this still a good video just needed some clarification.
@@catlance I agree with you. I feel LunaticJ's tone was too dismissive of the possibility. Stating that other explanations are more likely is more than enough to keep the question open and being dismissive diminishes his credibility, weakening his case rather than strengthening it. Another important point is that even very unlikely things happen all the time. With Mario 64 being streamed all around the world every day, we should fully expect to see all kinds of strange outcomes and glitches from time to time, including random hardware glitches, and yes, maybe even bit flips.
@@LunaticJIt's the tone. You sound really... smug throughout this video, and it really comes off as if you are dismissing the possibility entirely. You sound like you are mocking the idea. To be clear, I agree with you that it's unlikely but possible, but your tone was super off-putting.
Shh, keep quiet, I don't want them to find out I've been manipulating my runs for over a decade with that orphaned source I "aquired"... (Just kidding. I don't a have an orphaned source and I don't speedrun either. And if I had one, I'd probably be long dead by now.)
Don't. They'll silence us even more than they already are. Keep your thoughts to yourself. There'll be a time where Big Speedrun will fall and cosmic ray manipulation will be a widely used strat.
Satellite expert here. Bit flips are something we have to constantly design and shield against due to the extremely hostile nature of Space. However, these satalites are orbiting above the protective shielding our earth and atmosphere provide. On earth, bit flips are significantly rarer. EXCEPT! Our sun goes through an 11 year cycle of rising solar activity (solar flares, sunspots, coronal mass ejections etc.) Eventually culminating in the "solar maximum" before activity begins to decrease again. During the year or so before and after a solar maximum, computer issues caused by the sun are slightly more common. The last solar maximum was in 2014, meaning this run took place during a period of extremely high solar activity. Btw, the next solar maximum is building up and will be here by 2025, and experts are prediciting that it's going to be a rough one.
Even on earth, cosmic rays can cause hundreds of megabytes worth of RAM data to be scrambled every month. Of course, most modern hardware can comfortably deal with single bit flip events, cosmic ray or otherwise.
A coworker from my first job was testing a component that would end up on a satellite running our software. He would often start running tests on the device, then work on something else while those ran, and would later see that some random test failed. But he couldn't figure out why. He tried catching the failure by running the test suite with a debugger attached dozens of times over night, but came back in the morning to all the tests being green. So this was an enormously frustrating thing to deal with. How could the tests fail only when he was in the office? So he tried running the tests in the debugger once again in the morning, and got up to go get a bagel. But the debugger immediately paused on a failed test. It turned out that the slight electromagnetic impulse from his chair whenever he stood up/sat down/shifted his weight was enough to flip bits in the device's cache/RAM and cause a test to fail. With additional EM shielding the failures disappeared. The local environment can be a much better place to look for sources of spontaneous memory corruption than cosmic rays, though cosmic rays can cause rare bit flips on earth's surface.
Dram engineer here. My job is literally to debug these type of silicon bugs, but there isnt really enough info to determine what is failing in the memory chip. Could be capacitors losing charge due to lack of refresh cycles, unintentionally causing a row hammer attack, bad voltage reads on sense amps... Most likely, software issue as a bit flip in these cases could happen to any of the million bits in the array. Its weird how the bit flips would only happen to player coordinate.....
That is my impression too. If a bit has flipped somewhere in memory, the chances are good that it would just crash the game because it tries to read from different memory than it otherwise would, messes up a return address or a stack or something. The way that it seems to only affect the position of Mario is very strange and suggests that this is the result of a bug, not cosmic rays.
Huh, I'd never thought the community had the idea "It conclusively was a cosmic ray", but rather "We can't quite figure out what happened, and a cosmic ray was a possibility, since we can't figure out how else it might have happened yet."
yeah this wouldn't be the first time something similar happened and was observable i dont remember when exactly but in veritasium's video he talked about the small radioactive particles in intel chip's ceramic caused a single flip of bit 12 in a register which caused one of the candidates for an election to receive 4096 extra votes (noticeable and while it wasn't a cosmic ray the particles emitted from the uranium are the same)
14:00 Mario's position is stored as a floating-point number, which is stored vastly differently than an integer. A bit flip in the mantissa would change the value by a power of two, but a bit flip in the exponent would multiply or divide the value by a power of two. If the Y-position were, say, 1800, a bit flip in the lowest bit of the mantissa could change the value to 900, a difference of 900.
the part about Veritasium "using the article and video as an example of a bit flip definitely happening" in his video all he says is that no one has been able to replicate it and a bit flip from cosmic rays is the "best explanation anyone can come up with". does not seem like he was claiming it definitely happened to me.
If there is any channel I trust to do the proper research it is Veritasium. One of the most prolific scientific channels on the platform that routinely has experts in their field on to discuss their work.
While watching this video, I couldn’t help but grow annoyed by Lunatic constantly attacking the theory. He didn’t even provide a theory of his own. The idea at the time is that cosmic rays were the explanation that just made the most sense. And it’s true; it did make the most sense at the time. Aside from the cartridge or console just being faulty. And even coming out of the video, it isn’t exactly firmly debunked, either. Unlikely, but was it definitely not a cosmic ray? We don’t know.
@@richboy455hate to break it to you, but every other video Veritasium makes is an explicit sponsorship where the science is secondary to promoting the product. Not to say that this makes these videos completely inaccurate or anything, but you should be critical of the media you consume and shouldn't put blind trust in anything on UA-cam, especially in the pop science sphere
@@richboy455unless he is being sponsored to do the video. The one about dandruff, the one about self driving and the one done with Mercedes are utterly corporate propaganda nothing more.
"I refuse to believe it was cosmic rays because that's just too unlikely. It was definitely one of these other options that don't usually lead to bit flips or that we tested but didn't lead to a bit flip"
The purpose of the video is to say that we don't have an actual answer to that, there are lots of theories and that cosmic rays is the most improbable. That's literally (not using literally as an emphasis) the thesis of the video.
@@mommyukiExcept it's not. A cosmic bit flip literally affected an election before why is it so unlikely that it affected a speedrun out of the billions of speedruns
A bit flip is perfectly plausible, and in my opinion the most likely. The TAS recreation doesn't match perfectly, of course, but the timing is close enough that a bit flip should have caused that kind of displacement - I don't understand why he completely discards the hypothesis because of a few pixels in difference on what was supposed to be a *recreation* of the event and not the event itself. And with that, I think what actually caused one of those bits to flip is entirely irrelevant. There's a huge array of things that can cause that, especially on old hardware like n64s which don't have as much protection against this sort of malfunction.
@@mommyukiAnd he arrived at that conclusion using absolutely 0 credible arguments. Provided no alternate hypothesis and rejected the currently accepted hypothesis because it’s unlikely. Completely unscientific
I honestly have no idea why people keep saying that, although i still feel 2013 like yesterday. Technology wise, because i always go for low-specs, so when i try something that isn't, i'm amazed at technology's progress LOL.
There was an election in Belgium in 2003 where a candidate got 4096 extra votes (2^12) due to a cosmic ray bitflip. It was found and corrected, but strange and unlikely things happen given a large enough sample size and period of time.
@@Dante02d12 i can see why you would guess that, but circumstances dictate otherwise. It happened within the machine and was immediately spotted and fixed; she was among the least popular candidates out of 10ish, and as a minor candidate on a local election (iirc it was about 7000 votes for the whole race) had no ability to control the machines.
@@thethree13o epic gamer strats, bring 10 bajillion copies of a game into space, run them all at once, and keep them running and restarting until eventually you get a world record time
How can you possibly look at that TAS bit flip and not think that is EXACTLY what happened?!?! The only reason it's not a perfect 1-to-1 match is the not-perfect starting positions. But the height climbed is literally identical and Mario lands on the platform at the same time. Sure it may not be cosmic rays, but there's no way that wasn't caused by a bit flip. Also, with the downwarp later in the video. Why can't it be a bit flip? In floating point numbers there are ways to get non-power of 2 shifts (flips in the exponents), or there could be a frame correction where Mario's position is shifted because otherwise he'd be in the middle of a block or the ground etc. Bit shifts happen all the damn time, and for so many reasons. Simply having a bunch of other bits changing around one can have effects. It could well be that this bit location is in a hotspot location, and then cosmic rays pushed it over the edge. Every tech company realizes that cosmic rays can flip bits, so while it is true that the myth was never solved or proven, it is definitely not ruled out.
I don't think anyone's arguing that it isn't a bit flip(i hope not anyway), I think they're more arguing that it wasn't a cosmic ray. BUT I think its decently likely- not really a cosmic ray, but radiation of some kind.
Im not sayung it's cosmic rays but i just wanted to point out that cosmic ray bit flips are insanely common (unless you use ECC memory, they still happen but they get corrected 99% of the time). Like they happen to computers at an hourly rate, and if you gain elevation above sea level they massively increase in frequency. Aviation and Space deal with them by having so many redundancies and protections that the bit flips dont matter much. But a Nintendo 64 has no such protection, its bits will be flipped. Its not a matter of if they will happen but when. But this rarely (as in on an individual basis, on a provincial/state level they have caused issues all the time we just account for them nowadays) results in much issues let alone a bit flip like we saw. That said, this isnt a fact of the situation; cosmic ray bit flips may have had nothing to do with it. But the fact that the height change from one bit flip was nearly identical it feels safe assuming its a bit flip of some kind. But i wont say it as a fact since there is no such thing as 100%.
The electrical surge segment reminded me of a clip from Runnerguy2489. The framerate in his game chugs briefly at the exact same moment that lightning struck outside his house.
EDIT - Never mind, just saw the clip and it's a stream lag rather than a game lag. Oooh that's interesting - iirc the N64's CPU runs with a 1.5x multiplier on the base clock frequency. If Pin 112 is 3.3v, then it runs at 1.5x, however if it's grounded then it runs at 1x. I wonder if the strike somehow dropped the voltage enough on pin 112 that the CPU dropped to it's 1x base frequency, which resulted in the framerate chugging.
@@kargaroc386Just saw the clip, looks like it's a stream lag rather than a game lag unfortunately :( You can see the webcam freeze too, and immediately after the unfreeze you can also see a slight flash in the door frame.
@@Jademalo the N64 does have a rare bug in its clock generator(?). Maybe once every several thousand times you turn it on, it runs at half speed. The audio is all slowed down and spoopy and the video is a weird static (TV not properly syncing to a very-off-time signal). There are a couple clips of it on Twitch. While that evidently didn't happen here, I wonder if it ever could happen while the game is running? It wouldn't be helpful at all, though.
Okay yeah, but also the phrase: “Sun assisted speedrun” goes really fucking hard. It’s highly likely that it’s not true, as you’ve already mentioned, but god it sounds cool.
Just like aliens*, big foot and whatnot, we may keep making memes about that while ensuring people are informed that they're not real. *Aliens are real; but they're nowhere near Earth and we have yet to find any.
Am i too drunk or did he talk abt how flips don't cause displacement then shows a clip of flips displacing a piston, then saying that rays are a slim possiblity and at the end saying that it flat out wasn't reys because the mystery isn't solved? Like if the experts couldn't exactly replicate it without flipping (which is what rays do) and the only perfectly matching video is flipping then wouldn't it mean that it was rays? Since even the hardware wasn't faulty Again i may be too drunk but this video is not making sense to me
He never said flips don’t cause a displacement. A bit flip was probably the reason why the upwarp happened, but it’s MUCH more likely that it wasn’t cosmic rays. Think about how many bits are in an N64. 4 megabytes of RAM, 4-64 megabytes of storage. Assuming it was cosmic rays, how likely do you think it is that they somehow hit exactly that one bit at that time? Do cosmic rays flip bits constantly? No. So it’s incredibly unlikely that cosmic rays flipped that specific bit at the perfect time, especially considering the evidence of cartridge slapping corrupting the game and how he typically had to put in the game cartridge in a weird way, implying that his hardware could be the issue.
@@somenerd8139 1) You're assuming cosmic rays didn't flip other, less noticable bits. 2) You're assuming they don't do this all the time. 3) You're assuming that because a bit flip has meaning to us as human beings, it means that it is a more special and meaningful action than a bit flip that wouldn't be as noticable. 4) You're ignoring every other recorded SM64 playthrough where this DOESN'T happen, and focusing only on the exceptional playthrough where it did. 5) Why didn't the cartridge slap/corruption have many more noticeable effects than just a single upwarp?
Until I see definitive proof that outside rays didnt cause a bit flip, I'm sticking my N64 in a microwave at the start of TTC and hoping for good IRL RNG.
What you didn't mention is that most of the articles and some of the memes say the warp "saved time" or would have saved time, but it doesn't even do that. He was going for the red coins and it only moved him away from them and would have forced him to move back down to where the platforms were
Yeah the whole reason there was a bounty in the first place was for future use in A Press challenges or speedrun categories. The glitch is still in the first phase of speedrun metamorphosis but articles are treating it like a crazy time skip for a wr pace or something.
I really don’t understand why that is in the video. Does the editor think that a food processor/blender has significant electromagnetic radiation? Does bro get his food processor from aliens?
"The idea that a cosmic ray flipped a bit in a computer is highly unlikely." And yet, ECC memory is used in servers all around the world for... what, fun?
Yeah, insane to just rule out cosmic rays like that. I remember going about universities when I was younger, and it was almost guaranteed there'd be some demonstration about detecting cosmic rays with some spark chamber so you could visualize them.
@@plukerpluckit is so funny to me that people look up at the giant ball of flames in the sky that warms up this entire planet despite beings millions of miles away, one of literal infinite more of them in varying sizes from massively smaller to way bigger and think “nah nun’ gettin here”
There are simply way more common causes of it than cosmic rays. The N64 is old hardware with big transistors that take lots of energy. There are a million electrical issues that could have caused it. Hell ECC was first invented to handle the reading of punchcards and cosmic rays weren’t relevant.
Thought this was gonna prove the bitflip theory wrong but it just goes over all the likely theories and in the end the bitflip is the only one that has had any sort of evidence that could be reproduced, and you even bring up the dupdome instance which only reinforces the idea of a bitflip caused by a cosmic ray. In dupdome's scenario it is far more likely that a hardware error was in effect, as there were multiple bitflips occurring in predictable ways, but in the dota_teabag scenario we've taken apart the N64 and cartridge itself and in your own words "nothing notable popped up." Faulty hardware can be ruled out completely due to those tests. As for cartridge tilts and slaps, nothing in the dota_teabag clip indicates any sort of sudden harm to the console like in the atogami clip. In every other example shown of faulty cartridge connections, the interference is far greater than the upwarp and anomalies occur for far longer, animations and textures glitching out for several seconds at a time that isn't in line with the sudden, one frame change we see in the dota_teabag clip. Every alternate scenario you've presented has other caveats and such that make them differ too greatly from the dota_teabag clip. Faulty hardware has been disproven by further testing. Faulty connections cause wildly different scenarios from the one we are trying to recreate. The bitflip is the ONLY scenario that has shown a correlation to what actually happened in the clip. It is the only scenario that actually hits the mark, and no matter how unlikely a SEU is, the results suggest that it's still more likely than anything you've shown. Hell, the idea that cosmic rays are "unlikely" is even questioned in your vid. You say yourself that bitflips usually don't effect a computer in "a visually observable manner." This implies that they happen more often than we have observed, thus making our own perception of how likely they are to happen to be inaccurate. Which, of course, they do. Hardware and software usually just has ways to handle errors that the end user doesn't perceive, like simply crashing. I'm not gonna argue that poor journalism leads to myths, but if YOU are, then you should be doing your due diligence. You said that Veritasium used the clips as an example of a bitflip "DEFINITELY happening thanks to a cosmic ray." This is objectively incorrect and can be proven wrong by simply watching his video. His actual words were "the best explanation anyone could come up with is that a cosmic ray caused the glitch," which is a perfectly reasonable take that still leaves room for doubt. If you're gonna criticize him for poor journalistic integrity, criticize him for getting pannen's name wrong, don't shove words in his mouth. The fact that you're willing to put such blatant misinformation in your video calls in to question your own integrity. tldr, bitflip is still the most likely scenario as nothing else has come close to replicating it. Cosmic rays affecting hardware is more common than we think. You should stop making up quotes to make others look bad.
Perfect description of what's wrong with the video IMO. Essentially saying 'the cosmic ray bit flip theory "simply isn't true" (his words)' while providing evidence that a bit flip is the most plausible theory for that exact glitch is wild to me. Trying to call out bad journalism through _this_ video of all things is just peak irony.
Adding another comment here to try to push this comment to the top. People tend to open comment reply sections with more replies, and upon reading the replies are psychologically more likely to actually read the long original comment, and thus more likely to like said comment, pushing it further to the top, where more people can see it as it rightfully belongs to be.
I think this video adequatly covers ways other than cosmic rays this could have happened, however, it does not adequatly disprove the general idea of a bitflip to a degree that justifies the language used.
@@paytonhearn2502 that’s exactly how science works. If there is a likely solution backed by evidence(the faulty console) and someone comes up with another solution which is orders of magnitudes less likely, the science community doesn’t go „we can’t determine which one it is because both are possible“. They go „There is a very small change it _did_ happen like that, but the general consensus is that it happened like this“. If it wasn’t that way we’d have to consider that gravity might not exist because there is a possibility that all things just fall downwards through random quantum tunneling.
@@foximacentauri7891 What evidence? And how do you define which solution is more or less likely? If I was paying attention, the video said that upwarps have not been proven to happen because of faulty consoles. If that's the case, it's not obvious to say that faulty console is more likely source just because it can create other types of glitches. At least the bitflip has some kind of experimental evidence to back it up since the TAS recreation was close enough to not rule it out(i.e it was *really* close to what we saw). So we have one hypothesis about cosmic rays that can and was tested and it resulted in good but inconclusive evidence, and another hypothesis which has no evidence and is untestable and is basically just "faulty console can result in any kind of glitch, so it's probably that even though we don't know how". Either one could be the reason, but neither is "orders of magnitude less likely." because we lack data to make such claims
this happens in science journalism all the time. A journal is released where the author of the paper states that the study MIGHT show [xyz] to be true, and then people go around saying that [xyz] is DEFINITELY true and is PROVEN BY SCIENCE.
Doesn't help that terrible science education has left the vast majority of the population believing that science "proves" that things are "fact", rather than just being the process by which we eliminate impossibilities through repeated observation and analysis.
Bit flips caused by gamma rays happen much more often than you would think. The term "cosmic ray" used in English is offputting and makes it sound outlandishly unlikely. "Background radiation" is more accurate as it originates from space, earth, radioactive decay or any other thing producing exotic particles. Try turning on a geiger counter at home and see for yourself. Just because something sounds unbelievable to you doesn't mean it didn't happen.
The guy didn't even review the game, and even had positive impressions of it. Yet here we are years later using him as the frontliner example of "bad gaming journalism" because he was bad at a video game once
@@supra_sr Also worth noting that he wasn't a "regular" (for lack of a better word) gaming journalist, he mostly covered the industry-side of gaming, not gaming itself.
6:00 it's not nearly as rare as you think. just, most systems are memory managed and almost 0 bits of your memory are system critical so you don't ever notice. I've actually tested this with older ssds and have a few examples of where a single bit of a bios rom chip got flipped and disabled a whole computer. lucky for dual bios and flash back.
this is the biggest fault with the video imo. the experts that are cited are always speedrunners or tas engineers, never a person who is actually an expert on gaming hardware or storage or physics. if a bit flip seems plausible, then why not leave the video at "this is still unsolved" instead of "cosmic rays are impossible to have caused this"
@@angrymurloc7626Thing is, based on the other TTC anomalies experienced by a different person, cosmic rays being the cause gets less and less likely, bordering on being impossible
@@angrymurloc7626 He stated it was possible multiple times throughout the video, he was just rejecting the games journo myth that it was conclusively proven by a guy who's name they couldn't even get right since they're all too busy chasing clicks to worry about making sure they're not just repeating the same falsehoods.
I didn't really mention that when I was pulling bios off of boards to check them against clean bois install, I regularly found dozens of flips per year running. most of them were inconsequential but I'd be willing to bet the 20yr old problem of hardware just not working one day is somehow related. a large portion of your bios is just channel controllers for hardware.
This video really frustrates me. It keeps saying that cosmic rays are bad and implausible because of [some reason] (probably because it's fancy physics), so some mysterious other error has to be causing it. while i myself think that it's more likely that it's just faulty hardware flipping a bit for some electricity reason (N64s are old), it's just not rigorous or well explained at all. there's not even a single number on how rare cosmic rays altering bits on earth is, how often N64s fail, etc., there's nothing. it's just a cycle of "here's something kinda similar that happened that dosen't look like a bit flip" into "experts think it's dumb" repeat.
Yep. It is CLEARLY a bit flip from the TAS replication. I can't see how you could look at the replication and not think that. At which point the only question is what can cause bit flips, and this video basically doesn't go into that detail at all. Bits flip occur because of hardware degradation, electrical interference, random imperfections in the chips, or radiation. The last of those could be cosmic rays, which is one of the more common ways bits are flipped, and is the reason that all enterprise level hardware uses error correction to detect and fix those flips.
I like how you applied a double standard in how you use the fact that a bit flip didn't match exactly as evidence that is wasn't that while using equally dissimilar events as evidence that it was something else.
I have no idea why the SM64 glitch happened (the cache thing makes sense to me), but I don't know who told you that cosmic rays flipping memory bits is "rare". High performance servers, the main reason you pay extra money for error correcting RAM is in no small part because cosmic ray bit flips aren't terribly uncommon, particular if you have a lot of RAM. I'm not saying they happen all the time, but they absolutely do happen, and if you have data that you want to protect you need to actually consider it. Data centers use error correcting RAM specifically for this . Now, obviously, the N64 *does not* have a ton of RAM, but considering how many millions of hours have been spent playing SM64 it feels like a cosmic ray bit flip at some point is inevitable, considering that the N64 doesn't have error correcting memory (I'm not even sure it had been invented yet!).
It hadn't been invented yet because bitflips weren't as common. Memory nowadays is built on a much smaller process node, where less current is needed to flip a bit. More efficient, also more susceptible to bitflips from cosmic rays or quantum tunneling. Even non-ECC memory has some error correction nowadays.
I really don’t think you know what rare means if you think saying “they don’t happen all the time but they DO happen” somehow contradicts the claim that it’s rare. You’re basically saying “it’s not rare but essentially it’s rare”
@@zzodysseuszz no, I am agreeing it is rare. I am claiming that given enough hours of playtime an unlikely event becomes basically inevitable. The video had the tone of “it’s rare so it didn’t happen”, and I am arguing that even if they are rare, there have been untold millions of hours playing Mario 64, on systems that don’t have error correcting memory, so it wouldn’t be *weird* if it did happen eventually. Law of large numbers and whatnot.
@@zzodysseuszz The claim in the video is that it's practically impossible for a cosmic ray to flip the bits in an N64, despite the fact that it's a very well-known phenomena that happens plenty of times on other kinds of computers.
I will weigh in and say that it's extremely unlikely for a faulty cartridge read/cartridge tilt to cause an up-warp. There's a reason those glitches in games are often limited to animations, models and sounds/music freaking out. Cartridge tilting tends to not affect things that are always loaded in memory, such as the game's code and its associated memory. It would be very unlikely for that to be the cause of an upward warp. That's an effect you would only see if you used a ROM corruptor for amusement. I wouldn't dismiss the bit flip theory entirely, for that reason. The only other viable option at that point is some sort of undiscovered bug/quirk in the game's physics. I'm not saying the bit flip is necessarily caused by a cosmic ray, but it's exceedingly plausible that a bit was flipped to make that happen.
DOTA_Teabag made a YT video(a YT short for some reason?) showing the game crashing with Mario doing the classic corrupted animation with corrupted music. So it seems like hardware issues is(or are?) the culprit.
