It's utterly pathetic that my damned $60 RetroPie setup can emulate these games with less input lag than an official Capcom port on a $400 system. This is pure laziness and there's no excuse.
Console emulation tends to typically suck. Dont even waste your money, just emulate on PC or better yet, get the real carts, hardware and a CRT if you can afford to. Nothing beats a real authentic setup!
@@NintenPizza It's not about the console. The Steam version, where you are literally paying to emulate it on your PC, has 120ms input lag. If you decide to drive while adding that level of input lag (by, say, getting drunk) you can be prosecuted. it's not good enough.
In the past, I was never really aware of input latency. I first played Mega Man X on the Legacy Collection, and I noticed so many times where the input did not feel responsive at all. Out of curiosity, I tried Mega Man X on my SNES classic edition, and the difference was night and day. Amazing job on this video, I truly admire how thorough you were in investigating this, you guys are true scientists!
Input lag should be something game devs should take into account, because they clearly don't. Console versions should be on par with PC, heck even original 25 year old hardware, for an emulation collection. The fact Capcom can't even make 20+ year old games that responsive is appalling. People that playtested this collection were either incompetent testers or unskilled enough to tell how sloppy the games control.
Western QA would've likely got it. Noodal's sent me a Post Mortem on Black Ops 3 where they said the QA department complained about input lag. It's an interesting watch.
So... NVidia is better than AMD for lag? - I am very interested in this series, but not so interested in Megaman, since I would obviously play it on a Analogue Super NT. It is interesting that the 8bitDo is so slow, thats very disappointing. - Very keen to see how a non-emulated game like Sonic mania performs. Since the main confusion in this series is the difference between a baseline measurement for the tech (native game) VS Mega Man, which is specifically crap. IE - the difference between the platform being slow, and the programmers doing a bad job
Yea, I haven't owned an AMD card in a while and no one I know has a Free Sync display. We will definitely be doing Sonic Mania soon as a part of our tests.
Input Lag is actually hardware based, not something a game dev can just magically fix via the implementation of their game. You see, there are drivers for controllers that interface with the console and computer you are using. Those drivers along with the hardware itself is set to send input frequencies for the OS to read and do something with. In short, don’t blame game devs for the short coming of console hardware.
You're doing excellent work here. Input latency is an important factor for me when deciding which version of a game to pick up and for years I've wished I could just go on UA-cam and find a video like this to help me make my decision. This shit is just as important as Digital Foundry's framerate graphs. I know it's a huge amount of work, but I hope you guys can continue to produce videos like this.
I hope we can too. I was a little disappointed when Digital Foundry only did 2 videos with Noodall's input latency kit. We're definitely going to push to do more videos like this, with Sonic Mania likely being the next one we do.
I have never seen your videos before, but I do very much watch a lot of Digital Foundry videos, so that first 1:50 was the best introduction to your Charisma, Channel and Humor. I was absolutely dying!
Y'all are really hating on Dan for saying the word SNES? Do you people have lives? This is what you criticize in videos? I'll never understand the internet.
T Money Fun fact a lot of people in UK and Australia say it SNES. If you say NES add an S to the front then pronounce that combination is how and why it's said that way. Plus the quirks of accents. Fun times. I say all of this as an Australian.
Pronouncing each letter fully is only a North American thing from what I've seen. I have never heard anyone who is not from NA call it by each individual letter. It is always a single word, SNES. Living in the UK everyone pronounces it that way so you could argue the NA way is wrong to the rest of the world, but it is simply different dialects.
It's videos like this that remind me why I like your channel so much. So much time and dedication into something that many people don't even think about, but is the kind of thing I care about immensely.
Absolutely! Always great to have information. I really do think G-Sync is where it's at for PC gaming. Wish I could jump on the bandwagon but it's too costly for now.
Some of the numbers here are staggering. I swear the SNES Classic felt very similar to an actual SNES. It's always inspiring to see this much dedication aimed at such an important aspect of gaming (retro gaming collections as well in this case). Thank you Dan Videogames.
While this is really comprehensive and I appreciate how this validates my SNES Classic purchase, this is totally your "1.5 A Press Rolling Rocks" video.
Awesome work, GigaBoots! Having produced thousands of test samples for my own input lag testing, I know how much work it is to do this kind of thing. It's exhausting (and kind of addictive). Would have been great to see an optimized RetroArch setup in there, though. I know quite well where it would land from my own testing (around 5 ms higher than the original SNES), but still. :) Maybe for a future video!