Yeah I don't see how tilting the cartridge would affect the mario position variables stored on the N64 ram. I mean I guess hypothetically it could read a faulty instruction of the cartridge that would do that, but that's VERY unlikely.
@@Sad_Cirno Except it doesn't. Logic on N64 cannot be executed directly from ROM, it must be loaded into RAM first, and if you don't set up certain memory segments correctly, only the first megabyte of RAM can ever be used for executable code. Additional logic is never loaded in the middle of a level, only during transitions.
The actual impossible hypothesis is that it was a cartridge tilt. Mario's height is stored in RAM, not ROM, and is only modified by code that was already loaded at that point.
I literally stopped watching the video to see how more than 90% of the comments against argued in a polite and plausible way, I have rarely seen this, epic
I'd rather live in a world that mysterious waves of energy randomly travel across the universe to flip bits in my electronics than a world that doesn't
Good news for you, this guy has literally no idea what he's talking about and didn't disprove cosmic ray bitflips as a likely candidate in any way. To be honest he deconstructed his own argument, which was weird, by showcasing that a bitflip DID happen (The footage, despite his proclamations, matches perfectly. Of course when you overlay them they look different because 1. one is on an actual n64 and 2. the inputs leading up to the bitflip are different.) but also by showing that in competing theories there tend to be a LOT more bugs throughout the run than a single bit flipping.
@@EntitySteel no, he basically just has a problem with everyone accepting the cosmic ray theory as fact when 1) it wasnt ever proven or solved, and 2) there are MUCH more probable causes that no one gives credit to, simply because cosmic ray sounds cooler.
@@dogsbecute Except no one's accepting it as fact, and he's already considering it a "myth," simply dismissing the idea while refusing to acknowledge that it's neither been proven nor *disproven.*
@@yeetrepublic9142 .... im sure we didnt watch the same video, because he certainly acknowledged its existence and the probability compared to literally every other example he gave, as well as why it isnt so cut dry. Also, yes, people are literally taking it as a fact and are malding at the possibility that maybe...JUST MAYBE...it wasnt actually a cosmic ray flipping a bit. He even went into depth explaining different technology advances to prevent such a thing, which literally acknowledges its existence.
The repeated disregard for cosmic rays as a possible reason is unscientific, given how much they happen all the time. I suggest going to a lab that shows it in action, its actually quite fun, but on the more serious side, space agencies definitely have to account for it in their computers, where coding errors are accounted for and auto-repaired to insure failure is minimized as much as possible. The rarity that the rays specifically hitting computer bits is fine, given it's not just likely, its statistically guaranteed when enough people play it, especially for many hours, days, years.
Definitely a case of “technically possible” getting warped into “100% confirmed”. But look, either I’m an idiot who accidentally put a dirty bowl back in the cupboard, or someone broke in and put a dirty bowl in the cupboard. Both are possible, guess we’ll never know.
@@gloweye Bit flips from cosmic rays are very common. They just don't usually result in much as modern computers have ways for correcting the flip. Especially if you go above sea level, aviation and space get it the worst and it used to be a constant problem.
Copied this from a blog post by some guy (john d cook). This incident is well documented: . . . . Radiolab did an episode on the case of a cosmic bit flip changing the vote tally in a Belgian election in 2003. The error was caught because one candidate got more votes than was logically possible. A recount showed that the person in question got 4096 more votes in the first count than the second count. The difference of exactly 212 votes was a clue that there had been a bit flip. All the other counts remained unchanged when they reran the tally. It’s interesting that the cosmic ray-induced error was discovered presumably because the software quality was high. All software is subject to cosmic bit flipping, but most of it is so buggy that you couldn’t rule out other sources of error. Cosmic bit flipping is becoming more common because processors have become smaller and more energy efficient: the less energy it takes for a program to set a bit intentionally, the less energy it takes for radiation to set a bit accidentally.
This video is made well, however the arguments used dont really hold up. The main things we know are this: -Mario teleported up, tests have proven that this is almost certainly caused by a bit flip -the cause of the bit flip is unknown -cosmic rays can cause bit flips So really it could have been a cosmic ray, it also could have been a different interference. The fact is that no one knows how it occured but the cosmic ray theory is possible, just quite unlikely. But your point seems to be that it couldnt have posssibly been a cosmic ray, when it would make more sense if the point of the video was to discuss how the cosmic ray theory, while still being one of the possible causes, is not the only possible cause.
The bit flip from C5 to C4 actually looks quite convincing. The footage not matching 100% could be caused by a mismatch of input or velocity (only mario's position would have been tampered). However there can be several causes for a bit flip. I like the general spirit of the video, reminding everyone that we have never been able to tell for sure that it was a bit flip and that it was caused by a cosmic ray. However I think it goes a bit too far in the opposite direction as well. Cosmic rays are very real, even on Earth. In fact, that's one of the reasons why we use Error Correction memory sticks (ECC) on servers.
The way he talks about cosmic rays is like that of a person that doesn't belive the earth is round and that there are other planets in space, he is just neglecting the existence of something very real and pretty "common", I feel like he just reads "cosmic rays" and thinks it's some weird sci fi shit
@@superbeta1716 It kind of reminds me of this low-quality youtube channel about an ichthyologist (fish biologist) whose entire pitch is basically "if there's something cool that lives in the water, it doesn't exist"
The marginal difference between the C5 to C4 bitflip and the speedrun event in question could easily just be caused by the faulty cartridge. So say a cosmic ray DID hit this man's unique, janky, half-tilted game... the C5 bit flips to C4, and we see slightly different results because of slew of other possible unique oddities/issues going on within that decades-old faulty hardware. Whether or not the bit was flipped by a CR or some other event is fairly irrelevant, but at the end of the day Cosmic Rays are the only easily-accounted for phenomenon. You KNOW the earth is constantly being bombarded by cosmic rays, so you KNOW they were present and while unlikely to strike at that exact moment, they very well could have been the culprit despite the statistical odds. Meanwhile, everything else is purely speculation. DID his faulty N64 have a fault that would cause the event without some outside interference like CRs? We don't know. Was it simply have been some electrostatic discharge from nearby construction? We have no evidence for that specific scenario. Was it a simple power spike or electrical anomaly in the home or device? Again, we have ZERO conclusive evidence to suggest that was a factor. But do we have absolute proof that cosmic rays were being blasted at that area of the Earth at that time? Yes. Because the same is true for literally every square millimeter of Earth's surface at all times! So while it's not likely, we DO have solid evidence that it's a DISTINCT possibility. And as such, it can't be ruled out.
@@TheKhopesh Another interesting piece of information related to this is a 1996 scientific paper called "Single event upset at ground level" that states that (quoted from Wikipedia): "Electrical or magnetic interference inside a computer system can cause a single bit of dynamic random-access memory (DRAM) to spontaneously flip to the opposite state. It was initially thought that this was mainly due to alpha particles emitted by contaminants in chip packaging material, but research has shown that the majority of one-off soft errors in DRAM chips occur as a result of background radiation, chiefly neutrons from cosmic ray secondaries, which may change the contents of one or more memory cells or interfere with the circuitry used to read or write to them."
The footage you showed comparing DOTA's run to a TAS isn't even identical before the upwarp occurred, and the only difference after the bit flip is a slight change in the camera angle. The fact that, in both cases, DOTA and Panen reach the same height and hit the ground at the same time should be clear evidence that a bit WAS flipped. Even if it wasn't cosmic rays, the height that DOTA got in his speedrun was exactly the amount you would get switching the leading byte of height to from C5 to C4.
Thank you. Personally, I also dont find it too far fetched to think that with the millions of computers/consoles etc. running all the time, a cosmic ray could hit a person playing a game. Yes its improbable, but not impossible
"The clip of the bit switch and what happened is actually slightly different so it couldnt have happened" he says as the two clips match up perfectly except for the frame interpolation and fails to give anything else than even resembles the clip.
anyone claiming it is 100% not a cosmic ray is just as infuriating as anyone claiming it 100% is a cosmic ray. You can say it probably is or isn't all you want. But if you say 100%, then any slight probability of the opposite is enough to disprove the thesis.
Also, something i just thought of while reading other comments, who's to say it had never happened before DOTA? The vast majority of people playing Mario 64 for at least a few years if not longer would have been kids with either no way or no interest in recording. Just because we haven't seen it on camera doesn't mean it hasn't happened to some kid in 1997 who thought it was just a weird quirk of the game.
yes I agree, and what‘s bitter is that in the video itself he says it. We now still don‘t know what exactly caused it, there are many factors including the cosmic ray "theory", and like he said:"that a cosmic ray flipped a bit is highly unlikely", but then proceeds to at the end say that it is just false (so 100% untrue).
>complains about people taking "it could've happened" and making it "it did happen" >takes "it probably didn't happen" and makes it "it didn't happen" >refuses to elaborate >ends video
Let's be real though A cosmic ray traveling 8 light minutes through earths electromagnetic field into some guys house and into a very specific part of a circuit board at the exact perfect time causing mario end up on a higher platform isn't very likely.
@@lol-de4lo That's like saying a dart hitting an exact spot down to the millimeter isn't that likely. But throw a dart in hindsight, and wow, look at that. At exactly 1:13:004s I hit exactly 10.413cm by 5.143cm on the dart board, even though the probability of hitting that exact spot should be impossible!
@@lol-de4loYou're missing the point. Your devices get walloped by cosmic EM radiation all the time. Radios have static for a reason. The only reason computers don't glitch all the time is that they do, but programs handle errors like that all the time. And when they do fail, it just crashes. The fact that this glitch was blatantly visible is the only difference.
The BitFS downwarp being around 900 units doesn't automatically rule out a bit flip, because Mario's position is stored as 32-bit floating-point numbers, meaning the highest bit represents the sign, the next 8 represent the exponent, and the remaining 23 give the mantissa, to make up a number of the form (basically) mantissa * 2^exponent. Flipping one of the mantissa bits will add or subtract a power of 2, but flipping one of the exponent bits will *multiply* the previous value by some power of 2, meaning if the Y coordinate started as -900 and the last bit of the exponent was flipped from 0 to 1, it would become -1800, resulting in a 900-unit downwarp. I know the lower sections of BitFS are below Y=0 (because of the A press save on Wii VC) but I don't know if the coordinates perfectly match the clip
Exactly what I was thinking, he makes it sound like it's been for sure disproven but people still don't know what caused it. Went back and watched the Veritasium video and he does a great job at saying it's just a very likely scenario not a proven fact. The guy in this video act's like cosmic rays flipping bits is some outlandish thing but yet if he actually watched the Veritasium video he's criticizing he'd know it actually happens way more often than one would think, normally just not in such a visual way. I definitely agree that a lot of video game journalist do a very poor job at getting the facts right and it sucks, but calling this the biggest myth in speedrunning history when it's still a very plausible theory is just disingenuous.
The chances are very low, but it can't really be fully ruled out. It would be amazingly lucky for it to happen at that exact moment, but not impossible.
@@blueberrimuffin6682 Iirc, it was pretty damn close, no? And couldn't that be up to user error? Even for someone so experienced, I feel like it would be hard to replicate it perfectly. Unless there's some more tech I don't understand- I'll admit I'm not very in to this space.
You can't prove it wasn't a bug in the code either a ton of people looking at it and finding no such bug doesn't mean there isn't some insane edge case no one missed .. it's quite hard to prove something "wasnt" something
@@LiEnby thats the point. its an unprovable occurrence. it COULD happen. and also, i dont wanna be that bitch, but they HAVE decompiled the source code to this game. so, get a lookin :3
1. A bitflip seems to be the most plausible explanation to the upwarp considering how well it matches the movement of mario. 2. There was no better explanation given in the video. Just speculation about faulty hardware with no technical details as to how they could have caused a bitflip. Also all the examples mentioned about funny things happening in Mario 64, that cannot be explained by a bitflip are completely pointless. We all have experienced one glitch or another even casually. We know gaming systems aren't perfectly designed. 3. As for the reason the bit flipped: I don't know. Since there were similar discoveries on Tick Tock Clock, maybe the answer lies within this level. However I want to point out, that cosmic radiation is not as far fetched a theory as it is made to believe here. Especially if you assume this being the only instance in recorded Mario 64 history where the only good explanation was a bitflip. Since it could happen at any moment in time, even if it is highly unlikely, after collectively pouring enough hours in, someone somewhere in the world would experience a bitflip with visual effect sooner or later. There could be happening many we wouldn't notice until finally one did. Simply put: how likely is it to never ever happen to anyone recording Mario 64? Should we be dismissive when it eventually does?
does comic rays explain why bit errors only recorded in tick tock clock so far 4 fucking times instead of the other like 23 other stages and openworld that are likely played much more. Sounds to me like it's not fucking cosmic rays flipping bits it's just tick tock clock.
@@TheKingOfApples100 bro I've seen you comment this exact comment on like four people's replies. Are you okay? Like don't get me wrong I kind of agree with you to a degree but that is a really aggressive copy paste on such an unimportant topic.
@@papascronch so people responding to see it if you don't reply to the person's comment they have no reason to click on the video again this way they see it if they have notifications
@@TheKingOfApples100the comment you've responded to doesn't mention cosmic rays, just that it could have been a bit flip, which could happen due to reasons other than cosmic rays
If anything this video is a good argument FOR the cosmic ray theory (well the bit flip theory, as editor points out there are alternative causes of a flip). The examples of loose/dirty connections all lead to glitches caused by misreading of data from the cartridge which is different to the upwarp glitch. Hitting the cartridge when falling off the slide caused the system to load the wrong level's address for the painting. Bad connection causes audio/video data read by system to be garbled. Neither of these are similar to Mario randomly either gaining incredible speed for a tick or warping position, where is the memory read that would cause that. The examples of similar warping/glitches on that level that could be caused by bit flips were good evidence until you said they were reproducible by Dupdome. You specifically say in your video that it was likely a hardware fault with Dupdome's N64. This is unlike Dota_teabag's case where neither he nor the person who bought and took apart his N64 could replicate the warp on it. Also, you repeatedly said how it was so unlikely that a cosmic ray could have caused the bit flip, therefore that cant be the cause. But that discounts the sheer number of hours that people have played this game live on stream. Especially once you consider that it didnt have to this game, it could have been any game that is speedran and streamed where a bitflip could cause a stunning glitch that cannot be reproduced. Its similar to how you should be suprised and amazed when you win the lottery, but you shouldn't be surprised when SOMEONE wins the lottery.
@@chrisgaming9567 That's because if he showed the calculations, it would show that computer systems are bombarded to shit with cosmic rays that can flip bits and fuck around and the nintendo 64 is old enough that a lot of the modern (relatively speaking) redundencies just do not exist within any of the components.
@chrisgaming9567 I get that components that would mirror the N64 would be impossible to get today but there are definitely better methods than using an N64 and a blender
My only gripe with this video is how adamant it is, that "it is simply not true", while we still have no real idea what actually happened. It was a leading theory, it is not anymore. All that means (to me) is that we have found more probable explanations. But the original theory wasn't exactly DISPROVEN, we just realized it was not probable (but not technically impossible). Just another example of the media, and "the people", not really following expert discourse on the matter, and instead tightly holding onto, and parroting/reposting what gets the traffic in. Other than that, cool vidya. I get the youtube need for a "click-catchy title" too, so i'll excuse it, lolol.
damn i thought this video would be an actual discovery or something but it's just saying "actually it was probably something else and not a cosmic ray". how wildly informative
'the biggest myth' -> a photo of the up-warp myth, untruth, false, fake news, it was NOT a cosmic ray, it was a… maybe these other things but we don't know i personally expected to see the runner accused of cheating or the holy grail- replication i've never heard the myth of 'we've proven it was a cosmic bitflip' all i heard until now was 'one of our remaining theories is a cosmic bitflip' tbt if i had to bet my life on it, i'd think it was local EM noise or power grid noise, but this is such clickbait
What's funnier is the most likely outcome WAS a bit flip. He claims that because there's no 'proof' it was a bit flip, therefore it can't be a bit flip which isn't an actual argument at all. RAM back then did not have protections against random bit flips due to radiation or even just overheating. The TAS clip clearly shows that flipping the bit from C5 to C4 gives the same height as Teabag got, so it is 99% likely the bit was flipped. HOW it was flipped is obviously unknowable. But a bit WAS flipped.
@@m0rty161if I make a video saying the moon landing is fake, but then don't explain any of the footage or evidence, would you make that same comment lmfao.
@@alextheonewarrior the only "evidence" there is of the cosmic particle is "but it COULD be how it happened" which is the same mentality as "i COULD win the lottery today"
yeah honestly him being so aggressively against even the slightest possibility that a journalist might be right once that he denies the possibility altogether without even disproving it makes me just want to believe it more.
@@loafodisease614I know the feeling, but I think it’s very unlikely to be a cosmic Ray. LunaticJ really doesn’t do a good job of arguing his point, but the odds of one cosmic Ray striking an N64 and causing this particular glitch is already a very very low chance, and how that we’ve seen multiple TTC bit-flips, it seems incredibly unlikely to blame them all on cosmic rays. And if we find there’s an alternative cause for some of the bit-flips, it’s much more likely that it’s a similar cause for the original TTC upwarp. Technically it is possible that some of the bit-flips were caused by cosmic Rays still, but it becomes a lot more unlikely. I think the real takeaway from this is not to trust badly research journalist articles that contradict their sources, and not to latch onto one theory that doesn’t really have any evidence. But it’s also just as bad to ignore unlikely theories in favour of explanations that can be disproven, or don’t satisfactorily answer the question, like a bad cartridge read.
@@justanotheryoutubechannelwell i was making the implication that the result of the video is harmful, since his wild assertions are the exact same misinformation he claims to be challenging in the beginning of the video. The point is his certainty in said assertions is fueled by spite for gaming journalist sensationalism, not any actual observable proof that there is no possibility.
the only thing this video proves is that there are many POTENTIAL causes for the up-warp, but because nobody knows exactly what happened it could still be a cosmic ray event. just because it's unlikely doesn't mean it is impossible. so saying that it's a totally impossible myth isn't very accurate. It is unlikely but that does not mean impossible.
I work in biopharma with highly sensitive equipment. We get cosmic rays happening regularly causing outliers in our data. We have to account for it with our software. On a normal PC size machine this stuff happens once or twice a day sometimes while we are running tests. Tests only capture data for a minute or so. So really we are only capturing data for about 2 hours at most in a workday. And in that time we often see them. Sometimes it'll be a few days without observing one. If just one bit needs to be hit to cause this, then it's just as likely as any other bit being hit. I'd say it's plausible.
From my understanding it could have been a cosmic ray, but there are other similar events that point to other possible explanations. So it's possible if unlikely to have been a cosmic ray.
@@xOrynos you came to the conclusion that it was "literally anything other than a cosmic ray"....when the video very clearly made it known that the cosmic ray theory is implausible when put under scrutiny and compared with more tangible scenarios like a faulty console.
ok so why is the ionizing particle theory a myth though I was expecting some level of evidence to be weighed against it, but instead it's just touted as ridiculous, seemingly on principle?
I think the myth is that it was ever proven to have happened. It's still a possible explanation, but there's no hard evidence that (1) the glitch happened because of a bit flip or (2) that said bit flip was caused by an ionising particle. However, I personally think that the bit flip, since it is the only reproducible method to quite accurately achieve the same result, is the best existing theory. However, I know a decent amount about computer science, but next to nothing about particle physics, so I have no idea how likely it is that a bit flip would be caused by a cosmic ray.
@slynt_ The funny thing is that no one really knows how likely it is. Single Event Upsets are likely underreported, and even when they happen, you're never gonna be able to get evidence that it was via cosmic rays. Additionally, nowadays, there is typically redundancy to ensure Single Event Upsets do not affect anything, so we might not know when they do happen
@@byeguyssry It would be interesting to see scientific research into the probability of Single Event Upsets and to what extent different variables affect that probability. Maybe there is research on that, but probably not since it's a pretty niche thing to be interested in and like you say, it isn't really a practical problem because of redundancies we've developed.
@@slynt_ If you do know a decent amount about computer science, you should know that modern electronics are specifically designed to error correct for cosmic rays flipping bits in things like memory. AFAIK the nintendo 64 does not have any of this error correction, if a bit is flipped than the console just rolls with it.
Don't get me wrong, you've absolutely convinced me that hardware issues are the far more likely culprit here, but I feel like you MASSIVELY underestimate just how common cosmic rays are. They're not some out-there weird physics concept, you've had several pass by you in just the time you've read this alone. Single flip events are a good chunk of the reason error correction codes are as robust as they are.
@@Pupperpatsmany bits to flip, redundancies and fabrication techniques in modern technology to prevent single event upsets from affecting normal use, eec memory
Yep, if your PC has enough RAM there are actually scripts you can run to detect them based on bit flips. According to IBM, for each gigabyte of RAM your PC has, you will experience one cosmic ray induced bit flip per week. Though, this was in the 1990s and technology has changed since then, so the numbers might be a bit different now.
@@thelegend8570 Data corruption on hard drives is super common as well. Modern file systems do lots of integrity checks but even so people who are in the habit of doing extra validity checks on their data often find that errors have crept in!
Small correction for 13:50 No, we do not in fact "know for a fact that it was not a bit flip" because several bits could have been flipped, I believe there is no reason why several adjacent bits can't get flipped together instead of just one of them. In fact, three adjacent bits of 512 + 256 + 128 add up to 896 which is almost exactly 900, so if people do choose to believe it was a bit flip, it would likely be those 3 bits.
@@davibergamin5943 Depends on the particle and the physical structure of the cache line/RAM the the N64 uses. If a neutron comes in a the right angle I see no reason why it couldnt create a reaction that flips three bits. This is very unlikely, but that doesnt make it impossible.
@@davibergamin5943 while I do think it's because of faulty hardware, bit flips should not simply be disregarded. It's still a valid theory, even if unlikely. In fact, unlikely things happen all the time, simply because of how big the world is.
@@davibergamin59432 things 1. Couldn't a bit flip add to it, or is that not how bit flips work? 2. Since they are adjacent, could it be possible that the same event caused all the bit flips?
@@davibergamin5943 When thinking of impossibility consider first the odds of the Earth existing with us on it. Consider that either we are the impossibility or the visually empty universe is the impossibility. Anything above a flat 0% is possible and under a flat 100% can fail to happen.
Veritasium's video describes the event with much uncertainty, because it was never proven or disproven, which is good scientific communication. In his video he said: "This is *possibly* the rarest thing...", "...it *seems like* a newly discovered glitch...", and "The *best explanation* anyone can come up with is that a cosmic ray caused the glitch." That he cited The Gamer is cause for concern, but Veritasium never, as you suggested, used this incident "...as an example of a bit flip *definitely* happening..."
@@delcogoblin1. he wasnt angry in the video. 2. not proven either way? really? thats not an argument when youre defending a very unlikely argument instead of listening to the most likely scenario. as another commenter said "did i forget and put a dirty bowl in the cupboard, or did some complete stranger break into my house for no reason and plant a dirty bowl in my cupboard? well, no proof either way!". see how stupid it sounds to say "no proof either way" here? 3. veritasium did this exactly. "the best explanation anyone can come up with." really? science wannabe youtube commenters making an insane theory over the sensible and simpler explanation is "the best explanation anyone can come up with?"
I really, really wish you hadn't gone out and tried to say that SEU's caused by cosmic rays are an "out there" theory. It's in fact a very well studied, understood, and engineered against phenomenon. Whether or not THIS specific event was one of these is another matter, as they are much more rare down at the earth's surface, but you didn't need to go out and say something that you either didn't understand, or that you intended to mislead your audience with.
For real though, so many people in the comments think cosmic rays are something magical. do they ever wonder where the classic static pattern from, say, misaligned antennas on old TV's come from? It doesn't just exist or is programmed into the TV's, it's the antenna picking up random noise from the environment and a not insignificant amount of that noise comes from... cosmic rays.
yeah except that... bit flips caused by cosmic rays at ground level are extremely rare. The less atmosphere there is, the more likely that a photon will interact with a bit. it has happened for planes and for space hardware, but it's still a pretty "out there theory" that this is specifically the reason it happened and not just a hardware malfunction. yes the phenomena is well understood, and that's exactly why it's still valid to question the likelihood of this being the cause of the glitch.
@@hypstersaurus6828 My issue is not with the argument that it caused this specific glitch. Even if it's caused by a SEU, it's more likely to be things like LOCAL radiation sources or hardware defects etc. My issue is that the entire tone/idea behind this video seems to be trying to debunk the idea that they happen AT ALL which is a ludicrously ignorant position to take. And judging by the comments here, exactly what made me angry has happened: the careless, lazy, ignorant, and/or stupid way this video was written has misinformed many people about the subject, and they have walked away thinking that SEU's caused by cosmic rays are akin to big foot. The video did that because it's easier to make the argument that way, which is lazy and annoying.
To be clear, the glitch has still not been precisely replicated. And while you are correct that we do not and cannot know for certain that it was a bit flip that caused it, you also cannot dismiss the possibility that it was. There are definitely several other more likely explanations, but a bit flip remains possible, even if unlikely. The only myth is that anyone could have ever been certain about the cause of the glitch.
Not frame and pixel perfect replication, but anyone saying the bit flip recreation wasn't proof enough that somehow that's what happened, cosmic ray or not, is ridiculous. It's pretty obvious by the two videos side by side, unless there's another way to recreate very, very similar upwarps in that spot, it's a bit flip. Don't tell me that video side by side with the original isn't pretty conclusively showing damn near the same exact thing. Just because they weren't in the exact same spot for the warp doesn't make it any less incredibly, wildly close. Like so close it's shocking me that anyone thinks it's not a bit flip. The question isn't "was it a bit flip?" The question is "how did it happen? cosmic ray or not"
@@chrisgaming9567 What makes it unlikely is how improbable it is. It's not impossible, but there are other more likely things that could cause a bit flip than a cosmic ray. That all said, it has to be a bit flip. There have been no other remotely probable theories as to how he could have warped exactly that distance. There was never going to be a way to imitate the gameplay 100% perfectly side by side, even in a TAS, but it was more than close enough to show that the non cosmic ray bit flip theory almost perfectly matches what we saw in reality. Something I wonder about with the weird 900 unit down warp was if they tried to recreate it in TAS as well? If they did, maybe warping Mario down 1024 units makes him trigger some sort of in game anti clipping mechanic (the game checks for walls in marios movement path sort of) so he only appears to move 900ish units. It might be unlikely, but there is also potential that 2 bit flips combined made that happen. +1024 and -128 would be around 900 units. I don't think any bit flip we've seen in Mario 64 speedrunning has been caused by a cosmic ray, but I think bit flips do occur.
the fact that LunaticJ really made up the "proofs" like "cosmic rays are rare" and "look how it does match with a TAS ? i have decided that it actually doesn't mach lmfao" this video is a joke
HOW TO GOOD DOCUMENTARY IN 2 EASY STEPS!!!1! 1: Try literally everything in your power to make every point sound as biased as possible. 2: Fail to understand the definition of the word 'theory' Hope this helped
My dude, you're doing the inverse. Going on and on about how it's DEFINITELY NOT, CAN'T BE, COULDN'T BE a bit flip. Is a hardware issue plausible and also more probable than an SEU? Sure. Was a hardware issue proven to cause the upwarp? No. Is a bit flip still possible, even if it is highly unlikely? Yes. The only correct answer to what caused that upwarp is "we don't know."
Exactly! The whole time he was going off about cartridge tilts, I couldn’t help but think that the bit flip theory was still possible, if unbelievably unlikely. I only started to doubt that belief when that other runner had consistent bit flips from a hardware issue, I would have lead with that.
@@tinthatisfullofbeans Even then he already states that they found no noticeable issues with DOTA_Teabag's N64 nor were they capable of recreating an upwarp. Basically he says: The closest recreation of the original upwarp was done by someone flipping a bit, despite DOTA_Teabag's mention of sometimes tilting his cartridge to play there's no known instance of cartridge tilting flipping bits/causing upwarps, DOTA_Teabag's N64 also had no noticeable hardware issues / was unable to recreate the upwarp again, despite all this I'll just bring up an instance where someone else's N64 did likely have hardware issues and was believed to be glitching out levels through flipped bits. Therefore, despite having zero similarities to the DOTA_Teabag upwarp, this situation is proof that a cosmic ray absolutely did not flip a bit. Throughout the whole video he shows information that CAN suggest flipped-bits, if nothing else, but claims that they're not definitive so he looks at other situations/ideas where he acknowledges there's essentially full confirmation that these ideas can't/haven't caused upwarps or flipped bits yet goes on to say that they're magically a more plausible answer than flipped-bits (and especially cosmic rays causing it). like???
@@CobRaM4tRIX "there's no known instance of cartridge tilting flipping bits/causing upwarps," it doesnt seem too unlikely based on just logically like. if a pin on the cart was disconnected from the console it could defintely mean the console would then encounter undefined behaviour when trying to read the instruction, and if the instruction was something to update marios position. then maybe something like that could happen.
@@LiEnby I'm just quoting what he was saying in the video there. I don't know enough about that particular thing therefore I can't claim with certainty that it's too unlikely. Just from what he claims in the video he's explained that there is no instance of that kind of glitching caused by cartridge tilting (Things such as sounds, visual assets, etc, being corrupted or improperly executed due to a consistent stream of poor communication between the console and the cartridge makes sense. The idea that this constant stream of potential corruption/improper execution somehow doesn't affect the game at all in any other way yet somehow at one point randomly flipping a bit in the memory and never having problems again is not only extremely unlikely but also never has happened before.) From his claims/wordings, and from what I interpret it as with my limited understanding, is that cartridge tilting is theoretically just as unlikely to flip bits / cause upwarps and has never had any tangible evidence of it being able to do so. But he goes on to treat the general idea of flipped-bits (especially from cosmic rays) as such a laughable concept despite it having just as little "evidence" (though technically it does have more) as the failing hardware concept... Especially mixing that with the fact that the original console and cartridge that this glitch happened on never recreated a bug even remotely close to the upwarp since (and had no noteworthy hardware issues). Again I'm not an expert and have a limited understanding on this. But with that limited understanding, and through his own words alone, a tilted cartridge theoretically has no tangible proof of flipping bits especially in a manner like that. Almost implied in the video like it's theoretically impossible due to how the cartridge and console interact anyway. But it seems like he'd rather ignore that (and the other things) and suggest that it's a far more plausible option because...? I guess it's more satisfying to think it was an unexplainable general hardware error instead of cosmic rays? (which btw are not that uncommon nor anywhere near as "rare" as this video paints them to be). - EDIT: Just to reiterate. It seems that the tilting of a cartridge messes with multiple aspects of the game even if they're not immediately apparent. The major point being that there's often a very blatant issue with the game right away alongside any other small glitches caused by a tilted cartridge (broken skybox, broken music/sounds, broken animations, or even outright crashing, and so on). That would mean that somehow, despite the tilted cartridge, the game didn't show a single shred of bugginess for nearly an hour until it randomly decides to change Mario's height in a single instance then immediately operate as normal right after and never bugging out again. Even in that split instance not having a second of corrupted audio/visuals or anything. (which isn't required but it's just cherry on top for this extreme set of conveniences). And it's not like it accidentally sped Mario up immensely to throw him above the higher platform or anything like that (which wouldn't even make sense considering he was moving downwards in the first place). But rather it completely overwrote whatever his Y position was in that instance while preserving that downward velocity. Where he's suggesting that a tilted cartridge is more likely to have caused an overwriting of memory in that way (in spite of known behaviors and lack of any instance of this happening before) rather than the idea that a bit was flipped (with which its only behavior is to do exactly that kind of thing and only that kind of thing) even after showing instances of bit flipping affecting the game in that exact type of way due to hardware issues. He acknowledges that DOTA's game didn't have hardware issues worth noting but he decides the far more plausible scenario is that a tilted cartridge somehow perfectly affected the game in that exact type of way in that specific second despite establishing the idea that it's not happened before and conceptually shouldn't happen. Rather than believing that something outside of the hardware then affected the bit to flip it (a not uncommon and well-understood concept in computers). I'm not 100% team cosmic rays but like... it's not THAT implausible nor does it make sense to suggest that what he said is, not only more plausible, but by such a significant margin that the idea of cosmic rays is laughable.
Ok, SRL speedboomer here, the cosmic ray theory was around way before the 2020 clickbait article, i remember hearing about it when it happened in 2013. Furthermore, it was a common rumor back in the 80s and 90s back when computer hardware wasn't nearly as robust as it is today. The rumor goes (yes a rumor) that people would purchase RAM upgrades for their PCs, and when this RAM turned out to be faulty, people would angrily call the manufacturing companies of said RAM looking for answers. Instead of admitting that they sold faulty RAM to the customer, they would blame crazy things like cosmic rays for flipping bits. So yes, this Cosmic Ray Meme has been in computing for at least 40 years lol.
bc of solar flares during the time period which absolutely fried a bunch of computers and lead to an article about something people couldn't verify on their own
Physicist here. Without going into exactly what my work is, cosmic ray incidents are extremely common. It’s not a constant thing, but we expect a certain number of hits on our devices over a certain period of time and we have to account for it 😅
Reasons why the cosmic ray flip is a myth: - there's a different upwarp glitch - someone had many bit flips happening - some other person had a slightly similar thing happen that wasn't a bit flip - uhhh some people had faulty consoles that loaded textures wrong - uhhhhh someone slapped their console and the wrong level loaded also I ran my mixer next to my console
@@electroflame6188 You mean > person's game wouldn't start sometimes > someone tested the game and there were no bugs > but some other people had bugs > clearly, it proves that was a bug
I was the original author for the article on TheGamer. I can proudly say I no longer write for that website, nor do I do freelance gaming journalism anymore (and I wouldn't recommend you do it either). I want to clear up a few things about the ways that sites like TheGamer function, because I've grown a little tired about the amount of shit I've been given for this article in particular and others like it. I am very aware that the cosmic ray thing is just a theory, and I was very aware of it when I wrote it too. When I wrote for them (I can't confirm it still works like this), we would submit pitches in Slack, and those that were approved, we could write basically however we wanted to. With features/originals like this one, they weren't likely to pull in lots of page views without being turned into clickbait in some capacity (since it isn't "news"). What I want to make very clear is that turning articles into clickbaity feed-fillers is the job of the editors, not the authors. After I submit it, I no longer have any control over it. Editors re-work the titles and content and push them out. I could point to many articles I wrote that were published as much weaker articles than what I submitted, because the editors prioritize things being concise, rather than accurate, and things being SEO-friendly, rather than readable. When I submitted the article, it fell flat and barely got any page views for the first few months, which is rather annoying considering that authors only got paid for the page views from the first month, so I made basically nothing from the article, minus the flat rate for it ($10 USD I think it was for that word count?). Then of course, it blows up months later and becomes one of the best performing articles on site for a bit, and I make nothing from it, other than needing to hear about it from time to time in videos like this. TheGamer is actually the ONLY site I ever wrote for that paid extra for page views at all, so it always pisses me off when people give shit to freelance writers for just trying to get clicks. Overwhelmingly, freelance writers are just getting a flat rate so page views mean nothing to them. It wouldn't make sense for them to deliberately write clickbait. It's precisely the reason I had my name switched to a pseudonym after I stopped writing for them, and my go-to example of why no one should do freelance gaming journalism. It is a thankless job where editors twist the articles you put passion into, only for you to get shit for it when it's published under your name.
That's actually horrible of those editors. The fact that you never got compensation after the article blew up, regardless of quality, is really unjust. I want to make it clear I have nothing against you and I hope you've been able to move on from that thankless job. I deeply apologize if my tone was unfairly harsh, and I should have been more empathetic to your struggle of being a freelance journalist. the last thing i want to for more people to harass you over a trivial article. I made this video primarily because I saw people in the SM64 community get annoyed by people treating the TTC Upwarp as if it "definitely" happened, and wanted to make a comprehensive video on why they didn't believe it, and how the cosmic ray meme got out of hand. If you think I crossed any lines that may have upset you or caused you any trouble, I deeply apologize.
@@LunaticJ No hard feelings. My comment was more a response to other comments complaining about clickbait articles than it was to the video, and it also seemed like a good time and place to finally vent about my role in the cosmic ray saga. I just wish I never wrote that article to begin with.
@@throwawayblehblehblehYou do understand why criticizing the article is fair and reasonable, right? I'm perfectly happy to blame the editors instead of the authors.
Yeah, the cosmic ray theory being paraded as definitely true by media is egregious, but I don't understand the feeling of animosity towards the theory. Sure, it would be better if it wasn't a freak instance, as that may mean it could actually be replicated, but absent an actual technical explanation, it's just as valid a theory as any other considering what information we have.
@@FirstRecords204 Agreed. That being said I don't think you can accuse Veritasium of reporting it as fact. He merely pointed out that it was a possible example of a notable instance of a bit flip.
Thank you. This video feels like needless semantics. "Something happened that cannot be replicated through an emulator or normal console consistently, which caused a single bit of information to flip, teleporting Mario's positioning upwards." That's the factual story, and it's already super interesting. People went with the cosmic ray theory because it's the more interesting culprit between the faulty cartridge or electromagnetic waves. None of them can be disproven or proven, at all. Especially considering that it was the only bitflip to happen that stream. Yes, it's the least likely of them, and it got reported as factually being it instead of possibly being the culprit, but it's an interesting tale nonetheless. The fact that the footage doesn't match with the original 1:1 by upwarping through hacks can also be easily explained by consoles not always playing at 100% speed. It just comes off as the video not feeling like "it wasn't for sure a cosmic ray, but rather, a cosmic ray is still possible, but rather unlikely!" Wasn't substantive enough material for a video, and it needed a "it's FACTUALLY NOT TRUE that it was a cosmic ray!" Moment. And that's just the same bad journalism that the video criticized being repeated again. Sometimes stuff in life will forever be uncertain, and that's okay.
Yeah, this dude is completely dismissing the well-documented science on SEE. The answer should be: "no we don't know what happened and one of the theories is bit flip caused by cosmic rays" and not "cosmic rays totally didn't happen guys because that would be EXTRA silly."
@@ultimateman55 Using a shoddy article as a source is just bad journalism. It doesn't matter if he didn't explicitly sag it was absolute fact, he's presenting it as an example among a bunch of actual factual evidence, so it's going to be seen as such because he's a science communicator and is expected to be reporting things with a certain amount of rigor. If a video that is overall very scientific suddenly has something in it which doesn't meet the same standards, then that's a mistake.
As pointed out by the other comments, bit flips caused by cosmic rays or interference happen all the time in computers, which is why we have ECC memory and other redundancy schemes. So the idea pushed by this video of this being a super unlikely hypothesis that nobody took seriously is false. Plus, cartridge tilting can't cause single bit flips in the console's RAM or cache--he even shows someone saying this in the video--so that hypothesis is out. Finally, a console which suffers multiple bit flips in a short time probably has faulty memory or cache, but this video even explains that thorough testing was done on the actual N64 which had the upwarp, and no further issues were found. So the faulty hardware hardware hypothesis is also out. Yes, it would be accurate to revise the claim from just cosmic rays to "cosmic rays or other outside interference", but that's the best hypothesis we have so far.
ECC also helps with errors from faults or nearby interference ... similar to the explanations in the the video, there are many potential causes for single-bit errors
Exactly what I was thinking while watching this. Someone should really look into the redundancy schemes of the N64's cache to find out if a single bit flip could have made it through without other noticeable effects. Since the game's code is so deeply understood these days, we know it's not an anomaly or math error in the programming, and we know cartridge tilting cannot have this exact effect. A formal error testing of the offending N64's memory after heat soaking it should be conducted, instead of just trying to replicate the glitch on the console, as I'm sure there are other bits that could be flipped without noticeable effects like this. I think a power surge is a more likely explanation, or other types of ray interference, which the N64 is legally obligated to accept (and deal with) due to FCC laws.
@@TheKingOfApples100 well the OP's comment was only directed at the original case. The other n64 is confirmed to have memory issues, which couldn't be replicated on the upwarp console, so I don't think he was implying that one was being bombarded with cosmic rays enough to make the bit constantly flip as it does in the video
So a faulty cartridge, which you pointed out *isn't known to have this effect,* is more likely to cause this effect than a cosmic ray, which only has this effect?
My gripe exactly. I wanted to post a carefree comment about the cosmic ray hypothesis being more of a dated meme than a hard answer to the problem-at least to those of us who were following the story at the time-but this guy is taking it seriously and reaching so far in the opposite direction that it's not even funny. I'm walking away with *more* faith in the cosmic ray than when I was 10 years younger.
@@semisixx4967the cosmic ray is still a meme though. There are plenty of things that can cause bit flips. This is why they sell error correcting RAM after all. Cosmic rays are the least likely of those things, they just sounds the coolest.
Pointing at other examples where bit flips occured is like saying: "Because 10 / 10 murders we sampled happened with guns, all murders happened with guns" Just because this theory doesn't apply in other spaces, doesn't mean it doesn't apply to where it was originally suggested.
I think we should continue believing it was a cosmic ray just to spite this guy who seems to think he’s smarter than everyone else when he gets basic info about this wrong
"We don't know" not only means "it's not proven it DID happen", it also means "it's not proven it DID NOT happen". It's simply not yet proven on either side.
Telemetry data from Microsoft has shown that a noticeable portion of app crashes and blue screens are due to bit flips too. But even those can be explained through interference between memory cells and "in flight" between the DRAM modules and the memory controller, especially with any form of overclocking. Also with the gigabytes of memory that a PC has, any bit flip is much more likely to hit nonessential data.
The N64 will be using higher voltages and larger transistors than any windows machine, which makes bit flips much much harder. You need much more energy to flip a 5V bit than a 3.3V or 1.8V because the particle has to create that voltage on the memory cell
@@lomiificationCosmic rays are quite high energy objects. I don't have numbers off the top of my head but if a particle ionised by one has enough energy to flip a 3.3V it seems likely for them to have the energy to flip a 5V bit.
@@lomiification It's the *charge,* not the voltage. Okay, let's dig a bit into that. A 1GBit Samsung DRAM from 1995 with a supply voltage of 3.3V would internally store 1.8V in a DRAM cell. Each cell has a capacitance of around 25-30 fF (femto-Farad). That's a difference of maybe 300,000 electrons between the two sides of the capacitor (assuming I got my math right). The N64 had a 32MBit Rambus DRAM (36MBit actually, I'll explain further down). I don't have the technical details there, but a typical call voltage would be around 2.5V (supply voltage was 3.3V). Note that the Rambus uses differential 0.2V signals. The structures of such a 32MBit DRAM will be much larger than those of a 1GBit DRAM, so they can accommodate not only a larger voltage, but will also have a higher capacitance per cell - you have to convince much, much more electrons "to change sides" in a smaller DRAM. Nowadays, almost 30 years later, structures have shrunk even more and our modern DRAM is much more susceptible to cosmic radiation due to the smaller structures (that's one reason why quite a few space probes use outdated technology). This effect is however mitigated by better manufacturing processes and geometry. Technically, the RAM of the N64 would have been ECC RAM, but it wasn't used like that. The CPU could access only 8 of the 9 bits per byte (for obvious reasons), and the 9th bit was used by the GPU only. No ECC or error-checking there. Hence 36MBit inside the RAM, which looked like 32MBit to the CPU. A few decades ago, Microsoft claimed that about 30% of all Windows crashes were caused by bad RAM. Considering the overall stability of the Windows versions back then, this seems to be a very bold claim. Well, maybe Microsoft had a farm with thousands of PCs with bad RAM and had them crash continuously, to get to the 30%? Whatever. A Google study from 2009 (IIRC) showed that 8% of their RAM modules enjoyed one or more bit errors per year. I'm not sure what kind of memory modules Google used between 2007 and 2009, but I know that, in the past, Google had used very special memory modules which offered more memory per slot - and violated the specs. TL;DR version: Yes, i suspect that you're right when you wrote that it's probably harder for cosmic radiation to flip bits in an N64 RDRAM than in the more capable PC DRAM modules.
Bit flips are a well documented phenomenon. Sure, they *usually* happen to server farms with terabytes of ram, just because there's more stuff to go wrong, but it's bound to happen to an N64 eventually. Whether it was specifically caused by a cosmic ray is unknowable and frankly irrelevant... a single bit flip still seems the most probable cause.
Counterpoint: If the actual cause of the upwarp hasn’t been definitively proven, then you can’t say with certainty that the cosmic ray hypothesis is untrue. I choose to believe…
It’s maybe a counter argument but as you said, rooted in nothing but belief. There’s nothing scientific about saying “it could be this, so you should entertain it”. You go with the explanation that’s most likely, and remain open minded enough to change your mind if you see that something else is more likely.
@@Cosmalano Nope, wrong! Every other hypothesis has resisted demonstration despite having a much lower test-ability treshold, it's literally anti-scientific to refuse entertaining a perfectly plausible explanation on grounds of odds when none of the other more mundane ones can be proven despite having the exact offending hardware that experienced the glitch in the first place.
@@TheCamps10 that is not “anti-scientific,” let alone literally so. Science doesn’t depend on “proving” an idea right, it depends on proving them wrong. Cargo cult science seeks to prove things right. In science you always seek to disprove a hypothesis. The bit flip hypothesis did not produce the same effect when tested, which ruled it out. If you want to entirely rule out the faulty hardware explanation, then you need to demonstrate that it doesn’t work. If you’re not after ruling things out, then we’re back to my original point, whatever is most likely is the one you should go with, until it’s proven wrong. That’s the only way science works. Nothing in science has ever nor can ever be “proven right”. Even if the bit flip produced the same effect, this wouldn’t mean that a cosmic ray caused it, nor that this is indeed what happened. Just like how saying the Earth is flat works as a hypothesis when you see a couple of effects, but doesn’t hold up when you consider even more, where as the globe Earth model captures everything you could throw at it.
Well the best counterargument could be that the radiation/ waves/ rays came from earth. Storms, Uranium and other waves can still cause problem with tech. Or, ya know the giant light ball we call sun has regulary big outburst of radiation and solar flare, of wich 1 was in 2014!
Regardless of how you believe it happened, you really don't have to treat everyone who believed a widely spread story as if they're stupid for believing it when it was the only explanation given to them for years. Especially not while the actual reason why it happened is still unknown and the widely spread story you seem to dislike so much is still a possibility, however remote.
Exactly! How dare this UA-camr treat people who actively believe that particles from space flip a bit (which was never even proven to have caused this glitch in the first place) and actively defend this opinion when no proof and contradictions to that statement are shown online!