Awesome. By the way, it was also nice to see that your figures for the SNES Classic vs original lined up with my measurements. I had the SNES Classic at ~1.3 frames higher input lag than the original, which is within the margin of error compared to your result. Also interesting to see is that all of the pre-MMX LC releases of MMX 1-3 have decent input lag, while all of the new collections venture into the "kind of distracting" area. After lots of testing (of input lag in general), I'd personally say ~5 frames of total input lag or less is fine, while input lag above those levels starts to become increasingly distracting. One thing I've personally thought about is creating a video illustrating different levels of input lag in slow-motion. For example, showing how far in front of a ledge you need to press the jump button in order to not just run into the pit. Would be nice to see what that looks like with, for example, 3, 8, 12 frames of input lag. The problem is of course worse the faster the character can move. I've never gotten around to doing a video like this though, in large part due to a major lack of time recently.
That is precisely the next video we’re doing on this topic because people simply don’t understand and need it explained and illustrated. It’s really important that we make people care, IMO. Does your 5 frame estimate include display lag?
That will be a really awesome video. Can't wait! Yep, 5 frames of total input lag from button to pixel. Some people are certainly more sensitive than I am, but I personally feel that 5 frames is fine unless you run around and deliberately look for lag. It's obviously always better with less, though, since any lag will make you play objectively worse.
Capcom still hasn't fixed the lag issues on Switch. I recently got it for Switch and noticed the lag right away. I even did my own tests like you guys did with my SNES and other systems. A friend of mine swears im crazy and the issue is just my cart, but seeing your video is the proof i need to prove him wrong. Thanks for addressing this issue. I hope they patch it soon.
Marcos Danilo no, I think their timing is correct. Dante presses a button to make X jump (not start the music), and since there's so much lag he gets pissed and destroys the jukebox.
TotallyNotThatGuyYouJustBanned na I was looking at the joke as a DMC fan first and missed it till I looked at it with a broader perspective. As a DMC fan I expected the music to play once the button was presed. That's just me though
Its very noticeable in Ps4 many times i swear to god i press jump mega man just stands there. Still beat all 4 games but still im not getting the second half for this reason well unless they fix it
Thanks for the like & sub. You'll want to follow us over on GigaBoots Tech Lab where we're going to put our more technical videos going forward. It may take a bit before they start coming out though because of current production problems. ua-cam.com/channels/J7eptoy8sqcSI_3g8ofKNw.html
Imagine if game Devs had Dan as a consultant, I'm sure the latency would be as low as physically possible. I'm really impressed about the X8 on PC, I wish I had the means to playing that, twice as much responsiveness must be insane!
That would explain everything why I suddenly jumping most of the times into my death instead of jumping correctly out of danger . My favourite was x3 of the collection but what baffles me is that many times when I wanna jump from the wall ,while pressing the dash button , it doesnt register that im going to make a long jump. Oh and I forgot the charge-shots that I always shoot out after im getting hit by a boss to get the last hit or make huge damage in a fight .This infuriates me everytime because everytime I see the frame where x is getting out of the damage animation I always release the shoot button but the charge shot is suddenly gone . WTF?! Good that the ps1 games are more responsive but still they need to fix X1-3 from the frame latancey and the screen tearing as fast as possible . Thats a no go.
I noticed right away when I played MMX on the Legacy Collection (on a PS4 Pro) just how bad the input lag is. It is playable, it just took me awhile to adjust to. Especially after playing MMX on the Super NT, which really is a great system with next to zero input latency. It's my go to for playing my SNES games in hdmi 👌.
But is the Input Lag as bad on a Original PS4? I have the original PS4 and dispite being one of those that rarely ever notices input lags in games, I have the feeling that if the PS4 is as bad as the PS4 PRO, then I should definately notice it (especially since I played X1, 4 and 8 to death and should notice!). Weird how they include Original XBONE, and the XBOX (X Box One X, Clever Microsoft), they should have also included the original PS4 as well.
This is a fantastic video. I farted around with this on Switch last night and I was pretty sure that something felt off. It's good to know I wasn't just imagining an issue that doesn't exist. I hope that they get this patched. If not, I'll likely go back to the PS2 collection. Having this on Switch is brilliant, but I'm very sensitive to input lag, and its presence here would really ruin the experience.