@@kitterbug that's some good sarcasm there. thing is Lunatic never offered an alternative possibility, he just said duh this is unlikely, so it therefore must be impossible! yeah, I am not sure I believe the bitflip idea either, but just because you can't see it happening doesn't make it not true, thats an argument from incredulity
@@kitterbug The thing is, the video isn't even saying that the space ray CAN'T cause a bit flip - heck our modern civilization is one strong space fart away from crumbling as all of the electronic devices could potentially be bricked by such an event - the video is simply saying that it's unlikely that a cosmic ray would cause THIS particular bit to flip at this specific time. Which is... a bit like arguing that it's very unlikely that a person got hit by lightning because why would the lightning strike in the exact position and time that that person happened to be in
It could be a bit flip, just not from a cosmic ray, bit flips do happen more often then people think, just in small areas that make 0 difference usually. Bit flips are more common from old storage/memory, after storage (especially ram) gets old it has trouble retaining bits of information often causing bit flips. For storage in hard drives and SSDs bit flips take longer but will occur if something doesn't get used because it's just sitting there and the charged particles that create the bits can move (either to near by particles swapping bits or to an outside particle loosing the bit all together, the editor roughly mentioned how outside interference can also cause bit flips with outside particles charging the inside storage). This will often make old images have worse quality as colour(color if you're American) bits will get changed making the image more noisy, however, if you move the image around a lot the image will usually get compressed and in some cases will have generation loss as each copy loses information. These bit flips I'm talking about where the bits more corrode then flip (but can still happen in real time) are usually just called corruption (like when you get a corrupt file)
I agree with a lot of the other people in the comments saying that this video is a nothingburger and just as poor of video game journalism, because it claims that it definitively _wasn't_ cosmic rays without providing any evidence. Is it extremely unlikely? Yeah. But unlikely doesn't mean impossible. Saying that it definitely wasn't cosmic rays without providing evidence or clearly stating "it's possible but other things are much more likely" makes this the same kind of sensationalized garbage journalism as any article claiming it absolutely _was_ a cosmic ray bit flip.
He said "it's a leap in logic to assume that over an answer found in-game or a hardware malfunction", that is, to assume something unlikely happened when there are more likely explanations, at around 6:15 minute mark. Did I miss the part where he says something like "it definetly didn't happen" or "it's impossible"?
@@azzor4134The thumbnail and the title both strongly suggest that it definitively wasn't a cosmic ray bit flip (given the connotation usually attributed to the word "myth"), plus some extremely confident language at 18:20, saying "it just isn't true." And at about 18:45, he calls it a myth yet again. Additionally, his tone and wording throughout the video is straight-up hostile towards the idea that the bit flip theory in general is even slightly plausible, despite the fact that it still is as plausible as it ever has been. EDIT: Something I thought I remembered, and found verification elsewhere in the comments - the thumbnail has been changed. It originally straight-up said "IT WASN'T COSMIC RAYS." No nuance, no room for misinterpretation. And now that people who actually know what they're talking about are saying it's unlikely but certainly not nearly as impossible as he claims, he's backpedaling HARD.
So he said it. Though, as far as I remember, he said it was unlikely throughout the video until that point rather than 'false', and his pinned comment also states "I'm just saying it *probably* didn't happen", so I'm willing to let it pass as a small mistake. I understood 'myth' to mean not necessarily a false claim that is commonly held to be true, but an unproven claim that is commonly held to be true. *That may not be the exact dictionary definition of the word, but that's apparently what he meant by it*. So if the cosmic ray hypothesis is held to be true by many people, despite being still an unproven hypothesis, then it is a myth. He attempts to dispel it by showing alternative hypotheses which seem more likely (not a definitive answer, as he concludes by saying that it's still a mystery) and how well he did that is debatable, but he isn't wrong to call it a myth, by that definition. In my opinion he did a good job at showing that we don't known the answer, and how one answer came to be commonly accepted as the correct one, but a bad job at showing how alternative answers to the cosmic ray bit flip hypothesis are more likely. Rather, it still seems that any explanation is as good as any other.
@@azzor4134 To be fair, he edited the pinned comment _after_ me and a bunch of other people made comments calling him out. EDIT: He also edited the thumbnail to what it is from "IT WASN'T COSMIC RAYS." I think that says what you need to know. I admit that this may be a misunderstanding, but at the same time it rubs me the wrong way that he failed to do his due diligence while trying to come across as someone with journalistic integrity. He clearly came into this with a strong opinion against the bit flip theory based on his word choice and his tone, and yet provides a bunch of evidence that, if anything, _supports_ the bit flip theory - examples of hardware malfunctions that are often far more drastic, the N64 and cart used have been tested by people who know the system inside and out and nothing was found, _flipping one specific bit is identical to what happened,_ etc. I would have no problem with this video if it were titled better and used more agnostic tone/word choice throughout. But I think it's pretty clear that the implication is "the bit flip theory is incredibly dumb," when it's just as if not more plausible than it ever has been.
I work in the same office at NASA JPL as the radiation effects expert in Veritasium's video. If you have a GB of RAM, then you might be getting a bit flip once per day on earth on average caused by cosmic rays. Google estimated the DIMMs in their study were getting 10 per day. So there are billions of bit flips every day on earth induced by cosmic rays. A single event can cause one bit to flip or many. Radiation induced upset in memories is very well understood; there are particle accelerators all around the world used for testing electronics that run 24/7/365. If N64 had 4 MB of RAM, then a bit flip every 200-300 days would be expected (1/250th compared to 1 GB). So games who only play a couple hours per week may never see bit flips in their lives. But gamers playing 8 hours per day... they may not all see them on their 4 MB systems, but many (10-30%?) definitely saw them on N64 if they played a few hours per week for a few years. As someone whose job is to understand electronic interconnect reliability, solder joints, intrinsic memory cell reliability, and radiation effects; by far the most likely explanation is radiation induced upset rather than any of the other possibilities mentioned. Anything to do with interconnects or solder joints or anything at a package level (cart slap) would cause catastrophic malfunction; not a single bit flip. Intermittency on one or more signals between the cartridge and system does not cause a single bit flip. That leads to major memory corruption. The only other credible root cause in my opinion is a weak bit; one that is marginal and beginning to fail. This is common in memories. All electronics have a limited lifetime and will eventually fail; process marginalities in memories create a distribution of bit cell lifetimes with some failing before others. However, in that case, you would likely see further degradation over subsequent months and years of use. Not sure if that was the case for that user. If the glitch was isolated to a single run, it was most likely cosmic ray (not EM interference; that also does not flip specific bits...) This is not conjecture; it is 25 years of daily first-hand experience dealing with electronic component reliability issues including the ones mentioned in the video. (the explanation at 14:00 about powers of 2 is total nonsense; what about a bit flip that changes mario's position 512 to a position where there is no floor and he falls the rest of the way to 900? i'm not saying that's precisely the case, but it's not remotely as simple as described in that section of the video.)
Complete opposite for me back when this was still new. I first found a video talking about the cosmic rays (or the possibility of it being the case) before I heard about the bounty
It sounds like the cosmic ray theory is the most "complete" theory to explain the observed phenomenon. It's recognized that it is unlikely, but I think this has to be weighed against the evidence that we know cosmic rays do occasionally cause observable bit flips in computer systems. The alternative theory here is "something something to do with faulty hardware", but a complete theory isn't presented. I would say neither a single event upset nor a spurious hardware failure can be ruled out at this point. It seems unlikely given the limited evidence we'll ever be able to come to a fully definitive solution since the event happened in the past and we don't have access to the hardware state at the time. In that case the best we can hope for are some good theories, and I think both theories above are decent.
While I think the general consensus in the comments has already been established, I still would like to weigh in. This video is wonderful case study in how Not to go about a counter argument. To start most generally, in your attempt to use Occam's razor, you fell into the watchmaker's fallacy. That being, from within a complex system, and with hindsight, you cannot outright discredit an improbably based solely on the poor odds. Typically used when arguing theology, the idea goes that you find a watch on the beach. While in theory, there's a chance through extremely fortuitous odds that the ocean could have created a working watch, the more believable theory would be that there's a watchmaker who in fact made it. The problem comes from observation bias. If something improbable doesn't happen, you would never question the improbably of it. If something improbable does happen, and you're not there to see it, it results in the same lack of reaction. Rather than an observation of something not happening to disprove a theory, it's using an observation of something happening to disprove said thing from having happened. Secondly, experimental data will never be "one to one" even across the exact same tests done back to back; speaking from experience as a physics major. There's a reason the field of statistics is a thing, as it's no simple process to determine if variation is within impactful limits. To my uneducated eye- likely that of many of us- the bit flip test looked just about frame perfect, but with a different canera angle. Perhaps there's some meta data that says otherwise, but if so, that should've been shown. Additionally, one test of the bit-flip, in one location does not a data pool make; a data puddle more like. With the number of variables in a game like SM64, you need to test the primary variable arrangement many times over, and then begin to adjust other variables incase there are unknown linkages; perhaps your house in under an ozone weak point, or you have more radon in your soil, or maybe TTC has some poor code which causes incorrect bit calculations. These are all things which need testing before a bit-flip can be ruled out. Lastly, on cosmic rays themselves, as you said yourself, there is no concrete answer thus far and this is stuck in theory limbo. A misconception of scientific theory is that they are heeded because the evidence prove them; rather, a theory is heeded because there is yet decisive evidence to Disprove it. The theory of general relativity has many issues, mainly it's incompatibly with black holes. The reason we still use it is because we don't have a better explanation. The same goes for here. SSEs are shockingly common, as are cosmic rays, and while there are other causes for SSEs, it would seem thus far, none of them apply to this situation better than a cosmic ray. All told, this was a poorly written video with a combative tone, and which not only is predicted on fallacies, but falls victim to those it attempts to call out. This is how misinformation is amplified; a perhaps mildly incorrect statement countered by an even more incorrect statement, and the cycle repeats. If it weren't for the fact that this is involving SM64 speed running, I would be genuinely concerened that this was number 40 on trending. If this is something the SM64 community truly want to put to rest, screaming at the world that we're all wrong and uninformed, why don't you inform us. The burden of proof is on you, and imperical data carries infinity more weight than words. Show that this particular SN64 has poor radiation shielding, or that the original owners home has an unusually high count of ionizing radiation, or that the position data doesn't align with more than eyeballed videos, or that system impacts can induce bit flips, or that some code in TTC is broken, or any number of other counter theories with better proof than cosmic rays. Until that testing is done, until there is more counter evidence than "this seems improbable," then this argument is baseless.
"Rather than an observation of something not happening to disprove a theory, it's using an observation of something happening to disprove said thing from having happened." I get English might not be your native language but I am genuinely stumped at this. Watchmaker's fallacy is not used here, in fact it's a term completely made up to disprove the Watchmaker Argument developed by William Paley in the s.XIX. Both the argument and the rebuttal are only meant to use in one specific context, and are not applicable to other scenarios. As a clarification, I don't accept the watchmaker argument because it uses a man-made object to build the premise of a designed object. It's circular reasoning. It has nothing to do with probability or observability. A much better use of the argument is to argue design patterns using a natural object as a premise, like DNA strings. But then that would not be the watchmaker argument, but something else entirely.
@@FrcNeru I don't think english is the issue; his grammar and diction are more or less fine. The issue, I think, is that he is probably just trying to be hyper-specific with his wording, resulting in awkward phrasing. Speaking as an autist, I do it a lot, and that's what this feels like to me. I'm not 100% certain what he's trying to say, but I'm pretty sure part of what he's saying is that we are more likely to notice that improbable than the probable. For example, a meme in XCOM is that the odds of hitting are significantly lower than stated (eg. a displayed 98% chance to hit is actually a 50% chance to hit), however, in XCOM, the opposite is true; the game actually gives you hidden bonuses to make the odds of hitting more likely than shown rather than less likely. The reason it's a meme, is because everyone remembers the 1 time they missed a 98% chance to hit, and no one remembers the 49 times they hit the 98%. So, with that in mind, I think the meaning of the sentence is that you can't just say 'cosmic bit-flip is unlikely and therefore it didn't happen', because you are unknowingly discounting all of the chances it had to happen but didn't. It's not that I cheated to get the royal flush; it's that you don't remember all the games we had to play before I got it. Though, keep in mind, I didn't write his comment, so I'm probably off the mark at least a little bit.
Perhaps the point of this video is not that comsic rays "don't happen" rather the point was "mario 64 is an old glitchy game that has many many many documented instances of glitches, specifically on this one level, that appears to be a glitchy level, and this skip is just a rare glitch from a glitchy level, rather then a unprobeable cosmic ray".... but yeah this video did not make that point at all :)
I’m not sure you disproved the null in this video. Working around spacecraft, we have modeled these events on hardware, and while it is significantly harder the closer to earth you are for such an anomaly to occur, we have documented cases every year of this happening.
Please check the description if you want links to videos and the Ukikipedia pages featured in this video. i highly recommend checking those videos if you liked seeing those weird anomalies.
EDIT: Here's a link to the SM64 TASing and ABC Discord Server for those interested. Many SM64 experts are active there: discord.gg/ECskvyF Also I want to make it clear that cosmic rays have likely caused bitflips in computers like that one election in Belgium. I'm just saying it probably didn't happen to DOTA_Teabag in this specific scenario.
Ok
11 years* you round up from 6 months. we're in march now.
I find it hilarious how you call the glitches "anomalies". Keep up the good work man.
You've earned it 🏆
Hey Lunatic! Just want to say that it’s been great watching your channel grow, especially since you’ve had such high production quality right out of the gate. Can’t wait for that 100k!
....was this not a joke outside of literally the person who suggested it?
Dies in CCM, gets mad, slaps his N64, then goes to re-enter CCM but ends up in BBH instead.
The creepypasta writes itself.
Imagine a kid happening unto that, but in 1996, lolz.
@@glidershowerI literally wouldn’t go to sleep for like 2 weeks if this happened to me (even though I was born in 07). I had a deathly fear of boos when I was little.
Vindictive N64 sends abusive Japanese player on a wild goose chase!
*wild Ghost chase! ;)
Revenge of abused Mario
A cosmic ray traveled 92mil miles to flip a bit at Google HQ so this video got recommended to me.
Ok, but what if I told you it happened TWICE
@@kr1v twice you mean 3 times
A series of cosmis rays typed this comment
Thats crazy, because that JUST NOW happened to me.
Your here
As an aside, cosmic ray interactions are in fact very common when working with some equipment! They're a regular nuisance for Raman spectrometry, for example, as they interrupt your spectra by creating 'spikes' at random wavelengths.
Speedrunners, please take care while running Mario 64 on your CCD detectors.
They’re also a big problem for autopilot software on airplanes, since airplanes have less protection from the atmosphere when they’re in the air.
Whoa, I just made a comment talking exactly about raman spectroscopy! The first time I saw that huge spike I got excited! Then after having seen it over and over again it lost some magic.
Good to know since the lab I work at just got a Raman IR. I'll just blame the cosmos every time a sample fails 😅
Also, computers _do_ experience random bit flips fairly often (it's why ECC exists); just, most of them are caused by less interesting interference.
Sounds like it's Rayman speedrunners who should be concerned
I'm not very far into this video but just wanted to mention that cosmic rays flipping bits in RAM is not an "out there" concept. Server-grade RAM is literally designed to combat this for data integrity.
Cosmic rays being the cause of this is probably what is being questioned. To your point though, there is also something called "functional safety" in ISO and IEC standards. There are dictations to the way code must be written that assumes a bit could randomly be toggled. An example is that there cannot be any "hanging" or dormant code in which it has the potential of being executed even though it would never execute in any expected scenario. Even if that piece of code is commented out, it cannot be present at all if it doesn't belong there.
@@seinfan9 Nah the "cosmic rays are questioned" is just an excuse, do not follow the video creator blindly, the video clearly showed ways to neglect even the existence of cosmic rays and failed at it, the guy talks about it like it was something impossible and magical while it's actually not that uncommon.
I can't believe this video isn't mass disliked due to providing missinformation and confusion just for views and likes, everytime I come back I remember how stupid it is where he shows multiple recreations or possibilities that are already debunked and just says "well cosmic rays don't exist so it didn't happend".
Actual moron he is
@@seinfan9 When you compile code, comments are stripped out, so any commented-out code is not only not able to be triggered; it literally is not in the compiled program, and never would have been even in the past. Modern linkers are also good at cutting out dead code, but ISO 26262 might predate that behavior.
In either case, that requirement sounds less like it's striving for resilience against bit flips and more like it's striving for the code being easier for a human to understand. That requirement is basically saying, "Keep your workspace tidy and throw out things that aren't useful anymore so that it's easier to organize the things that matter."
@@seinfan9 "bits" aren't code, it's electronics
@@jimmymcgee9374 You wanna try that thought again I think you failed several times
It was originally "speed run", until we found a glitch to drop the space, so now its "speedrun"
Space skip
Going for that spedrun any % e skip
saved 1 second off the initial run but not a pb
Verb
spdrn
***WORLD RECORD*** 3-7-24 😂
your opponent getting hit by a solar flare would be one hell of a way to lose a speedrun race
I lost the run to CME, dude!
For real!!
Krillin coming in with a solar flare
@SteveNeubauer smh not even giving credit to Tenshinhan.
PANNENKOEK 🥞🥞🥞🥞🥞🥞🥞🥞🥞🥞
I'm an electronics engineer.
It's definitely unlikely to be a cosmic ray bitflip, but it could be as simple as environmental radiation or contamination in the packaging of the semiconductor chips, or a variety of other semiconductor device physics stuff.
Fun fact: did you know that trace contamination of certain radioactive metals in the solder used to put chips down can cause bitflips?
Interesting. There's a 1996 scientific paper called "Single event upset at ground level" that states that (quoting Wikipedia):
"Electrical or magnetic interference inside a computer system can cause a single bit of dynamic random-access memory (DRAM) to spontaneously flip to the opposite state. It was initially thought that this was mainly due to alpha particles emitted by contaminants in chip packaging material, but research has shown that the majority of one-off soft errors in DRAM chips occur as a result of background radiation, chiefly neutrons from cosmic ray secondaries, which may change the contents of one or more memory cells or interfere with the circuitry used to read or write to them."
I've talked to astrophysicists about this before and they disagree, cosmic rays (especially neutrinos) are very capable of flipping bits and should actually do it fairly often (over years duh) statistically speaking. You and an astrophysicist should get together to come out with a truly informed opinion.
Where do you think the classic static noise pattern comes from? Radiation, yea, but a lot of that radiation is in fact cosmic in nature (fun fact, a not insignificant portion of that noise comes from the CMB.)
And yes, before you go ''ackchually'', yes, i am aware modern electronic devices including chips are built to self correct enough to make that noise negligible, but that design does not cover all energy levels of particles (and even types of particles themselves) that the chip will realistically be exposed to.
@@thelelanatorlol3978 About 100 trillion neutrinos pass through your body every second. With 8 billion people and many many many many more electronic devices, a cosmic particle hitting a chip is not rare at all. Just happens continiously.
hey, weighing in with this is a programmer and systems administrator, bit flips happen constantly, be it from heat, radiation or cosmic rays and are usually caught by inbuilt error correction, the n64 used 4mb of RDRAM running at 400mhz on a 9 bit bus, using it's built in error correction bit for the GPU instead. early versions of this RAM standard had no error checking but the standard used a pretty big transistor which would require more energy comparative to the phone in your pocket or your laptop's modern hardware, so a high energy particle is more likely than you may think.
changing this bit at the right time seems to me like the most likely cause of the error. weather its ionising radiation from space or because the passively cooled memory was overheating or a solar flare is completely academic and unknowable, but IS more likely to be the case because of the energy requirement to flip a bit in that kind of memory module and it is a compelling story, and one I will continue to tell to comfort people when they can't open their word documents due to data corruption
tl;dr we live in a world where ECC memory has made us forget that this isn't even unlikely anymore, it just goes unnoticed because Richard Hamming is an underappreciated genius
edit: removed erroneous assumption about hexadecimal conversion because it was incorrect and not really relevant, I couldn't even remember what I meant by it when it was pointed out. I also corrected myself from "a bit flip is more likely" to "A high energy particle is more likely than you might think" because its what I meant and it didn't really make a lot of sense before
people have also pointed out that I say in the TL;DR it's because of ECC memory and when they google ECC memory they find the Hamming module and find out that isn't applied to consumer hardware, then assume consumer grade memory has *no* protection from errors which isn't the case, every stick of memory since DDR4 has had a Cyclic redundancy check (CRC) and parity bits on different parts of the memory module, and DDR5 has a sort of on die ECC even in it's "non ecc" variants, the ecc variants send additional parity bits to the cpu for a second check (unless you're on intel, since I don't think their consumer CPU's even support true ECC), which will catch any errors in transport, because the density of a ddr5 chip is kind of insane now, it bit flips constantly. there's also all of this on the non-volatile storage side which as I understand it wouldn't be a problem for this error, if there was an unhandled error in the non volatile storage it would be consistently reproduced, and I don't think it could happen at all in a variable because the CPU creates and references these in memory
also wtf are you talking about when you call those clips too different, they match perfectly, except one is running on native hardware with shitty 90s frame generation
@@gunslingerspartan I think his point is that they don't overlay 1 to 1 over each other, but the outcome shown is completely identical.
A rather silly gripe to claim "Not identical"
A flipped bit? In MY N64 game? Its more likely than you think
Fking thank you man. Frustrated me watching this video him claiming "this definitely wasn't caused by a cosmic ray" while citing 0 evidence that it wasn't. When the TAS practically lines up 1 to 1, the bit flip is pretty much what caused it, and again, whether it was caused by a cosmic ray or not is unknowable. But to make an entire video essentially making the claim that "No, it was 100% not a cosmic ray" is infuriating when it's essentially unknowable. ECC memory and error correction is so commonplace nowadays in all software that this essentially doesn't happen anymore, but this did use to happen quite a bit and caused data corruption all the time. It seems like we've forgotten that, goes to show how unbelievably effective these algorithms are at protecting our data from random corruption.
Might be the most genuine and educated comments I've seen on UA-cam since like, 2014.
Hi, Cosmic-Radiation-Bit-Flip researcher here (I am not making that up, that is actually my job). So to keep a long story short: computers are weird and trying to model anything happening on hardware exactly is practically impossible. I disagree with the statement that a cosmic bit flip cannot be the cause, simply as it is very hard to prove that it isn't and it is within the realm of (at least theoretical) possibilities. However, there are two other things we should consider that make the cosmic ray theory very unlikely. The first is the very low amount of radiation we observe on earth. A satellite within the Low Earth Orbit will still only get a small number of radiation events in the mission time, for a device on earth the possibility is negligible. On satellites we still need to consider the effects of radiation if hardware is safety critical, we have a very bad setup (high amounts of radiation expected), or we have long missions and aging effects, but that is satellite mission stuff. Secondly, we shouldn't forget how much hardware just doesn't work. Modern computer chips have a huge percentage of the chip that just doesn't work due to manufacturing problems. If you produce a CPU with 12 cores but only 10 work: sell it as a 10 core CPU, that's how it's done. Thus hardware malfunctions are definitely not unlikely and again really really hard to identify, even if you have the hardware in your hands.
TLDR: is the cosmic ray theory possible? Yes, it is possible. Is the theory likely: No, it's not likely. Even though it's not impossible, the odds are very low. So we shouldn't say it's impossible, but also not act as if it was definitely the solution.
Well said. this mirrors my thoughts, but it bothers me that people are thinking i don't believe cosmic rays can affect computers when all I'm saying is it probably didn't happen in this specific instance
@@LunaticJI do agree with both of these comments I just think maybe next time use your words more carefully if that makes sense
this is just feedback and not a way to make you look bad, just saying it's a myth and it's rare is not best things to say when making an statement, next time maybe not put that people opinions as much and make people look dumb, im not sure it was your intentions and I know everyone has different views just this is what I think. Over all even if I disagreed and thought sometimes you was a bit dramatic this still a good video just needed some clarification.
@@catlance I agree with you. I feel LunaticJ's tone was too dismissive of the possibility. Stating that other explanations are more likely is more than enough to keep the question open and being dismissive diminishes his credibility, weakening his case rather than strengthening it. Another important point is that even very unlikely things happen all the time. With Mario 64 being streamed all around the world every day, we should fully expect to see all kinds of strange outcomes and glitches from time to time, including random hardware glitches, and yes, maybe even bit flips.
@@LunaticJIt's the tone. You sound really... smug throughout this video, and it really comes off as if you are dismissing the possibility entirely. You sound like you are mocking the idea. To be clear, I agree with you that it's unlikely but possible, but your tone was super off-putting.
@@fiddleronthenet3360 It's the Veritasium complex.
I can’t believe Big Speedrun is trying to gatekeep the common tech of cosmic ray manipulation.
Shh, keep quiet, I don't want them to find out I've been manipulating my runs for over a decade with that orphaned source I "aquired"...
(Just kidding. I don't a have an orphaned source and I don't speedrun either. And if I had one, I'd probably be long dead by now.)
Don't. They'll silence us even more than they already are. Keep your thoughts to yourself. There'll be a time where Big Speedrun will fall and cosmic ray manipulation will be a widely used strat.
@c0d3warrior riiiiiight. Just admit you do speed runs in space to manipulate memory
Big Speedrun made my day. Cause the concept is crazy😂😂
@@c0d3warriorAh yes, the caesium-137 strat
Satellite expert here. Bit flips are something we have to constantly design and shield against due to the extremely hostile nature of Space. However, these satalites are orbiting above the protective shielding our earth and atmosphere provide. On earth, bit flips are significantly rarer.
EXCEPT! Our sun goes through an 11 year cycle of rising solar activity (solar flares, sunspots, coronal mass ejections etc.) Eventually culminating in the "solar maximum" before activity begins to decrease again. During the year or so before and after a solar maximum, computer issues caused by the sun are slightly more common. The last solar maximum was in 2014, meaning this run took place during a period of extremely high solar activity.
Btw, the next solar maximum is building up and will be here by 2025, and experts are prediciting that it's going to be a rough one.
Even on earth, cosmic rays can cause hundreds of megabytes worth of RAM data to be scrambled every month. Of course, most modern hardware can comfortably deal with single bit flip events, cosmic ray or otherwise.
@@thelelanatorlol3978the problem is, a N64 is not modern. that leaves the theory open
Satellite GOD here : you don't know what the fuck you are talking about. Please shut the fuck up. Thank You.
A coworker from my first job was testing a component that would end up on a satellite running our software. He would often start running tests on the device, then work on something else while those ran, and would later see that some random test failed. But he couldn't figure out why. He tried catching the failure by running the test suite with a debugger attached dozens of times over night, but came back in the morning to all the tests being green. So this was an enormously frustrating thing to deal with.
How could the tests fail only when he was in the office?
So he tried running the tests in the debugger once again in the morning, and got up to go get a bagel. But the debugger immediately paused on a failed test. It turned out that the slight electromagnetic impulse from his chair whenever he stood up/sat down/shifted his weight was enough to flip bits in the device's cache/RAM and cause a test to fail. With additional EM shielding the failures disappeared.
The local environment can be a much better place to look for sources of spontaneous memory corruption than cosmic rays, though cosmic rays can cause rare bit flips on earth's surface.
i guess 2025 is the time to get into mario 64 speedrunning
Dram engineer here. My job is literally to debug these type of silicon bugs, but there isnt really enough info to determine what is failing in the memory chip. Could be capacitors losing charge due to lack of refresh cycles, unintentionally causing a row hammer attack, bad voltage reads on sense amps...
Most likely, software issue as a bit flip in these cases could happen to any of the million bits in the array. Its weird how the bit flips would only happen to player coordinate.....
That is my impression too. If a bit has flipped somewhere in memory, the chances are good that it would just crash the game because it tries to read from different memory than it otherwise would, messes up a return address or a stack or something. The way that it seems to only affect the position of Mario is very strange and suggests that this is the result of a bug, not cosmic rays.
@@somekindofbox264however that doesnt mean its impossible
@@somekindofbox264and that happens significantly more often than the upwarp
Do you solve practical problems?
@@firebird-two-onegoing off by what's "possible" instead of probable is stupid and conspiratorial
Huh, I'd never thought the community had the idea "It conclusively was a cosmic ray", but rather "We can't quite figure out what happened, and a cosmic ray was a possibility, since we can't figure out how else it might have happened yet."
Someone plagiarized a yt video into a short and said it was the only possible explanation and went viral a few weeks ago
yeah this wouldn't be the first time something similar happened and was observable
i dont remember when exactly but in veritasium's video he talked about the small radioactive particles in intel chip's ceramic caused a single flip of bit 12 in a register which caused one of the candidates for an election to receive 4096 extra votes (noticeable and while it wasn't a cosmic ray the particles emitted from the uranium are the same)
@@kadupse yeah, it could have been just a fun video about "what probably happened" but instead was weirdly hostile about the whole thing.
@@korumannbecause videos of accusatory tones get pedalled more to viewers
Basically LunaticJ does a lot of projecting as he does exactly the same which he blames other for doing. Making things up.
14:00
Mario's position is stored as a floating-point number, which is stored vastly differently than an integer. A bit flip in the mantissa would change the value by a power of two, but a bit flip in the exponent would multiply or divide the value by a power of two. If the Y-position were, say, 1800, a bit flip in the lowest bit of the mantissa could change the value to 900, a difference of 900.
Nice catch, that seems reasonable!
No, stop! You are scaring him with first semester computer-science knowledge!
the part about Veritasium "using the article and video as an example of a bit flip definitely happening"
in his video all he says is that no one has been able to replicate it and a bit flip from cosmic rays is the "best explanation anyone can come up with". does not seem like he was claiming it definitely happened to me.
But this guy ran a blender by his N64 so he clearly knows what he's talking about.
If there is any channel I trust to do the proper research it is Veritasium. One of the most prolific scientific channels on the platform that routinely has experts in their field on to discuss their work.
While watching this video, I couldn’t help but grow annoyed by Lunatic constantly attacking the theory. He didn’t even provide a theory of his own.
The idea at the time is that cosmic rays were the explanation that just made the most sense. And it’s true; it did make the most sense at the time. Aside from the cartridge or console just being faulty.
And even coming out of the video, it isn’t exactly firmly debunked, either. Unlikely, but was it definitely not a cosmic ray? We don’t know.
@@richboy455hate to break it to you, but every other video Veritasium makes is an explicit sponsorship where the science is secondary to promoting the product. Not to say that this makes these videos completely inaccurate or anything, but you should be critical of the media you consume and shouldn't put blind trust in anything on UA-cam, especially in the pop science sphere
@@richboy455unless he is being sponsored to do the video. The one about dandruff, the one about self driving and the one done with Mercedes are utterly corporate propaganda nothing more.
"I refuse to believe it was cosmic rays because that's just too unlikely. It was definitely one of these other options that don't usually lead to bit flips or that we tested but didn't lead to a bit flip"
lmao yes, I dont understand what he's trying to point in this video
The purpose of the video is to say that we don't have an actual answer to that, there are lots of theories and that cosmic rays is the most improbable.
That's literally (not using literally as an emphasis) the thesis of the video.
@@mommyukiExcept it's not. A cosmic bit flip literally affected an election before why is it so unlikely that it affected a speedrun out of the billions of speedruns
A bit flip is perfectly plausible, and in my opinion the most likely. The TAS recreation doesn't match perfectly, of course, but the timing is close enough that a bit flip should have caused that kind of displacement - I don't understand why he completely discards the hypothesis because of a few pixels in difference on what was supposed to be a *recreation* of the event and not the event itself. And with that, I think what actually caused one of those bits to flip is entirely irrelevant. There's a huge array of things that can cause that, especially on old hardware like n64s which don't have as much protection against this sort of malfunction.
@@mommyukiAnd he arrived at that conclusion using absolutely 0 credible arguments. Provided no alternate hypothesis and rejected the currently accepted hypothesis because it’s unlikely. Completely unscientific
LJ: "10 years ago..."
Me: "Oh so like 2003"
LJ "September 21st 2013"
Me: Turns to dust
I honestly have no idea why people keep saying that, although i still feel 2013 like yesterday.
Technology wise, because i always go for low-specs, so when i try something that isn't, i'm amazed at technology's progress LOL.
@@manuelkfc7916the other day i got an ebook and im still baffled at how it works
Same age as Pokémon X and Y.
We get it, we're old. I swear I see this exact comment on every video that has even a singular date. Get over it.
@@Briskeeenrude for no reason
There was an election in Belgium in 2003 where a candidate got 4096 extra votes (2^12) due to a cosmic ray bitflip. It was found and corrected, but strange and unlikely things happen given a large enough sample size and period of time.
From an outsider to this situation, it sounds like a guy tried to cheat then paid media to paint it as an accident when he got caught, lol.
@@Dante02d12 i can see why you would guess that, but circumstances dictate otherwise. It happened within the machine and was immediately spotted and fixed; she was among the least popular candidates out of 10ish, and as a minor candidate on a local election (iirc it was about 7000 votes for the whole race) had no ability to control the machines.
@@alexanderolson6622 Thanks for the added context!
computer calculating voting counts doesnt have ecc ram? nice.
also all ram has parity bit so it should've given error
New speedrun strat: bring your console with you onto the ISS and play up there where cosmic rays are way, way, way more common and reap the rewards
And then it flips the wrong bit and switches you to the game over screen lol
@@thethree13o epic gamer strats, bring 10 bajillion copies of a game into space, run them all at once, and keep them running and restarting until eventually you get a world record time
i just did that and the iss is now falling down , woops
New speedrunning strat: play with a bag of cesium over your N64
ISS orbits below Earth's magnetosphere
How can you possibly look at that TAS bit flip and not think that is EXACTLY what happened?!?! The only reason it's not a perfect 1-to-1 match is the not-perfect starting positions. But the height climbed is literally identical and Mario lands on the platform at the same time. Sure it may not be cosmic rays, but there's no way that wasn't caused by a bit flip.
Also, with the downwarp later in the video. Why can't it be a bit flip? In floating point numbers there are ways to get non-power of 2 shifts (flips in the exponents), or there could be a frame correction where Mario's position is shifted because otherwise he'd be in the middle of a block or the ground etc.
Bit shifts happen all the damn time, and for so many reasons. Simply having a bunch of other bits changing around one can have effects. It could well be that this bit location is in a hotspot location, and then cosmic rays pushed it over the edge. Every tech company realizes that cosmic rays can flip bits, so while it is true that the myth was never solved or proven, it is definitely not ruled out.
EXACTLY
I don't think anyone's arguing that it isn't a bit flip(i hope not anyway), I think they're more arguing that it wasn't a cosmic ray. BUT I think its decently likely- not really a cosmic ray, but radiation of some kind.
Im not sayung it's cosmic rays but i just wanted to point out that cosmic ray bit flips are insanely common (unless you use ECC memory, they still happen but they get corrected 99% of the time). Like they happen to computers at an hourly rate, and if you gain elevation above sea level they massively increase in frequency. Aviation and Space deal with them by having so many redundancies and protections that the bit flips dont matter much. But a Nintendo 64 has no such protection, its bits will be flipped. Its not a matter of if they will happen but when. But this rarely (as in on an individual basis, on a provincial/state level they have caused issues all the time we just account for them nowadays) results in much issues let alone a bit flip like we saw.
That said, this isnt a fact of the situation; cosmic ray bit flips may have had nothing to do with it. But the fact that the height change from one bit flip was nearly identical it feels safe assuming its a bit flip of some kind. But i wont say it as a fact since there is no such thing as 100%.
The electrical surge segment reminded me of a clip from Runnerguy2489. The framerate in his game chugs briefly at the exact same moment that lightning struck outside his house.
EDIT - Never mind, just saw the clip and it's a stream lag rather than a game lag.
Oooh that's interesting - iirc the N64's CPU runs with a 1.5x multiplier on the base clock frequency. If Pin 112 is 3.3v, then it runs at 1.5x, however if it's grounded then it runs at 1x. I wonder if the strike somehow dropped the voltage enough on pin 112 that the CPU dropped to it's 1x base frequency, which resulted in the framerate chugging.
"Was that lightning?" I think it was called.
@@kargaroc386Just saw the clip, looks like it's a stream lag rather than a game lag unfortunately :(
You can see the webcam freeze too, and immediately after the unfreeze you can also see a slight flash in the door frame.
@@Jademalo the N64 does have a rare bug in its clock generator(?). Maybe once every several thousand times you turn it on, it runs at half speed. The audio is all slowed down and spoopy and the video is a weird static (TV not properly syncing to a very-off-time signal). There are a couple clips of it on Twitch.
While that evidently didn't happen here, I wonder if it ever could happen while the game is running? It wouldn't be helpful at all, though.
@@Jademalo Heh, if it's Stream lag maybe the lighting still caused it, maybe via packet loss from interference somewhere
Okay yeah, but also the phrase: “Sun assisted speedrun” goes really fucking hard. It’s highly likely that it’s not true, as you’ve already mentioned, but god it sounds cool.
Gotta get some of then neutrinos
Just like aliens*, big foot and whatnot, we may keep making memes about that while ensuring people are informed that they're not real.
*Aliens are real; but they're nowhere near Earth and we have yet to find any.
Tfw you RNGmanip the SUN.
@@0Cazador RNGmanipulating sun so your crops get sunlight for the entire day
Vertassium (I think that's the channel spelling, or at least close) has a great video on bit flips
Am i too drunk or did he talk abt how flips don't cause displacement then shows a clip of flips displacing a piston, then saying that rays are a slim possiblity and at the end saying that it flat out wasn't reys because the mystery isn't solved?
Like if the experts couldn't exactly replicate it without flipping (which is what rays do) and the only perfectly matching video is flipping then wouldn't it mean that it was rays?
Since even the hardware wasn't faulty
Again i may be too drunk but this video is not making sense to me
The video makes so little sense even someone who was drunk could see the obvious fallacies. Lmao.
He never said flips don’t cause a displacement. A bit flip was probably the reason why the upwarp happened, but it’s MUCH more likely that it wasn’t cosmic rays. Think about how many bits are in an N64. 4 megabytes of RAM, 4-64 megabytes of storage. Assuming it was cosmic rays, how likely do you think it is that they somehow hit exactly that one bit at that time? Do cosmic rays flip bits constantly? No. So it’s incredibly unlikely that cosmic rays flipped that specific bit at the perfect time, especially considering the evidence of cartridge slapping corrupting the game and how he typically had to put in the game cartridge in a weird way, implying that his hardware could be the issue.
@@somenerd8139 1) You're assuming cosmic rays didn't flip other, less noticable bits. 2) You're assuming they don't do this all the time. 3) You're assuming that because a bit flip has meaning to us as human beings, it means that it is a more special and meaningful action than a bit flip that wouldn't be as noticable. 4) You're ignoring every other recorded SM64 playthrough where this DOESN'T happen, and focusing only on the exceptional playthrough where it did. 5) Why didn't the cartridge slap/corruption have many more noticeable effects than just a single upwarp?
@@EntitySteel meme theory being regurgitated by tourists. What is new
Until I see definitive proof that outside rays didnt cause a bit flip, I'm sticking my N64 in a microwave at the start of TTC and hoping for good IRL RNG.
gl bro wish you luck 🙏🏿
Gl man wish you the best
But microwaves aren't ionizing radiation 😂
@@R3SerialDreams 🤓
@@R3SerialDreamssimply place some uranium ore on the console then.
What you didn't mention is that most of the articles and some of the memes say the warp "saved time" or would have saved time, but it doesn't even do that. He was going for the red coins and it only moved him away from them and would have forced him to move back down to where the platforms were
Yeah he went for "Get A Hand" as a backup. If he got that star already, it certainly would've lost more time
Yeah the whole reason there was a bounty in the first place was for future use in A Press challenges or speedrun categories. The glitch is still in the first phase of speedrun metamorphosis but articles are treating it like a crazy time skip for a wr pace or something.
@@beangorl7005if we had a cosmic ray gun that could change bits on demand it could be used for a WR run.
But it could help if there was some way to replicate it elsewhere in the game, I imagine
13:21 Guys, he tried to run a blender next to the console and nothing happened. I'm convinced. No more tests needed.
I really don’t understand why that is in the video. Does the editor think that a food processor/blender has significant electromagnetic radiation?
Does bro get his food processor from aliens?
@@zaxtonhong3958its to try and cause a power surge some high power blenders are known to cause them, not saying it was a good test of course
New M64 category: 70 stars w/ powersurge
@@zaxtonhong3958 Blender%
I NEED MORE VOLTS thousands millions billions trillions
"The idea that a cosmic ray flipped a bit in a computer is highly unlikely."
And yet, ECC memory is used in servers all around the world for... what, fun?
Yeah, insane to just rule out cosmic rays like that. I remember going about universities when I was younger, and it was almost guaranteed there'd be some demonstration about detecting cosmic rays with some spark chamber so you could visualize them.
@@plukerpluckit is so funny to me that people look up at the giant ball of flames in the sky that warms up this entire planet despite beings millions of miles away, one of literal infinite more of them in varying sizes from massively smaller to way bigger and think “nah nun’ gettin here”
There are simply way more common causes of it than cosmic rays. The N64 is old hardware with big transistors that take lots of energy. There are a million electrical issues that could have caused it. Hell ECC was first invented to handle the reading of punchcards and cosmic rays weren’t relevant.
Thought this was gonna prove the bitflip theory wrong but it just goes over all the likely theories and in the end the bitflip is the only one that has had any sort of evidence that could be reproduced, and you even bring up the dupdome instance which only reinforces the idea of a bitflip caused by a cosmic ray. In dupdome's scenario it is far more likely that a hardware error was in effect, as there were multiple bitflips occurring in predictable ways, but in the dota_teabag scenario we've taken apart the N64 and cartridge itself and in your own words "nothing notable popped up." Faulty hardware can be ruled out completely due to those tests. As for cartridge tilts and slaps, nothing in the dota_teabag clip indicates any sort of sudden harm to the console like in the atogami clip. In every other example shown of faulty cartridge connections, the interference is far greater than the upwarp and anomalies occur for far longer, animations and textures glitching out for several seconds at a time that isn't in line with the sudden, one frame change we see in the dota_teabag clip.
Every alternate scenario you've presented has other caveats and such that make them differ too greatly from the dota_teabag clip. Faulty hardware has been disproven by further testing. Faulty connections cause wildly different scenarios from the one we are trying to recreate. The bitflip is the ONLY scenario that has shown a correlation to what actually happened in the clip. It is the only scenario that actually hits the mark, and no matter how unlikely a SEU is, the results suggest that it's still more likely than anything you've shown. Hell, the idea that cosmic rays are "unlikely" is even questioned in your vid. You say yourself that bitflips usually don't effect a computer in "a visually observable manner." This implies that they happen more often than we have observed, thus making our own perception of how likely they are to happen to be inaccurate. Which, of course, they do. Hardware and software usually just has ways to handle errors that the end user doesn't perceive, like simply crashing.
I'm not gonna argue that poor journalism leads to myths, but if YOU are, then you should be doing your due diligence. You said that Veritasium used the clips as an example of a bitflip "DEFINITELY happening thanks to a cosmic ray." This is objectively incorrect and can be proven wrong by simply watching his video. His actual words were "the best explanation anyone could come up with is that a cosmic ray caused the glitch," which is a perfectly reasonable take that still leaves room for doubt. If you're gonna criticize him for poor journalistic integrity, criticize him for getting pannen's name wrong, don't shove words in his mouth. The fact that you're willing to put such blatant misinformation in your video calls in to question your own integrity.
tldr, bitflip is still the most likely scenario as nothing else has come close to replicating it. Cosmic rays affecting hardware is more common than we think. You should stop making up quotes to make others look bad.
This should be at the very top of the comment section
Perfect description of what's wrong with the video IMO.
Essentially saying 'the cosmic ray bit flip theory "simply isn't true" (his words)' while providing evidence that a bit flip is the most plausible theory for that exact glitch is wild to me. Trying to call out bad journalism through _this_ video of all things is just peak irony.
Adding another comment here to try to push this comment to the top. People tend to open comment reply sections with more replies, and upon reading the replies are psychologically more likely to actually read the long original comment, and thus more likely to like said comment, pushing it further to the top, where more people can see it as it rightfully belongs to be.
Doing my bit to push this comment up.
This
I think this video adequatly covers ways other than cosmic rays this could have happened, however, it does not adequatly disprove the general idea of a bitflip to a degree that justifies the language used.
Came here looking for this comment. If it's not solved, you can't say it WASN'T a cosmic ray just like you can't say it was.
@@paytonhearn2502 How else would he get to bring up his tired "ethics in gaming journalism" points?
It's not even relevant here too
@@paytonhearn2502 that’s exactly how science works. If there is a likely solution backed by evidence(the faulty console) and someone comes up with another solution which is orders of magnitudes less likely, the science community doesn’t go „we can’t determine which one it is because both are possible“. They go „There is a very small change it _did_ happen like that, but the general consensus is that it happened like this“.
If it wasn’t that way we’d have to consider that gravity might not exist because there is a possibility that all things just fall downwards through random quantum tunneling.
@@foximacentauri7891 What evidence? And how do you define which solution is more or less likely? If I was paying attention, the video said that upwarps have not been proven to happen because of faulty consoles. If that's the case, it's not obvious to say that faulty console is more likely source just because it can create other types of glitches.
At least the bitflip has some kind of experimental evidence to back it up since the TAS recreation was close enough to not rule it out(i.e it was *really* close to what we saw).
So we have one hypothesis about cosmic rays that can and was tested and it resulted in good but inconclusive evidence, and another hypothesis which has no evidence and is untestable and is basically just "faulty console can result in any kind of glitch, so it's probably that even though we don't know how". Either one could be the reason, but neither is "orders of magnitude less likely." because we lack data to make such claims
this happens in science journalism all the time. A journal is released where the author of the paper states that the study MIGHT show [xyz] to be true, and then people go around saying that [xyz] is DEFINITELY true and is PROVEN BY SCIENCE.
Science is dead, it used to be about finding the truth, now it's "take my word for it and TRUST THE SCIENCE or you're an evil conspiracy theorist"
"the science"
That’s why science journalism articles are no credible sources for dissertations and papers and why scientists read the studies itself.
Doesn't help that terrible science education has left the vast majority of the population believing that science "proves" that things are "fact", rather than just being the process by which we eliminate impossibilities through repeated observation and analysis.
Here's my life hack, look for the words "peer reviewed".
Peer-reviewed = the science version of "pics or it didn't happen"
Bit flips caused by gamma rays happen much more often than you would think. The term "cosmic ray" used in English is offputting and makes it sound outlandishly unlikely. "Background radiation" is more accurate as it originates from space, earth, radioactive decay or any other thing producing exotic particles. Try turning on a geiger counter at home and see for yourself. Just because something sounds unbelievable to you doesn't mean it didn't happen.
that cuphead journalist is never going to live that down.
The guy didn't even review the game, and even had positive impressions of it. Yet here we are years later using him as the frontliner example of "bad gaming journalism" because he was bad at a video game once
@@supra_sr Also worth noting that he wasn't a "regular" (for lack of a better word) gaming journalist, he mostly covered the industry-side of gaming, not gaming itself.
@@supra_sr Gamers don't let skill issue slide, you know the rules and so do I
@@supra_sr deserved
Have they finished the tutorial yet?
6:00
it's not nearly as rare as you think. just, most systems are memory managed and almost 0 bits of your memory are system critical so you don't ever notice.
I've actually tested this with older ssds and have a few examples of where a single bit of a bios rom chip got flipped and disabled a whole computer. lucky for dual bios and flash back.
this is the biggest fault with the video imo. the experts that are cited are always speedrunners or tas engineers, never a person who is actually an expert on gaming hardware or storage or physics. if a bit flip seems plausible, then why not leave the video at "this is still unsolved" instead of "cosmic rays are impossible to have caused this"
@@angrymurloc7626Thing is, based on the other TTC anomalies experienced by a different person, cosmic rays being the cause gets less and less likely, bordering on being impossible
@@angrymurloc7626 He stated it was possible multiple times throughout the video, he was just rejecting the games journo myth that it was conclusively proven by a guy who's name they couldn't even get right since they're all too busy chasing clicks to worry about making sure they're not just repeating the same falsehoods.