I was back-and-forth on whether I wanted to get it on PS4 or Switch. Eventually decided to go for Switch since the graphics don't really suffer from playing them portable. Looks like I happened to make the right choice for latency as well, though the Switch still has input lag that seems noticeable even to me (or maybe its just because I've seen videos like this saying it does...).
w0ng™ yeah man, i feel you. same people who get pissed when someone says "wally" instead of "waldo". Here in Brazil, the sega "genesis" is know as "mega drive". And thats how i say it, and yet some people make fun because "genesis is the correct name". Lol
@@ricmagus Joke's on them, "Where's Wally?" and "Sega Mega Drive" are the original names before their American localisations. So they can stick that in their pipes and smoke it.
Thankfuly, it's plays just fine on Mega Man Zero/ZX Legacy Collection. At least on PC, where I played, I do not feel any input lag like MMXLC1. It's smooth. :)
Thank you for this video. I was waiting for it before buying it on Steam. I don't think I will buy it now. Input lag is something I have little tolerance for! Awesome channel! Keep up the good work! :D
Anyone have any idea how the lag compares to the Megaman (Classic) Legacy Collection? I was able to get through all of those games on Switch, and found it pretty responsive in portable mode, even though I'm usually pretty sensitive to lag.
PogueSquadron megaman legacy collection classic, part one has fairly noticeable input lag on the switch. I own the pc and switch version and I could instantly tell the difference. However part 2 of the collection is fine and much more responsive.
Just picked this collection up on PS4 and immediately noticed the input lag, and stumbled upon this video. I’m glad you made it so I know I’m not crazy 😅 sad to have wasted $20 though
You know fellas, I thought you guys were crying over something silly. I never had an issue with input lag. Until I started really watching you guys, then I started noticing it. Good job. This shits stupid. Keep it up guys! 👍
GigaBoots Damn, All I wanted to do was show my support and post a intentionally humorous comment. Forgot you can't run from the hate and cancer in UA-cam. Sorry for starting a war Dan!
Next time a guy at Whataburger interrupts your conversation you better pull this video up and make him watch the whole thing. Great job though, this exhaustive amount of effort and detail in your videos are why so many of us love you guys.
Input lag is the reason I use ZSNES over Snes9x to play games like the Mega Man X series (inb4 muh accuracy). ZSNES with v-sync enabled is less than half of the delay that other emulators like Snes9x produce. Input lag should be the number one priority when developers make/re-release games. It's probably the most infuriating problem you can have with games, especially platformers.
It's a bit bewildering how much of a mixed bag this collection is. On one hand, sure it's a great value, far better than Nintendo's VC where each of these games would cost $8+ cool high res scans, X-Challenge is ...alright and all that But the games have slowdown ...where the X-Collection on PS2 didn't, the save system is also worse than that collection. How are we having trouble with emulation on hardware THIS POWERFUL, I can emulate these games better on my old Pentium 4 laptop than an officially done product. That's not good.
Two odd questions #1. Did you test the PS4 and XB1X with the cord physically connected? I heard that solves some latency. #2. At the time, the games didn't feel much different to me, but I just tried to play through the legacy collection again, just for fun, and everything felt.. off. MMX I've never had an issue with, but all of a sudden, I was dying to Sigma 1's tower climbs. I could just be getting old (44)
So the test with DS4 controllers + vsync on pc seems to point at the controller itself causing the input lag. I actually wonder if that holds true across other cross platform games.
The results of your tests with involving the DualShock 4 controllers made me wonder if there was some sort of input latency that was just endemic to the controller's design itself considering the almost identical results between the PS4 Pro and PC w/ DS4 results. I'm really interested in the development of this collection, mostly because while X1-X3 seem to have been proven as ROMs, there's too much that suggests that X4-X8 may be all ports to MT Framework.
X7 & X8 are definitely native ports. I didn’t realize there was a theory about x4-x6, that’s really interesting! As for the dualshock 4, it likely has a slower polling speed (250Hz Wired, ~500Hz wireless) but on PC we’re using an input wrapper called input mapper. That layer is probably adding latency, we will do some tests later this year on what the ideal way to play is and if that is adding latency. I think some of the settings that Sony requires from developers is accountable for the additional input lag over switch. Triple buffered V-Sync, and all.