I didn't really mention that when I was pulling bios off of boards to check them against clean bois install, I regularly found dozens of flips per year running. most of them were inconsequential but I'd be willing to bet the 20yr old problem of hardware just not working one day is somehow related. a large portion of your bios is just channel controllers for hardware.
@@valseedian After 20 years, I would suspect capacitor issues.
This video really frustrates me. It keeps saying that cosmic rays are bad and implausible because of [some reason] (probably because it's fancy physics), so some mysterious other error has to be causing it. while i myself think that it's more likely that it's just faulty hardware flipping a bit for some electricity reason (N64s are old), it's just not rigorous or well explained at all. there's not even a single number on how rare cosmic rays altering bits on earth is, how often N64s fail, etc., there's nothing. it's just a cycle of "here's something kinda similar that happened that dosen't look like a bit flip" into "experts think it's dumb" repeat.
Yep. It is CLEARLY a bit flip from the TAS replication. I can't see how you could look at the replication and not think that. At which point the only question is what can cause bit flips, and this video basically doesn't go into that detail at all.
Bits flip occur because of hardware degradation, electrical interference, random imperfections in the chips, or radiation. The last of those could be cosmic rays, which is one of the more common ways bits are flipped, and is the reason that all enterprise level hardware uses error correction to detect and fix those flips.
It’s just shitting on a cool theory for no reason other than “ackshually 🤓”
I like how you applied a double standard in how you use the fact that a bit flip didn't match exactly as evidence that is wasn't that while using equally dissimilar events as evidence that it was something else.
The question we should really be asking is what does he have against the bit flip theory
I can't read this sentence, my brain won't let me.
@@terrybuckley2850 fr tho why does this guy sound so upset abt the cosmic ray thimg
this sentence is unreadable
What’s crazy is that it basically looked the exact same just that one person was moving while the other one stopped
I have no idea why the SM64 glitch happened (the cache thing makes sense to me), but I don't know who told you that cosmic rays flipping memory bits is "rare". High performance servers, the main reason you pay extra money for error correcting RAM is in no small part because cosmic ray bit flips aren't terribly uncommon, particular if you have a lot of RAM. I'm not saying they happen all the time, but they absolutely do happen, and if you have data that you want to protect you need to actually consider it. Data centers use error correcting RAM specifically for this .
Now, obviously, the N64 *does not* have a ton of RAM, but considering how many millions of hours have been spent playing SM64 it feels like a cosmic ray bit flip at some point is inevitable, considering that the N64 doesn't have error correcting memory (I'm not even sure it had been invented yet!).
It hadn't been invented yet because bitflips weren't as common. Memory nowadays is built on a much smaller process node, where less current is needed to flip a bit. More efficient, also more susceptible to bitflips from cosmic rays or quantum tunneling. Even non-ECC memory has some error correction nowadays.
This is wrong, parity in memory is as old as memory. IBM had commercially available ECC RAM in the 50s
I really don’t think you know what rare means if you think saying “they don’t happen all the time but they DO happen” somehow contradicts the claim that it’s rare.
You’re basically saying “it’s not rare but essentially it’s rare”
@@zzodysseuszz no, I am agreeing it is rare. I am claiming that given enough hours of playtime an unlikely event becomes basically inevitable. The video had the tone of “it’s rare so it didn’t happen”, and I am arguing that even if they are rare, there have been untold millions of hours playing Mario 64, on systems that don’t have error correcting memory, so it wouldn’t be *weird* if it did happen eventually. Law of large numbers and whatnot.
@@zzodysseuszz The claim in the video is that it's practically impossible for a cosmic ray to flip the bits in an N64, despite the fact that it's a very well-known phenomena that happens plenty of times on other kinds of computers.
I will weigh in and say that it's extremely unlikely for a faulty cartridge read/cartridge tilt to cause an up-warp. There's a reason those glitches in games are often limited to animations, models and sounds/music freaking out. Cartridge tilting tends to not affect things that are always loaded in memory, such as the game's code and its associated memory. It would be very unlikely for that to be the cause of an upward warp. That's an effect you would only see if you used a ROM corruptor for amusement.
I wouldn't dismiss the bit flip theory entirely, for that reason. The only other viable option at that point is some sort of undiscovered bug/quirk in the game's physics. I'm not saying the bit flip is necessarily caused by a cosmic ray, but it's exceedingly plausible that a bit was flipped to make that happen.
DOTA_Teabag made a YT video(a YT short for some reason?) showing the game crashing with Mario doing the classic corrupted animation with corrupted music. So it seems like hardware issues is(or are?) the culprit.
9:57
Yeah I don't see how tilting the cartridge would affect the mario position variables stored on the N64 ram. I mean I guess hypothetically it could read a faulty instruction of the cartridge that would do that, but that's VERY unlikely.
@@Sad_Cirno Except it doesn't. Logic on N64 cannot be executed directly from ROM, it must be loaded into RAM first, and if you don't set up certain memory segments correctly, only the first megabyte of RAM can ever be used for executable code. Additional logic is never loaded in the middle of a level, only during transitions.
The actual impossible hypothesis is that it was a cartridge tilt. Mario's height is stored in RAM, not ROM, and is only modified by code that was already loaded at that point.
I literally stopped watching the video to see how more than 90% of the comments against argued in a polite and plausible way, I have rarely seen this, epic
nah fr everyones pretty chill with disagreeing with the video and the like
I dislike it
They may be respectful but some are delusional, (unintentionally) making it easier to argue against it by twisting his words (strawman).
I'd rather live in a world that mysterious waves of energy randomly travel across the universe to flip bits in my electronics than a world that doesn't
Good news for you, this guy has literally no idea what he's talking about and didn't disprove cosmic ray bitflips as a likely candidate in any way. To be honest he deconstructed his own argument, which was weird, by showcasing that a bitflip DID happen (The footage, despite his proclamations, matches perfectly. Of course when you overlay them they look different because 1. one is on an actual n64 and 2. the inputs leading up to the bitflip are different.) but also by showing that in competing theories there tend to be a LOT more bugs throughout the run than a single bit flipping.
W gold truth
@@EntitySteel no, he basically just has a problem with everyone accepting the cosmic ray theory as fact when 1) it wasnt ever proven or solved, and 2) there are MUCH more probable causes that no one gives credit to, simply because cosmic ray sounds cooler.
@@dogsbecute Except no one's accepting it as fact, and he's already considering it a "myth," simply dismissing the idea while refusing to acknowledge that it's neither been proven nor *disproven.*
@@yeetrepublic9142 .... im sure we didnt watch the same video, because he certainly acknowledged its existence and the probability compared to literally every other example he gave, as well as why it isnt so cut dry. Also, yes, people are literally taking it as a fact and are malding at the possibility that maybe...JUST MAYBE...it wasnt actually a cosmic ray flipping a bit. He even went into depth explaining different technology advances to prevent such a thing, which literally acknowledges its existence.
The repeated disregard for cosmic rays as a possible reason is unscientific, given how much they happen all the time. I suggest going to a lab that shows it in action, its actually quite fun, but on the more serious side, space agencies definitely have to account for it in their computers, where coding errors are accounted for and auto-repaired to insure failure is minimized as much as possible. The rarity that the rays specifically hitting computer bits is fine, given it's not just likely, its statistically guaranteed when enough people play it, especially for many hours, days, years.
Hell, the existence of ECC RAM alone debunks half this guy's video
Teacher: "Why didn't you do your homework"
Me: "A cosmic ray ate my dog"
@@TheGameMakeGuy after that the rumbing started
I spelled dog backwards and he created a reality without homework.
@@abloogywoogywoo Everyone deserves to live in that reality. I don't think I have to do the thing I do in school at home lol
"why were you drunk driving?"
"a cosmic ray bitflipped me"
"A cosmic ray deleted my presentation"
Definitely a case of “technically possible” getting warped into “100% confirmed”. But look, either I’m an idiot who accidentally put a dirty bowl back in the cupboard, or someone broke in and put a dirty bowl in the cupboard. Both are possible, guess we’ll never know.
Probably not even technically possible. Cosmic rays are pretty common, and I've never heard of a confirmed bit flip as result.
someone broke in 100%
@@gloweye Bit flips from cosmic rays are very common. They just don't usually result in much as modern computers have ways for correcting the flip. Especially if you go above sea level, aviation and space get it the worst and it used to be a constant problem.
Copied this from a blog post by some guy (john d cook). This incident is well documented:
.
.
.
.
Radiolab did an episode on the case of a cosmic bit flip changing the vote tally in a Belgian election in 2003. The error was caught because one candidate got more votes than was logically possible. A recount showed that the person in question got 4096 more votes in the first count than the second count. The difference of exactly 212 votes was a clue that there had been a bit flip. All the other counts remained unchanged when they reran the tally.
It’s interesting that the cosmic ray-induced error was discovered presumably because the software quality was high. All software is subject to cosmic bit flipping, but most of it is so buggy that you couldn’t rule out other sources of error.
Cosmic bit flipping is becoming more common because processors have become smaller and more energy efficient: the less energy it takes for a program to set a bit intentionally, the less energy it takes for radiation to set a bit accidentally.
@@gloweyeWell that's just a matter of you not hearing about them
This video is made well, however the arguments used dont really hold up. The main things we know are this:
-Mario teleported up, tests have proven that this is almost certainly caused by a bit flip
-the cause of the bit flip is unknown
-cosmic rays can cause bit flips
So really it could have been a cosmic ray, it also could have been a different interference. The fact is that no one knows how it occured but the cosmic ray theory is possible, just quite unlikely.
But your point seems to be that it couldnt have posssibly been a cosmic ray, when it would make more sense if the point of the video was to discuss how the cosmic ray theory, while still being one of the possible causes, is not the only possible cause.
Finally, a reasonable comment
The bit flip from C5 to C4 actually looks quite convincing. The footage not matching 100% could be caused by a mismatch of input or velocity (only mario's position would have been tampered). However there can be several causes for a bit flip.
I like the general spirit of the video, reminding everyone that we have never been able to tell for sure that it was a bit flip and that it was caused by a cosmic ray. However I think it goes a bit too far in the opposite direction as well. Cosmic rays are very real, even on Earth. In fact, that's one of the reasons why we use Error Correction memory sticks (ECC) on servers.
The way he talks about cosmic rays is like that of a person that doesn't belive the earth is round and that there are other planets in space, he is just neglecting the existence of something very real and pretty "common", I feel like he just reads "cosmic rays" and thinks it's some weird sci fi shit
@@superbeta1716 It kind of reminds me of this low-quality youtube channel about an ichthyologist (fish biologist) whose entire pitch is basically "if there's something cool that lives in the water, it doesn't exist"
The marginal difference between the C5 to C4 bitflip and the speedrun event in question could easily just be caused by the faulty cartridge.
So say a cosmic ray DID hit this man's unique, janky, half-tilted game... the C5 bit flips to C4, and we see slightly different results because of slew of other possible unique oddities/issues going on within that decades-old faulty hardware.
Whether or not the bit was flipped by a CR or some other event is fairly irrelevant, but at the end of the day Cosmic Rays are the only easily-accounted for phenomenon.
You KNOW the earth is constantly being bombarded by cosmic rays, so you KNOW they were present and while unlikely to strike at that exact moment, they very well could have been the culprit despite the statistical odds.
Meanwhile, everything else is purely speculation.
DID his faulty N64 have a fault that would cause the event without some outside interference like CRs?
We don't know.
Was it simply have been some electrostatic discharge from nearby construction?
We have no evidence for that specific scenario.
Was it a simple power spike or electrical anomaly in the home or device?
Again, we have ZERO conclusive evidence to suggest that was a factor.
But do we have absolute proof that cosmic rays were being blasted at that area of the Earth at that time?
Yes.
Because the same is true for literally every square millimeter of Earth's surface at all times!
So while it's not likely, we DO have solid evidence that it's a DISTINCT possibility.
And as such, it can't be ruled out.
@@TheKhopesh Another interesting piece of information related to this is a 1996 scientific paper called "Single event upset at ground level" that states that (quoted from Wikipedia):
"Electrical or magnetic interference inside a computer system can cause a single bit of dynamic random-access memory (DRAM) to spontaneously flip to the opposite state. It was initially thought that this was mainly due to alpha particles emitted by contaminants in chip packaging material, but research has shown that the majority of one-off soft errors in DRAM chips occur as a result of background radiation, chiefly neutrons from cosmic ray secondaries, which may change the contents of one or more memory cells or interfere with the circuitry used to read or write to them."
In fact, pannenkoek himself brings up three reasons why the footage may not perfectly match up, which are stream lag, n64 lag, and choppy footage.
The footage you showed comparing DOTA's run to a TAS isn't even identical before the upwarp occurred, and the only difference after the bit flip is a slight change in the camera angle. The fact that, in both cases, DOTA and Panen reach the same height and hit the ground at the same time should be clear evidence that a bit WAS flipped. Even if it wasn't cosmic rays, the height that DOTA got in his speedrun was exactly the amount you would get switching the leading byte of height to from C5 to C4.
This seems like a concrete explanation.
This. And as the editor noted, a bitflip can happen because of the noise in the powerline.
I noticed that when mario took the same damage
Yeah it seemed exactly the same
Thank you. Personally, I also dont find it too far fetched to think that with the millions of computers/consoles etc. running all the time, a cosmic ray could hit a person playing a game. Yes its improbable, but not impossible
"The clip of the bit switch and what happened is actually slightly different so it couldnt have happened" he says as the two clips match up perfectly except for the frame interpolation and fails to give anything else than even resembles the clip.
They’re so similar it left me confused as to why he was claiming they were different.
Exactly, either if it was a cosmic ray or not, its hard to believe it was anything other than a bit flip.
nope.
IT IS SLIGHTLY LONGER BEFORE DAMAGE IS TAKEN
NOT BY MUCH BUT THEY ARENT IDENTICAL
weve moved on from fucking click bait to lengthy bullshit videos, it seems. The worst part, i dont think buddy is in on his own scheme.:(
@kevinnone4728 This is not new. Now the entire internet is filled with pseudo intellectual content creators and audiences alike.
anyone claiming it is 100% not a cosmic ray is just as infuriating as anyone claiming it 100% is a cosmic ray. You can say it probably is or isn't all you want. But if you say 100%, then any slight probability of the opposite is enough to disprove the thesis.
Also, something i just thought of while reading other comments, who's to say it had never happened before DOTA? The vast majority of people playing Mario 64 for at least a few years if not longer would have been kids with either no way or no interest in recording. Just because we haven't seen it on camera doesn't mean it hasn't happened to some kid in 1997 who thought it was just a weird quirk of the game.
yes I agree, and what‘s bitter is that in the video itself he says it. We now still don‘t know what exactly caused it, there are many factors including the cosmic ray "theory", and like he said:"that a cosmic ray flipped a bit is highly unlikely", but then proceeds to at the end say that it is just false (so 100% untrue).
>complains about people taking "it could've happened" and making it "it did happen"
>takes "it probably didn't happen" and makes it "it didn't happen"
>refuses to elaborate
>ends video
Let's be real though A cosmic ray traveling 8 light minutes through earths electromagnetic field into some guys house and into a very specific part of a circuit board at the exact perfect time causing mario end up on a higher platform isn't very likely.
@@lol-de4lo That's like saying a dart hitting an exact spot down to the millimeter isn't that likely. But throw a dart in hindsight, and wow, look at that. At exactly 1:13:004s I hit exactly 10.413cm by 5.143cm on the dart board, even though the probability of hitting that exact spot should be impossible!
His smug tone while fighting misinformation with more misinformation really makes me angry.
@@lol-de4loYou're missing the point. Your devices get walloped by cosmic EM radiation all the time. Radios have static for a reason. The only reason computers don't glitch all the time is that they do, but programs handle errors like that all the time. And when they do fail, it just crashes. The fact that this glitch was blatantly visible is the only difference.
@@Darkra98
I feel like those are very very very very different levels of unlikely/impossible/improbable & aren't comparable in the slightest.
The BitFS downwarp being around 900 units doesn't automatically rule out a bit flip, because Mario's position is stored as 32-bit floating-point numbers, meaning the highest bit represents the sign, the next 8 represent the exponent, and the remaining 23 give the mantissa, to make up a number of the form (basically) mantissa * 2^exponent. Flipping one of the mantissa bits will add or subtract a power of 2, but flipping one of the exponent bits will *multiply* the previous value by some power of 2, meaning if the Y coordinate started as -900 and the last bit of the exponent was flipped from 0 to 1, it would become -1800, resulting in a 900-unit downwarp. I know the lower sections of BitFS are below Y=0 (because of the A press save on Wii VC) but I don't know if the coordinates perfectly match the clip
I thought this was a solid proof it is not cosmic rays. I'm disapointed, it could still be cosmic rays, it is not ruled out
hahahaha he had to change the thumbnail
Exactly what I was thinking, he makes it sound like it's been for sure disproven but people still don't know what caused it. Went back and watched the Veritasium video and he does a great job at saying it's just a very likely scenario not a proven fact. The guy in this video act's like cosmic rays flipping bits is some outlandish thing but yet if he actually watched the Veritasium video he's criticizing he'd know it actually happens way more often than one would think, normally just not in such a visual way. I definitely agree that a lot of video game journalist do a very poor job at getting the facts right and it sucks, but calling this the biggest myth in speedrunning history when it's still a very plausible theory is just disingenuous.
The chances are very low, but it can't really be fully ruled out. It would be amazingly lucky for it to happen at that exact moment, but not impossible.
What about the fact that by RECREATING the bit flip it didn't match? That's pretty solid evidence.
@@blueberrimuffin6682 Iirc, it was pretty damn close, no? And couldn't that be up to user error? Even for someone so experienced, I feel like it would be hard to replicate it perfectly. Unless there's some more tech I don't understand- I'll admit I'm not very in to this space.
"UHHHH WELLL, YOU CANT PROVE IT WASSSSS" Yeah, and YOU cant prove it wasn't, collar boy.
That’s a really shitty argument
You can't prove it wasn't a bug in the code either a ton of people looking at it and finding no such bug doesn't mean there isn't some insane edge case no one missed .. it's quite hard to prove something "wasnt" something
@@LiEnby thats the point. its an unprovable occurrence. it COULD happen. and also, i dont wanna be that bitch, but they HAVE decompiled the source code to this game. so, get a lookin :3
Devils proof
1. A bitflip seems to be the most plausible explanation to the upwarp considering how well it matches the movement of mario.
2. There was no better explanation given in the video. Just speculation about faulty hardware with no technical details as to how they could have caused a bitflip. Also all the examples mentioned about funny things happening in Mario 64, that cannot be explained by a bitflip are completely pointless. We all have experienced one glitch or another even casually. We know gaming systems aren't perfectly designed.
3. As for the reason the bit flipped: I don't know. Since there were similar discoveries on Tick Tock Clock, maybe the answer lies within this level.
However I want to point out, that cosmic radiation is not as far fetched a theory as it is made to believe here. Especially if you assume this being the only instance in recorded Mario 64 history where the only good explanation was a bitflip. Since it could happen at any moment in time, even if it is highly unlikely, after collectively pouring enough hours in, someone somewhere in the world would experience a bitflip with visual effect sooner or later. There could be happening many we wouldn't notice until finally one did.
Simply put: how likely is it to never ever happen to anyone recording Mario 64? Should we be dismissive when it eventually does?
does comic rays explain why bit errors only recorded in tick tock clock so far 4 fucking times instead of the other like 23 other stages and openworld that are likely played much more.
Sounds to me like it's not fucking cosmic rays flipping bits it's just tick tock clock.
@@TheKingOfApples100 bro I've seen you comment this exact comment on like four people's replies. Are you okay? Like don't get me wrong I kind of agree with you to a degree but that is a really aggressive copy paste on such an unimportant topic.
@@TheKingOfApples100signs of mental disability:
@@papascronch so people responding to see it if you don't reply to the person's comment they have no reason to click on the video again this way they see it if they have notifications
@@TheKingOfApples100the comment you've responded to doesn't mention cosmic rays, just that it could have been a bit flip, which could happen due to reasons other than cosmic rays
If anything this video is a good argument FOR the cosmic ray theory (well the bit flip theory, as editor points out there are alternative causes of a flip).
The examples of loose/dirty connections all lead to glitches caused by misreading of data from the cartridge which is different to the upwarp glitch. Hitting the cartridge when falling off the slide caused the system to load the wrong level's address for the painting. Bad connection causes audio/video data read by system to be garbled. Neither of these are similar to Mario randomly either gaining incredible speed for a tick or warping position, where is the memory read that would cause that.
The examples of similar warping/glitches on that level that could be caused by bit flips were good evidence until you said they were reproducible by Dupdome. You specifically say in your video that it was likely a hardware fault with Dupdome's N64. This is unlike Dota_teabag's case where neither he nor the person who bought and took apart his N64 could replicate the warp on it.
Also, you repeatedly said how it was so unlikely that a cosmic ray could have caused the bit flip, therefore that cant be the cause. But that discounts the sheer number of hours that people have played this game live on stream. Especially once you consider that it didnt have to this game, it could have been any game that is speedran and streamed where a bitflip could cause a stunning glitch that cannot be reproduced. Its similar to how you should be suprised and amazed when you win the lottery, but you shouldn't be surprised when SOMEONE wins the lottery.
Also, he never showed any calculations to prove that it's unlikely
@@chrisgaming9567 That's because if he showed the calculations, it would show that computer systems are bombarded to shit with cosmic rays that can flip bits and fuck around and the nintendo 64 is old enough that a lot of the modern (relatively speaking) redundencies just do not exist within any of the components.
@chrisgaming9567 I get that components that would mirror the N64 would be impossible to get today but there are definitely better methods than using an N64 and a blender
My only gripe with this video is how adamant it is, that "it is simply not true", while we still have no real idea what actually happened. It was a leading theory, it is not anymore. All that means (to me) is that we have found more probable explanations. But the original theory wasn't exactly DISPROVEN, we just realized it was not probable (but not technically impossible). Just another example of the media, and "the people", not really following expert discourse on the matter, and instead tightly holding onto, and parroting/reposting what gets the traffic in. Other than that, cool vidya. I get the youtube need for a "click-catchy title" too, so i'll excuse it, lolol.
damn i thought this video would be an actual discovery or something but it's just saying "actually it was probably something else and not a cosmic ray". how wildly informative
if you thought that then that's your fault, there was nothing in the title or thumbnail claiming that we found the cause
'the biggest myth' -> a photo of the up-warp
myth, untruth, false, fake news, it was NOT a cosmic ray, it was a… maybe these other things but we don't know
i personally expected to see the runner accused of cheating or the holy grail- replication
i've never heard the myth of 'we've proven it was a cosmic bitflip'
all i heard until now was 'one of our remaining theories is a cosmic bitflip'
tbt if i had to bet my life on it, i'd think it was local EM noise or power grid noise, but this is such clickbait
What's funnier is the most likely outcome WAS a bit flip. He claims that because there's no 'proof' it was a bit flip, therefore it can't be a bit flip which isn't an actual argument at all.
RAM back then did not have protections against random bit flips due to radiation or even just overheating. The TAS clip clearly shows that flipping the bit from C5 to C4 gives the same height as Teabag got, so it is 99% likely the bit was flipped. HOW it was flipped is obviously unknowable. But a bit WAS flipped.
@@m0rty161if I make a video saying the moon landing is fake, but then don't explain any of the footage or evidence, would you make that same comment lmfao.
@@alextheonewarrior the only "evidence" there is of the cosmic particle is "but it COULD be how it happened" which is the same mentality as "i COULD win the lottery today"
So it was definitely, absolutely, without a shadow of a doubt, 100% a cosmic-ray-induced bit flip. Got it.
yeah honestly him being so aggressively against even the slightest possibility that a journalist might be right once that he denies the possibility altogether without even disproving it makes me just want to believe it more.
@@loafodisease614I know the feeling, but I think it’s very unlikely to be a cosmic Ray. LunaticJ really doesn’t do a good job of arguing his point, but the odds of one cosmic Ray striking an N64 and causing this particular glitch is already a very very low chance, and how that we’ve seen multiple TTC bit-flips, it seems incredibly unlikely to blame them all on cosmic rays. And if we find there’s an alternative cause for some of the bit-flips, it’s much more likely that it’s a similar cause for the original TTC upwarp. Technically it is possible that some of the bit-flips were caused by cosmic Rays still, but it becomes a lot more unlikely.
I think the real takeaway from this is not to trust badly research journalist articles that contradict their sources, and not to latch onto one theory that doesn’t really have any evidence. But it’s also just as bad to ignore unlikely theories in favour of explanations that can be disproven, or don’t satisfactorily answer the question, like a bad cartridge read.
@@justanotheryoutubechannelwell i was making the implication that the result of the video is harmful, since his wild assertions are the exact same misinformation he claims to be challenging in the beginning of the video. The point is his certainty in said assertions is fueled by spite for gaming journalist sensationalism, not any actual observable proof that there is no possibility.
wtf, did literally any of us watch the same video?
@@chillmadude There's an obvious joke I could make here, but... just use your imagination
the only thing this video proves is that there are many POTENTIAL causes for the up-warp, but because nobody knows exactly what happened it could still be a cosmic ray event. just because it's unlikely doesn't mean it is impossible. so saying that it's a totally impossible myth isn't very accurate. It is unlikely but that does not mean impossible.
I work in biopharma with highly sensitive equipment. We get cosmic rays happening regularly causing outliers in our data. We have to account for it with our software. On a normal PC size machine this stuff happens once or twice a day sometimes while we are running tests. Tests only capture data for a minute or so. So really we are only capturing data for about 2 hours at most in a workday. And in that time we often see them. Sometimes it'll be a few days without observing one. If just one bit needs to be hit to cause this, then it's just as likely as any other bit being hit. I'd say it's plausible.
Video conclusion: I have no fucking clue why it happened, just literally anything other than a cosmic ray.
every copy is personalized
From my understanding it could have been a cosmic ray, but there are other similar events that point to other possible explanations. So it's possible if unlikely to have been a cosmic ray.
@@lathalassa what are you on about, I didn't say it was impossible to have been a comic ray. Besides it was just a joke
@@xOrynos you came to the conclusion that it was "literally anything other than a cosmic ray"....when the video very clearly made it known that the cosmic ray theory is implausible when put under scrutiny and compared with more tangible scenarios like a faulty console.
ok so why is the ionizing particle theory a myth though
I was expecting some level of evidence to be weighed against it, but instead it's just touted as ridiculous, seemingly on principle?
I think the myth is that it was ever proven to have happened. It's still a possible explanation, but there's no hard evidence that (1) the glitch happened because of a bit flip or (2) that said bit flip was caused by an ionising particle.
However, I personally think that the bit flip, since it is the only reproducible method to quite accurately achieve the same result, is the best existing theory. However, I know a decent amount about computer science, but next to nothing about particle physics, so I have no idea how likely it is that a bit flip would be caused by a cosmic ray.
@slynt_ The funny thing is that no one really knows how likely it is. Single Event Upsets are likely underreported, and even when they happen, you're never gonna be able to get evidence that it was via cosmic rays. Additionally, nowadays, there is typically redundancy to ensure Single Event Upsets do not affect anything, so we might not know when they do happen
@@byeguyssry It would be interesting to see scientific research into the probability of Single Event Upsets and to what extent different variables affect that probability. Maybe there is research on that, but probably not since it's a pretty niche thing to be interested in and like you say, it isn't really a practical problem because of redundancies we've developed.
@@slynt_ If you do know a decent amount about computer science, you should know that modern electronics are specifically designed to error correct for cosmic rays flipping bits in things like memory. AFAIK the nintendo 64 does not have any of this error correction, if a bit is flipped than the console just rolls with it.
Of course, the vast majority of bitflips will not cause any visible effects at all so you'll be unlikely to see it happening.
Don't get me wrong, you've absolutely convinced me that hardware issues are the far more likely culprit here, but I feel like you MASSIVELY underestimate just how common cosmic rays are. They're not some out-there weird physics concept, you've had several pass by you in just the time you've read this alone. Single flip events are a good chunk of the reason error correction codes are as robust as they are.
That's also why cosmic rays couldn't be the culprit. They pass every console everywhere.
@@Pupperpatsmany bits to flip, redundancies and fabrication techniques in modern technology to prevent single event upsets from affecting normal use, eec memory
@@Pupperpats most of those rays aren't strong enough to cause a bit flip, but some of them are. That's why servers use ECC memory
Yep, if your PC has enough RAM there are actually scripts you can run to detect them based on bit flips.
According to IBM, for each gigabyte of RAM your PC has, you will experience one cosmic ray induced bit flip per week. Though, this was in the 1990s and technology has changed since then, so the numbers might be a bit different now.
@@thelegend8570 Data corruption on hard drives is super common as well. Modern file systems do lots of integrity checks but even so people who are in the habit of doing extra validity checks on their data often find that errors have crept in!
Small correction for 13:50
No, we do not in fact "know for a fact that it was not a bit flip" because several bits could have been flipped, I believe there is no reason why several adjacent bits can't get flipped together instead of just one of them. In fact, three adjacent bits of 512 + 256 + 128 add up to 896 which is almost exactly 900, so if people do choose to believe it was a bit flip, it would likely be those 3 bits.
1 bit flip is already very unlikely, 2 bit flips at the same time are pratically impossible
@@davibergamin5943 Depends on the particle and the physical structure of the cache line/RAM the the N64 uses. If a neutron comes in a the right angle I see no reason why it couldnt create a reaction that flips three bits. This is very unlikely, but that doesnt make it impossible.
@@davibergamin5943 while I do think it's because of faulty hardware, bit flips should not simply be disregarded. It's still a valid theory, even if unlikely. In fact, unlikely things happen all the time, simply because of how big the world is.
@@davibergamin59432 things
1. Couldn't a bit flip add to it, or is that not how bit flips work?
2. Since they are adjacent, could it be possible that the same event caused all the bit flips?
@@davibergamin5943 When thinking of impossibility consider first the odds of the Earth existing with us on it. Consider that either we are the impossibility or the visually empty universe is the impossibility.
Anything above a flat 0% is possible and under a flat 100% can fail to happen.
A "high power blender" what a joke this video is
Veritasium's video describes the event with much uncertainty, because it was never proven or disproven, which is good scientific communication. In his video he said: "This is *possibly* the rarest thing...", "...it *seems like* a newly discovered glitch...", and "The *best explanation* anyone can come up with is that a cosmic ray caused the glitch." That he cited The Gamer is cause for concern, but Veritasium never, as you suggested, used this incident "...as an example of a bit flip *definitely* happening..."
it is astounding how this guy (lunaticJ) has gotten so angry over people not agreeing with him, over something that has not been proven either way.
@@delcogoblin name fits i guess
@@delcogoblin1. he wasnt angry in the video. 2. not proven either way? really? thats not an argument when youre defending a very unlikely argument instead of listening to the most likely scenario. as another commenter said "did i forget and put a dirty bowl in the cupboard, or did some complete stranger break into my house for no reason and plant a dirty bowl in my cupboard? well, no proof either way!". see how stupid it sounds to say "no proof either way" here? 3. veritasium did this exactly. "the best explanation anyone can come up with." really? science wannabe youtube commenters making an insane theory over the sensible and simpler explanation is "the best explanation anyone can come up with?"
@@ouravantgarde so what's your sensible and simpler explanation, Señor D. Ryder?
@@delcogoblinI think that is a hardware malfunction.
I really, really wish you hadn't gone out and tried to say that SEU's caused by cosmic rays are an "out there" theory. It's in fact a very well studied, understood, and engineered against phenomenon. Whether or not THIS specific event was one of these is another matter, as they are much more rare down at the earth's surface, but you didn't need to go out and say something that you either didn't understand, or that you intended to mislead your audience with.
For real though, so many people in the comments think cosmic rays are something magical. do they ever wonder where the classic static pattern from, say, misaligned antennas on old TV's come from? It doesn't just exist or is programmed into the TV's, it's the antenna picking up random noise from the environment and a not insignificant amount of that noise comes from... cosmic rays.
@@thelelanatorlol3978 indeed. even a measurable amount of that static is the Cosmic Microwave Background radiation.
yeah except that... bit flips caused by cosmic rays at ground level are extremely rare. The less atmosphere there is, the more likely that a photon will interact with a bit. it has happened for planes and for space hardware, but it's still a pretty "out there theory" that this is specifically the reason it happened and not just a hardware malfunction. yes the phenomena is well understood, and that's exactly why it's still valid to question the likelihood of this being the cause of the glitch.
@@hypstersaurus6828 My issue is not with the argument that it caused this specific glitch. Even if it's caused by a SEU, it's more likely to be things like LOCAL radiation sources or hardware defects etc.
My issue is that the entire tone/idea behind this video seems to be trying to debunk the idea that they happen AT ALL which is a ludicrously ignorant position to take.
And judging by the comments here, exactly what made me angry has happened: the careless, lazy, ignorant, and/or stupid way this video was written has misinformed many people about the subject, and they have walked away thinking that SEU's caused by cosmic rays are akin to big foot.
The video did that because it's easier to make the argument that way, which is lazy and annoying.
@@jordanledoux197Yeah, it's extremely ironic for a video that claims from the onset to be an attempt at fighting misinformation.
To be clear, the glitch has still not been precisely replicated. And while you are correct that we do not and cannot know for certain that it was a bit flip that caused it, you also cannot dismiss the possibility that it was. There are definitely several other more likely explanations, but a bit flip remains possible, even if unlikely. The only myth is that anyone could have ever been certain about the cause of the glitch.
Not frame and pixel perfect replication, but anyone saying the bit flip recreation wasn't proof enough that somehow that's what happened, cosmic ray or not, is ridiculous. It's pretty obvious by the two videos side by side, unless there's another way to recreate very, very similar upwarps in that spot, it's a bit flip. Don't tell me that video side by side with the original isn't pretty conclusively showing damn near the same exact thing. Just because they weren't in the exact same spot for the warp doesn't make it any less incredibly, wildly close. Like so close it's shocking me that anyone thinks it's not a bit flip. The question isn't "was it a bit flip?" The question is "how did it happen? cosmic ray or not"
this whole vid seems like some wierd cope ngl, bit flipping is a very real thing@@Sahxocnsba
Hey bit flip is a good candidate cause, it's just... not necessarily of cosmic origin.
I still have yet to hear anyone say what even makes the Cosmic Ray hypothesis unlikely at all.
@@chrisgaming9567 What makes it unlikely is how improbable it is. It's not impossible, but there are other more likely things that could cause a bit flip than a cosmic ray. That all said, it has to be a bit flip. There have been no other remotely probable theories as to how he could have warped exactly that distance. There was never going to be a way to imitate the gameplay 100% perfectly side by side, even in a TAS, but it was more than close enough to show that the non cosmic ray bit flip theory almost perfectly matches what we saw in reality. Something I wonder about with the weird 900 unit down warp was if they tried to recreate it in TAS as well? If they did, maybe warping Mario down 1024 units makes him trigger some sort of in game anti clipping mechanic (the game checks for walls in marios movement path sort of) so he only appears to move 900ish units. It might be unlikely, but there is also potential that 2 bit flips combined made that happen. +1024 and -128 would be around 900 units. I don't think any bit flip we've seen in Mario 64 speedrunning has been caused by a cosmic ray, but I think bit flips do occur.
I feel like I just got dragged by the ankles into a dark wet cave full of confirmation bias.
the fact that LunaticJ really made up the "proofs" like "cosmic rays are rare" and "look how it does match with a TAS ? i have decided that it actually doesn't mach lmfao"
this video is a joke
HOW TO GOOD DOCUMENTARY IN 2 EASY STEPS!!!1!
1: Try literally everything in your power to make every point sound as biased as possible.
2: Fail to understand the definition of the word 'theory'
Hope this helped
honestly hate this video. biggest yap session of all time
My dude, you're doing the inverse. Going on and on about how it's DEFINITELY NOT, CAN'T BE, COULDN'T BE a bit flip.
Is a hardware issue plausible and also more probable than an SEU? Sure.
Was a hardware issue proven to cause the upwarp? No.
Is a bit flip still possible, even if it is highly unlikely? Yes.
The only correct answer to what caused that upwarp is "we don't know."
Exactly! The whole time he was going off about cartridge tilts, I couldn’t help but think that the bit flip theory was still possible, if unbelievably unlikely. I only started to doubt that belief when that other runner had consistent bit flips from a hardware issue, I would have lead with that.
@@tinthatisfullofbeans Even then he already states that they found no noticeable issues with DOTA_Teabag's N64 nor were they capable of recreating an upwarp.
Basically he says: The closest recreation of the original upwarp was done by someone flipping a bit, despite DOTA_Teabag's mention of sometimes tilting his cartridge to play there's no known instance of cartridge tilting flipping bits/causing upwarps, DOTA_Teabag's N64 also had no noticeable hardware issues / was unable to recreate the upwarp again, despite all this I'll just bring up an instance where someone else's N64 did likely have hardware issues and was believed to be glitching out levels through flipped bits.
Therefore, despite having zero similarities to the DOTA_Teabag upwarp, this situation is proof that a cosmic ray absolutely did not flip a bit.
Throughout the whole video he shows information that CAN suggest flipped-bits, if nothing else, but claims that they're not definitive so he looks at other situations/ideas where he acknowledges there's essentially full confirmation that these ideas can't/haven't caused upwarps or flipped bits yet goes on to say that they're magically a more plausible answer than flipped-bits (and especially cosmic rays causing it).
like???
@@CobRaM4tRIX "there's no known instance of cartridge tilting flipping bits/causing upwarps," it doesnt seem too unlikely based on just logically like. if a pin on the cart was disconnected from the console it could defintely mean the console would then encounter undefined behaviour when trying to read the instruction, and if the instruction was something to update marios position. then maybe something like that could happen.
@@LiEnby I think a faulty connection would have way more than just 1 bit flipped
@@LiEnby I'm just quoting what he was saying in the video there. I don't know enough about that particular thing therefore I can't claim with certainty that it's too unlikely. Just from what he claims in the video he's explained that there is no instance of that kind of glitching caused by cartridge tilting (Things such as sounds, visual assets, etc, being corrupted or improperly executed due to a consistent stream of poor communication between the console and the cartridge makes sense. The idea that this constant stream of potential corruption/improper execution somehow doesn't affect the game at all in any other way yet somehow at one point randomly flipping a bit in the memory and never having problems again is not only extremely unlikely but also never has happened before.)
From his claims/wordings, and from what I interpret it as with my limited understanding, is that cartridge tilting is theoretically just as unlikely to flip bits / cause upwarps and has never had any tangible evidence of it being able to do so. But he goes on to treat the general idea of flipped-bits (especially from cosmic rays) as such a laughable concept despite it having just as little "evidence" (though technically it does have more) as the failing hardware concept...
Especially mixing that with the fact that the original console and cartridge that this glitch happened on never recreated a bug even remotely close to the upwarp since (and had no noteworthy hardware issues).
Again I'm not an expert and have a limited understanding on this. But with that limited understanding, and through his own words alone, a tilted cartridge theoretically has no tangible proof of flipping bits especially in a manner like that. Almost implied in the video like it's theoretically impossible due to how the cartridge and console interact anyway.
But it seems like he'd rather ignore that (and the other things) and suggest that it's a far more plausible option because...? I guess it's more satisfying to think it was an unexplainable general hardware error instead of cosmic rays? (which btw are not that uncommon nor anywhere near as "rare" as this video paints them to be).
-
EDIT:
Just to reiterate.
It seems that the tilting of a cartridge messes with multiple aspects of the game even if they're not immediately apparent. The major point being that there's often a very blatant issue with the game right away alongside any other small glitches caused by a tilted cartridge (broken skybox, broken music/sounds, broken animations, or even outright crashing, and so on).
That would mean that somehow, despite the tilted cartridge, the game didn't show a single shred of bugginess for nearly an hour until it randomly decides to change Mario's height in a single instance then immediately operate as normal right after and never bugging out again. Even in that split instance not having a second of corrupted audio/visuals or anything. (which isn't required but it's just cherry on top for this extreme set of conveniences). And it's not like it accidentally sped Mario up immensely to throw him above the higher platform or anything like that (which wouldn't even make sense considering he was moving downwards in the first place). But rather it completely overwrote whatever his Y position was in that instance while preserving that downward velocity.
Where he's suggesting that a tilted cartridge is more likely to have caused an overwriting of memory in that way (in spite of known behaviors and lack of any instance of this happening before) rather than the idea that a bit was flipped (with which its only behavior is to do exactly that kind of thing and only that kind of thing) even after showing instances of bit flipping affecting the game in that exact type of way due to hardware issues. He acknowledges that DOTA's game didn't have hardware issues worth noting but he decides the far more plausible scenario is that a tilted cartridge somehow perfectly affected the game in that exact type of way in that specific second despite establishing the idea that it's not happened before and conceptually shouldn't happen. Rather than believing that something outside of the hardware then affected the bit to flip it (a not uncommon and well-understood concept in computers).
I'm not 100% team cosmic rays but like... it's not THAT implausible nor does it make sense to suggest that what he said is, not only more plausible, but by such a significant margin that the idea of cosmic rays is laughable.
Ok, SRL speedboomer here, the cosmic ray theory was around way before the 2020 clickbait article, i remember hearing about it when it happened in 2013. Furthermore, it was a common rumor back in the 80s and 90s back when computer hardware wasn't nearly as robust as it is today. The rumor goes (yes a rumor) that people would purchase RAM upgrades for their PCs, and when this RAM turned out to be faulty, people would angrily call the manufacturing companies of said RAM looking for answers. Instead of admitting that they sold faulty RAM to the customer, they would blame crazy things like cosmic rays for flipping bits. So yes, this Cosmic Ray Meme has been in computing for at least 40 years lol.
bc of solar flares during the time period which absolutely fried a bunch of computers and lead to an article about something people couldn't verify on their own
Wow
Physicist here. Without going into exactly what my work is, cosmic ray incidents are extremely common. It’s not a constant thing, but we expect a certain number of hits on our devices over a certain period of time and we have to account for it 😅
Reasons why the cosmic ray flip is a myth:
- there's a different upwarp glitch
- someone had many bit flips happening
- some other person had a slightly similar thing happen that wasn't a bit flip
- uhhh some people had faulty consoles that loaded textures wrong
- uhhhhh someone slapped their console and the wrong level loaded
also I ran my mixer next to my console
>person's game has a history of weird bugs
>weird bug happens in person's game
>hurr must be cosmic rays
@@electroflame6188 You mean
> person's game wouldn't start sometimes
> someone tested the game and there were no bugs
> but some other people had bugs
> clearly, it proves that was a bug
well yes, proof by contradiction
@@spicytaker Not sure u know what that is
@@spicytaker It's not proof by contradiction, none of the cited examples for causes of glitches in sm64 apply to this case.
I was the original author for the article on TheGamer. I can proudly say I no longer write for that website, nor do I do freelance gaming journalism anymore (and I wouldn't recommend you do it either). I want to clear up a few things about the ways that sites like TheGamer function, because I've grown a little tired about the amount of shit I've been given for this article in particular and others like it. I am very aware that the cosmic ray thing is just a theory, and I was very aware of it when I wrote it too.
When I wrote for them (I can't confirm it still works like this), we would submit pitches in Slack, and those that were approved, we could write basically however we wanted to. With features/originals like this one, they weren't likely to pull in lots of page views without being turned into clickbait in some capacity (since it isn't "news"). What I want to make very clear is that turning articles into clickbaity feed-fillers is the job of the editors, not the authors. After I submit it, I no longer have any control over it. Editors re-work the titles and content and push them out. I could point to many articles I wrote that were published as much weaker articles than what I submitted, because the editors prioritize things being concise, rather than accurate, and things being SEO-friendly, rather than readable.
When I submitted the article, it fell flat and barely got any page views for the first few months, which is rather annoying considering that authors only got paid for the page views from the first month, so I made basically nothing from the article, minus the flat rate for it ($10 USD I think it was for that word count?). Then of course, it blows up months later and becomes one of the best performing articles on site for a bit, and I make nothing from it, other than needing to hear about it from time to time in videos like this. TheGamer is actually the ONLY site I ever wrote for that paid extra for page views at all, so it always pisses me off when people give shit to freelance writers for just trying to get clicks. Overwhelmingly, freelance writers are just getting a flat rate so page views mean nothing to them. It wouldn't make sense for them to deliberately write clickbait.
It's precisely the reason I had my name switched to a pseudonym after I stopped writing for them, and my go-to example of why no one should do freelance gaming journalism. It is a thankless job where editors twist the articles you put passion into, only for you to get shit for it when it's published under your name.
That's actually horrible of those editors. The fact that you never got compensation after the article blew up, regardless of quality, is really unjust. I want to make it clear I have nothing against you and I hope you've been able to move on from that thankless job. I deeply apologize if my tone was unfairly harsh, and I should have been more empathetic to your struggle of being a freelance journalist. the last thing i want to for more people to harass you over a trivial article.
I made this video primarily because I saw people in the SM64 community get annoyed by people treating the TTC Upwarp as if it "definitely" happened, and wanted to make a comprehensive video on why they didn't believe it, and how the cosmic ray meme got out of hand. If you think I crossed any lines that may have upset you or caused you any trouble, I deeply apologize.
@@LunaticJ No hard feelings. My comment was more a response to other comments complaining about clickbait articles than it was to the video, and it also seemed like a good time and place to finally vent about my role in the cosmic ray saga. I just wish I never wrote that article to begin with.
@@throwawayblehblehblehYou do understand why criticizing the article is fair and reasonable, right? I'm perfectly happy to blame the editors instead of the authors.
Yeah, the cosmic ray theory being paraded as definitely true by media is egregious, but I don't understand the feeling of animosity towards the theory. Sure, it would be better if it wasn't a freak instance, as that may mean it could actually be replicated, but absent an actual technical explanation, it's just as valid a theory as any other considering what information we have.
the problem is that it was pretty universally reported as fact because nobody bothered to question it before reporting on it
@@FirstRecords204 Agreed. That being said I don't think you can accuse Veritasium of reporting it as fact. He merely pointed out that it was a possible example of a notable instance of a bit flip.
Thank you. This video feels like needless semantics. "Something happened that cannot be replicated through an emulator or normal console consistently, which caused a single bit of information to flip, teleporting Mario's positioning upwards." That's the factual story, and it's already super interesting.
People went with the cosmic ray theory because it's the more interesting culprit between the faulty cartridge or electromagnetic waves. None of them can be disproven or proven, at all. Especially considering that it was the only bitflip to happen that stream. Yes, it's the least likely of them, and it got reported as factually being it instead of possibly being the culprit, but it's an interesting tale nonetheless. The fact that the footage doesn't match with the original 1:1 by upwarping through hacks can also be easily explained by consoles not always playing at 100% speed.
It just comes off as the video not feeling like "it wasn't for sure a cosmic ray, but rather, a cosmic ray is still possible, but rather unlikely!" Wasn't substantive enough material for a video, and it needed a "it's FACTUALLY NOT TRUE that it was a cosmic ray!" Moment. And that's just the same bad journalism that the video criticized being repeated again. Sometimes stuff in life will forever be uncertain, and that's okay.
Yeah, this dude is completely dismissing the well-documented science on SEE. The answer should be: "no we don't know what happened and one of the theories is bit flip caused by cosmic rays" and not "cosmic rays totally didn't happen guys because that would be EXTRA silly."
@@ultimateman55
Using a shoddy article as a source is just bad journalism. It doesn't matter if he didn't explicitly sag it was absolute fact, he's presenting it as an example among a bunch of actual factual evidence, so it's going to be seen as such because he's a science communicator and is expected to be reporting things with a certain amount of rigor. If a video that is overall very scientific suddenly has something in it which doesn't meet the same standards, then that's a mistake.
Him showing footage from a Veritasium video that directly contradicts what he is saying is hilarious
I never expected to see a person so angry at cosmic rays in my life, the internet is truly boundless
who's angry?
@@TweedleDeem LunaticJ, watch the video
He's not mad about cosmic rays. He is just correcting misinformation.
@@Gigamex2 it's just a joke don't take it seriously
@@osparav since you read my comment telling you to watch the video, read that same comment to find out who was angry
As pointed out by the other comments, bit flips caused by cosmic rays or interference happen all the time in computers, which is why we have ECC memory and other redundancy schemes. So the idea pushed by this video of this being a super unlikely hypothesis that nobody took seriously is false. Plus, cartridge tilting can't cause single bit flips in the console's RAM or cache--he even shows someone saying this in the video--so that hypothesis is out. Finally, a console which suffers multiple bit flips in a short time probably has faulty memory or cache, but this video even explains that thorough testing was done on the actual N64 which had the upwarp, and no further issues were found. So the faulty hardware hardware hypothesis is also out. Yes, it would be accurate to revise the claim from just cosmic rays to "cosmic rays or other outside interference", but that's the best hypothesis we have so far.
ECC also helps with errors from faults or nearby interference ... similar to the explanations in the the video, there are many potential causes for single-bit errors
it's not even proven to be a bit flip though
Exactly what I was thinking while watching this. Someone should really look into the redundancy schemes of the N64's cache to find out if a single bit flip could have made it through without other noticeable effects. Since the game's code is so deeply understood these days, we know it's not an anomaly or math error in the programming, and we know cartridge tilting cannot have this exact effect. A formal error testing of the offending N64's memory after heat soaking it should be conducted, instead of just trying to replicate the glitch on the console, as I'm sure there are other bits that could be flipped without noticeable effects like this. I think a power surge is a more likely explanation, or other types of ray interference, which the N64 is legally obligated to accept (and deal with) due to FCC laws.
@@TheKingOfApples100 reading comprehension check
@@TheKingOfApples100 well the OP's comment was only directed at the original case. The other n64 is confirmed to have memory issues, which couldn't be replicated on the upwarp console, so I don't think he was implying that one was being bombarded with cosmic rays enough to make the bit constantly flip as it does in the video
So a faulty cartridge, which you pointed out *isn't known to have this effect,* is more likely to cause this effect than a cosmic ray, which only has this effect?
My gripe exactly. I wanted to post a carefree comment about the cosmic ray hypothesis being more of a dated meme than a hard answer to the problem-at least to those of us who were following the story at the time-but this guy is taking it seriously and reaching so far in the opposite direction that it's not even funny. I'm walking away with *more* faith in the cosmic ray than when I was 10 years younger.
Most cosmic rays have no effect.
@@Chapien Doesn't matter. It is the only plausible theory.
Mind you, it could be interference from some other radiation source too.
@@semisixx4967the cosmic ray is still a meme though. There are plenty of things that can cause bit flips. This is why they sell error correcting RAM after all. Cosmic rays are the least likely of those things, they just sounds the coolest.
Pointing at other examples where bit flips occured is like saying:
"Because 10 / 10 murders we sampled happened with guns, all murders happened with guns"
Just because this theory doesn't apply in other spaces, doesn't mean it doesn't apply to where it was originally suggested.
Bro that glitch of the Japanese player going to big boo haunt, if that happened to me as a kid I would have been absolutely terrified
Right? That's some B3313 crap
I would never hit my console again
Dies in CCM, gets mad, slaps his N64, then goes to re-enter CCM but ends up in BBH instead. He gets what he asked for; a different level.
"Fuck this level man *hits console*
Here ya go bud, a better level
Every-
A bit flip clearly happened there, the footage proves that, it's just what specifically caused the bit to flip is unknown
I think we should continue believing it was a cosmic ray just to spite this guy who seems to think he’s smarter than everyone else when he gets basic info about this wrong
he really does have the "erm akshually" vibe, trying so hard it kinda hurts
"We don't know" not only means "it's not proven it DID happen", it also means "it's not proven it DID NOT happen". It's simply not yet proven on either side.
Yeah, but it's worth getting Occam's Razor involved there too.
burden of proof is on the person who claimed it happened, given the incredibly low probability of it happening
Are u slow
@gmdnikoyou should do it first pal
@gmdnikoChill, I think he might be artistic
Telemetry data from Microsoft has shown that a noticeable portion of app crashes and blue screens are due to bit flips too. But even those can be explained through interference between memory cells and "in flight" between the DRAM modules and the memory controller, especially with any form of overclocking. Also with the gigabytes of memory that a PC has, any bit flip is much more likely to hit nonessential data.
The N64 will be using higher voltages and larger transistors than any windows machine, which makes bit flips much much harder. You need much more energy to flip a 5V bit than a 3.3V or 1.8V because the particle has to create that voltage on the memory cell
@@lomiificationCosmic rays are quite high energy objects. I don't have numbers off the top of my head but if a particle ionised by one has enough energy to flip a 3.3V it seems likely for them to have the energy to flip a 5V bit.
@@lomiification It's the *charge,* not the voltage. Okay, let's dig a bit into that.
A 1GBit Samsung DRAM from 1995 with a supply voltage of 3.3V would internally store 1.8V in a DRAM cell. Each cell has a capacitance of around 25-30 fF (femto-Farad). That's a difference of maybe 300,000 electrons between the two sides of the capacitor (assuming I got my math right).
The N64 had a 32MBit Rambus DRAM (36MBit actually, I'll explain further down). I don't have the technical details there, but a typical call voltage would be around 2.5V (supply voltage was 3.3V). Note that the Rambus uses differential 0.2V signals. The structures of such a 32MBit DRAM will be much larger than those of a 1GBit DRAM, so they can accommodate not only a larger voltage, but will also have a higher capacitance per cell - you have to convince much, much more electrons "to change sides" in a smaller DRAM.
Nowadays, almost 30 years later, structures have shrunk even more and our modern DRAM is much more susceptible to cosmic radiation due to the smaller structures (that's one reason why quite a few space probes use outdated technology). This effect is however mitigated by better manufacturing processes and geometry.
Technically, the RAM of the N64 would have been ECC RAM, but it wasn't used like that. The CPU could access only 8 of the 9 bits per byte (for obvious reasons), and the 9th bit was used by the GPU only. No ECC or error-checking there. Hence 36MBit inside the RAM, which looked like 32MBit to the CPU.
A few decades ago, Microsoft claimed that about 30% of all Windows crashes were caused by bad RAM. Considering the overall stability of the Windows versions back then, this seems to be a very bold claim. Well, maybe Microsoft had a farm with thousands of PCs with bad RAM and had them crash continuously, to get to the 30%?
Whatever. A Google study from 2009 (IIRC) showed that 8% of their RAM modules enjoyed one or more bit errors per year. I'm not sure what kind of memory modules Google used between 2007 and 2009, but I know that, in the past, Google had used very special memory modules which offered more memory per slot - and violated the specs.
TL;DR version: Yes, i suspect that you're right when you wrote that it's probably harder for cosmic radiation to flip bits in an N64 RDRAM than in the more capable PC DRAM modules.
Bit flips are a well documented phenomenon. Sure, they *usually* happen to server farms with terabytes of ram, just because there's more stuff to go wrong, but it's bound to happen to an N64 eventually. Whether it was specifically caused by a cosmic ray is unknowable and frankly irrelevant... a single bit flip still seems the most probable cause.
Counterpoint: If the actual cause of the upwarp hasn’t been definitively proven, then you can’t say with certainty that the cosmic ray hypothesis is untrue.
I choose to believe…
It’s maybe a counter argument but as you said, rooted in nothing but belief. There’s nothing scientific about saying “it could be this, so you should entertain it”. You go with the explanation that’s most likely, and remain open minded enough to change your mind if you see that something else is more likely.
@@Cosmalano Nope, wrong! Every other hypothesis has resisted demonstration despite having a much lower test-ability treshold, it's literally anti-scientific to refuse entertaining a perfectly plausible explanation on grounds of odds when none of the other more mundane ones can be proven despite having the exact offending hardware that experienced the glitch in the first place.
@@TheCamps10 that is not “anti-scientific,” let alone literally so. Science doesn’t depend on “proving” an idea right, it depends on proving them wrong. Cargo cult science seeks to prove things right. In science you always seek to disprove a hypothesis.
The bit flip hypothesis did not produce the same effect when tested, which ruled it out. If you want to entirely rule out the faulty hardware explanation, then you need to demonstrate that it doesn’t work. If you’re not after ruling things out, then we’re back to my original point, whatever is most likely is the one you should go with, until it’s proven wrong. That’s the only way science works. Nothing in science has ever nor can ever be “proven right”.
Even if the bit flip produced the same effect, this wouldn’t mean that a cosmic ray caused it, nor that this is indeed what happened. Just like how saying the Earth is flat works as a hypothesis when you see a couple of effects, but doesn’t hold up when you consider even more, where as the globe Earth model captures everything you could throw at it.
Well the best counterargument could be that the radiation/ waves/ rays came from earth. Storms, Uranium and other waves can still cause problem with tech. Or, ya know the giant light ball we call sun has regulary big outburst of radiation and solar flare, of wich 1 was in 2014!
W gold truth
Regardless of how you believe it happened, you really don't have to treat everyone who believed a widely spread story as if they're stupid for believing it when it was the only explanation given to them for years. Especially not while the actual reason why it happened is still unknown and the widely spread story you seem to dislike so much is still a possibility, however remote.
Exactly! How dare this UA-camr treat people who actively believe that particles from space flip a bit (which was never even proven to have caused this glitch in the first place) and actively defend this opinion when no proof and contradictions to that statement are shown online!
@@kitterbug that's some good sarcasm there. thing is Lunatic never offered an alternative possibility, he just said duh this is unlikely, so it therefore must be impossible! yeah, I am not sure I believe the bitflip idea either, but just because you can't see it happening doesn't make it not true, thats an argument from incredulity
@@kitterbug The thing is, the video isn't even saying that the space ray CAN'T cause a bit flip - heck our modern civilization is one strong space fart away from crumbling as all of the electronic devices could potentially be bricked by such an event - the video is simply saying that it's unlikely that a cosmic ray would cause THIS particular bit to flip at this specific time. Which is... a bit like arguing that it's very unlikely that a person got hit by lightning because why would the lightning strike in the exact position and time that that person happened to be in
@@kitterbug are you literally like 10?
It could be a bit flip, just not from a cosmic ray, bit flips do happen more often then people think, just in small areas that make 0 difference usually.
Bit flips are more common from old storage/memory, after storage (especially ram) gets old it has trouble retaining bits of information often causing bit flips.
For storage in hard drives and SSDs bit flips take longer but will occur if something doesn't get used because it's just sitting there and the charged particles that create the bits can move (either to near by particles swapping bits or to an outside particle loosing the bit all together, the editor roughly mentioned how outside interference can also cause bit flips with outside particles charging the inside storage).
This will often make old images have worse quality as colour(color if you're American) bits will get changed making the image more noisy, however, if you move the image around a lot the image will usually get compressed and in some cases will have generation loss as each copy loses information.
These bit flips I'm talking about where the bits more corrode then flip (but can still happen in real time) are usually just called corruption (like when you get a corrupt file)
I agree with a lot of the other people in the comments saying that this video is a nothingburger and just as poor of video game journalism, because it claims that it definitively _wasn't_ cosmic rays without providing any evidence. Is it extremely unlikely? Yeah. But unlikely doesn't mean impossible.
Saying that it definitely wasn't cosmic rays without providing evidence or clearly stating "it's possible but other things are much more likely" makes this the same kind of sensationalized garbage journalism as any article claiming it absolutely _was_ a cosmic ray bit flip.
He said "it's a leap in logic to assume that over an answer found in-game or a hardware malfunction", that is, to assume something unlikely happened when there are more likely explanations, at around 6:15 minute mark. Did I miss the part where he says something like "it definetly didn't happen" or "it's impossible"?
@@azzor4134The thumbnail and the title both strongly suggest that it definitively wasn't a cosmic ray bit flip (given the connotation usually attributed to the word "myth"), plus some extremely confident language at 18:20, saying "it just isn't true." And at about 18:45, he calls it a myth yet again.
Additionally, his tone and wording throughout the video is straight-up hostile towards the idea that the bit flip theory in general is even slightly plausible, despite the fact that it still is as plausible as it ever has been.
EDIT: Something I thought I remembered, and found verification elsewhere in the comments - the thumbnail has been changed. It originally straight-up said "IT WASN'T COSMIC RAYS." No nuance, no room for misinterpretation. And now that people who actually know what they're talking about are saying it's unlikely but certainly not nearly as impossible as he claims, he's backpedaling HARD.
So he said it. Though, as far as I remember, he said it was unlikely throughout the video until that point rather than 'false', and his pinned comment also states "I'm just saying it *probably* didn't happen", so I'm willing to let it pass as a small mistake.
I understood 'myth' to mean not necessarily a false claim that is commonly held to be true, but an unproven claim that is commonly held to be true. *That may not be the exact dictionary definition of the word, but that's apparently what he meant by it*. So if the cosmic ray hypothesis is held to be true by many people, despite being still an unproven hypothesis, then it is a myth. He attempts to dispel it by showing alternative hypotheses which seem more likely (not a definitive answer, as he concludes by saying that it's still a mystery) and how well he did that is debatable, but he isn't wrong to call it a myth, by that definition.
In my opinion he did a good job at showing that we don't known the answer, and how one answer came to be commonly accepted as the correct one, but a bad job at showing how alternative answers to the cosmic ray bit flip hypothesis are more likely. Rather, it still seems that any explanation is as good as any other.
@@azzor4134 To be fair, he edited the pinned comment _after_ me and a bunch of other people made comments calling him out.
EDIT: He also edited the thumbnail to what it is from "IT WASN'T COSMIC RAYS." I think that says what you need to know.
I admit that this may be a misunderstanding, but at the same time it rubs me the wrong way that he failed to do his due diligence while trying to come across as someone with journalistic integrity. He clearly came into this with a strong opinion against the bit flip theory based on his word choice and his tone, and yet provides a bunch of evidence that, if anything, _supports_ the bit flip theory - examples of hardware malfunctions that are often far more drastic, the N64 and cart used have been tested by people who know the system inside and out and nothing was found, _flipping one specific bit is identical to what happened,_ etc.
I would have no problem with this video if it were titled better and used more agnostic tone/word choice throughout. But I think it's pretty clear that the implication is "the bit flip theory is incredibly dumb," when it's just as if not more plausible than it ever has been.
I work in the same office at NASA JPL as the radiation effects expert in Veritasium's video. If you have a GB of RAM, then you might be getting a bit flip once per day on earth on average caused by cosmic rays. Google estimated the DIMMs in their study were getting 10 per day. So there are billions of bit flips every day on earth induced by cosmic rays. A single event can cause one bit to flip or many. Radiation induced upset in memories is very well understood; there are particle accelerators all around the world used for testing electronics that run 24/7/365. If N64 had 4 MB of RAM, then a bit flip every 200-300 days would be expected (1/250th compared to 1 GB). So games who only play a couple hours per week may never see bit flips in their lives. But gamers playing 8 hours per day... they may not all see them on their 4 MB systems, but many (10-30%?) definitely saw them on N64 if they played a few hours per week for a few years. As someone whose job is to understand electronic interconnect reliability, solder joints, intrinsic memory cell reliability, and radiation effects; by far the most likely explanation is radiation induced upset rather than any of the other possibilities mentioned. Anything to do with interconnects or solder joints or anything at a package level (cart slap) would cause catastrophic malfunction; not a single bit flip. Intermittency on one or more signals between the cartridge and system does not cause a single bit flip. That leads to major memory corruption. The only other credible root cause in my opinion is a weak bit; one that is marginal and beginning to fail. This is common in memories. All electronics have a limited lifetime and will eventually fail; process marginalities in memories create a distribution of bit cell lifetimes with some failing before others. However, in that case, you would likely see further degradation over subsequent months and years of use. Not sure if that was the case for that user. If the glitch was isolated to a single run, it was most likely cosmic ray (not EM interference; that also does not flip specific bits...) This is not conjecture; it is 25 years of daily first-hand experience dealing with electronic component reliability issues including the ones mentioned in the video. (the explanation at 14:00 about powers of 2 is total nonsense; what about a bit flip that changes mario's position 512 to a position where there is no floor and he falls the rest of the way to 900? i'm not saying that's precisely the case, but it's not remotely as simple as described in that section of the video.)
The best part about the TheGamer article is that the upwarp didn't even save him time, it did the exact opposite.
the complete opposite
i remember when pannen put out the bounty for this, but i NEVER heard anything about cosmic rays lmao. this is wild
I only ever heard about the cosmic ray
Complete opposite for me back when this was still new.
I first found a video talking about the cosmic rays (or the possibility of it being the case) before I heard about the bounty
I saw the video in 2016, returned in 2017 to see comments talking about the cosmic ray.
Im shaking and crying the gamer would never spread misinformatiom on the internet
so we all gave a nothing burger video a million views? okay. yey us.
It sounds like the cosmic ray theory is the most "complete" theory to explain the observed phenomenon. It's recognized that it is unlikely, but I think this has to be weighed against the evidence that we know cosmic rays do occasionally cause observable bit flips in computer systems.
The alternative theory here is "something something to do with faulty hardware", but a complete theory isn't presented.
I would say neither a single event upset nor a spurious hardware failure can be ruled out at this point. It seems unlikely given the limited evidence we'll ever be able to come to a fully definitive solution since the event happened in the past and we don't have access to the hardware state at the time. In that case the best we can hope for are some good theories, and I think both theories above are decent.
While I think the general consensus in the comments has already been established, I still would like to weigh in.
This video is wonderful case study in how Not to go about a counter argument.
To start most generally, in your attempt to use Occam's razor, you fell into the watchmaker's fallacy. That being, from within a complex system, and with hindsight, you cannot outright discredit an improbably based solely on the poor odds. Typically used when arguing theology, the idea goes that you find a watch on the beach. While in theory, there's a chance through extremely fortuitous odds that the ocean could have created a working watch, the more believable theory would be that there's a watchmaker who in fact made it. The problem comes from observation bias. If something improbable doesn't happen, you would never question the improbably of it. If something improbable does happen, and you're not there to see it, it results in the same lack of reaction. Rather than an observation of something not happening to disprove a theory, it's using an observation of something happening to disprove said thing from having happened.
Secondly, experimental data will never be "one to one" even across the exact same tests done back to back; speaking from experience as a physics major. There's a reason the field of statistics is a thing, as it's no simple process to determine if variation is within impactful limits. To my uneducated eye- likely that of many of us- the bit flip test looked just about frame perfect, but with a different canera angle. Perhaps there's some meta data that says otherwise, but if so, that should've been shown. Additionally, one test of the bit-flip, in one location does not a data pool make; a data puddle more like. With the number of variables in a game like SM64, you need to test the primary variable arrangement many times over, and then begin to adjust other variables incase there are unknown linkages; perhaps your house in under an ozone weak point, or you have more radon in your soil, or maybe TTC has some poor code which causes incorrect bit calculations. These are all things which need testing before a bit-flip can be ruled out.
Lastly, on cosmic rays themselves, as you said yourself, there is no concrete answer thus far and this is stuck in theory limbo. A misconception of scientific theory is that they are heeded because the evidence prove them; rather, a theory is heeded because there is yet decisive evidence to Disprove it. The theory of general relativity has many issues, mainly it's incompatibly with black holes. The reason we still use it is because we don't have a better explanation. The same goes for here. SSEs are shockingly common, as are cosmic rays, and while there are other causes for SSEs, it would seem thus far, none of them apply to this situation better than a cosmic ray.
All told, this was a poorly written video with a combative tone, and which not only is predicted on fallacies, but falls victim to those it attempts to call out. This is how misinformation is amplified; a perhaps mildly incorrect statement countered by an even more incorrect statement, and the cycle repeats.
If it weren't for the fact that this is involving SM64 speed running, I would be genuinely concerened that this was number 40 on trending.
If this is something the SM64 community truly want to put to rest, screaming at the world that we're all wrong and uninformed, why don't you inform us. The burden of proof is on you, and imperical data carries infinity more weight than words. Show that this particular SN64 has poor radiation shielding, or that the original owners home has an unusually high count of ionizing radiation, or that the position data doesn't align with more than eyeballed videos, or that system impacts can induce bit flips, or that some code in TTC is broken, or any number of other counter theories with better proof than cosmic rays. Until that testing is done, until there is more counter evidence than "this seems improbable," then this argument is baseless.
"Rather than an observation of something not happening to disprove a theory, it's using an observation of something happening to disprove said thing from having happened."
I get English might not be your native language but I am genuinely stumped at this. Watchmaker's fallacy is not used here, in fact it's a term completely made up to disprove the Watchmaker Argument developed by William Paley in the s.XIX. Both the argument and the rebuttal are only meant to use in one specific context, and are not applicable to other scenarios.
As a clarification, I don't accept the watchmaker argument because it uses a man-made object to build the premise of a designed object. It's circular reasoning. It has nothing to do with probability or observability.
A much better use of the argument is to argue design patterns using a natural object as a premise, like DNA strings. But then that would not be the watchmaker argument, but something else entirely.
@@FrcNeru I don't think english is the issue; his grammar and diction are more or less fine. The issue, I think, is that he is probably just trying to be hyper-specific with his wording, resulting in awkward phrasing. Speaking as an autist, I do it a lot, and that's what this feels like to me.
I'm not 100% certain what he's trying to say, but I'm pretty sure part of what he's saying is that we are more likely to notice that improbable than the probable. For example, a meme in XCOM is that the odds of hitting are significantly lower than stated (eg. a displayed 98% chance to hit is actually a 50% chance to hit), however, in XCOM, the opposite is true; the game actually gives you hidden bonuses to make the odds of hitting more likely than shown rather than less likely. The reason it's a meme, is because everyone remembers the 1 time they missed a 98% chance to hit, and no one remembers the 49 times they hit the 98%.
So, with that in mind, I think the meaning of the sentence is that you can't just say 'cosmic bit-flip is unlikely and therefore it didn't happen', because you are unknowingly discounting all of the chances it had to happen but didn't. It's not that I cheated to get the royal flush; it's that you don't remember all the games we had to play before I got it.
Though, keep in mind, I didn't write his comment, so I'm probably off the mark at least a little bit.
Perhaps the point of this video is not that comsic rays "don't happen" rather the point was "mario 64 is an old glitchy game that has many many many documented instances of glitches, specifically on this one level, that appears to be a glitchy level, and this skip is just a rare glitch from a glitchy level, rather then a unprobeable cosmic ray".... but yeah this video did not make that point at all :)
I’m not sure you disproved the null in this video. Working around spacecraft, we have modeled these events on hardware, and while it is significantly harder the closer to earth you are for such an anomaly to occur, we have documented cases every year of this happening.