It's a little less than theory, I've been gathering information from so many sources, and doing my own testing as well. For one, X4-X6 all universally have the faster loading times, it just seems that the fade ins for the "Now Loading" screens are hardcoded to appear on all platforms, so the benefit isn't as immediately apparent. Being familiar with the three game's file structures due to doing graphics rips from them, I've also found that the files for X4 have had some of their names changed. For example, the Ultimate Armor version of X is a separate character file, named PL00_U.arc in the PS1 original and the PS2 collection, but in XLC it's named PL02.arc. As well, the individual files for X5 and X6 are all in a container format in their PS1 originals, named ROCK_X5.bin and ROCK_X6.bin respectively. In XLC2, that isn't the case and they're all in individual files like X4 was natively, with short little nicknames for all the bosses. I tried to investigate these files using my usual set of tools (GGD, PSound) but everything's been converted into a new format altogether. As well, there's also the differences that are in practice itself. X4 has a lot of sounds that are completely different between the PS1 and XLC2 versions, most notably the Ride Chaser's blaster or whatever. Or, in X5, Zero's saber has a random pitch variation to it that the PS1 original does not have at all, and some audio effects that are present in XLC2 aren't in the PS1 version of X5. Not to mention that either X7 was more broken than we remember or a lot of port-specific issues popped up.
Late to the party. Got X Legacy on my PS5 and boy that X1 latency issue is serious. It took a while to get used to be at least I was able to complete all secrets just adjusting my reaction speed. Just started X2 and getting that capsule on the crocodile stage is infuriating. I simply can't get the timing right
It's a shame that Capcom doesn't seem to have any plans to release a patch at all. If they had, they would already have done so, and the fact that there is not a single word from Capcom on the Steam forums about any of the problems people are having really just tells me that this is a situation that isn't ever getting better, just staying the same.
This was a really interesting video, and I appreciate all the work you guys did to make this happen. This has actually got me wondering if it is just this instance on PS4 where the latency is bad or if this is happening to other titles as well on the console.
It really is unfortunate that if you want to avoid all the input lag you have to go with the console originals or go with emulation. You can't even pay to have a near console perfect collection of games that are over 20 years's old. I highly doubt this will be fixed with a patch because as of release Capcom has said nothing and nothing will be said. It's...sad.
Ok, not-native english speaker here so the language barrier is preventing me from understanding: so how should his graphics be interpreted? Bigger latency means it's less responsive? Is that correct?
Thanks for uncovering this scam. I bought both collections on a new Switch not knowing and immediately felt something was very crappy. Guess now I can at least emulate the OG games on PC guilt-free.
In MMXC for the PS2, X1 and X2 are really not so badly emulated, since the emu only ads to them one more frame of lag. From X4 to X6, they are simply the PSX original releases thrown at one disc for the PS2, since it can run most PSX games with varying levels of audio and graphical incompatibilities.
Thank you for the comprehensive testing. I wish I saw this video before I just bought the PS4 copy. What a sham, X1 feels like ass and I am playing on PS5. Never would have thought a collection like this would have input lag. I just tried X1 on my SNES on a CRT and it is butter smooth, I wish I could get a refund.
Do the new-ish NVIDIA Control Panel "Low Latency Mode" inspire another round of testing? This isn't just vsync on/off or gsync. GPUs often queue up frames (PS4 does this a lot). I know how subjective this can be (without your objective testing method) so take this lightly but setting Low Latency Mode to Ultra (which effectively means 0 frame queue) it felt like the input latency in X1 (and others) improved.
Thanks for the hard work you put into this. Do the unmentioned games (X2, 3, 5 and 6) have relatively similar lag to their nearest sequels that you did test? Or does anyone else know?
All charts are referring to whole 60FPS frames. 1200fps is so fast that the whole screen doesn't even have time to change. If you watch the OLED footage then you can see the refresh going down the screen over and over.
You're doing god's work here. Could you also do some frame rate analysis? That's a trickier test setup because to compare frame time profiles you need a way to automate inputs from game start.
I've done some of my own work on it out of curiosity. The repeatable input is by far the hardest part. Data collection and analysis is trivial. In the first link the imgur album at the top is the most important bit. The rest of the post is just details and analysis. In the second link the most important part is the edit in the middle. I measured a difference when doing manual runs but not when using a low framerate case that didn't use human input. It's difficult to draw actual conclusions from that and repeatable inputs would help greatly. www.reddit.com/r/3dshacks/comments/4hgwg3/performance_and_battery_life_test_of_luma_3ds/ www.reddit.com/r/NintendoSwitch/comments/87b47g/nintendo_switch_seems_to_have_lower_frame_rates/
I've noticed the input lag in the switch version, it made playing x1 and 4 more frustrating, and judging from what I hear, x6 is nearly unplayable. In a game like megaman X, you don't want alot of input lag, and capcom should know and address this.
What about MMX1 on a SNES Emulator? like Retroarch's SNES Core? By the way, what kind of game company releases a collection that is actually worse in many aspects? and they don't even managed to get rid of Slowdowns (MMLC have a option now atleast)
I appreciate this information very much and would love to see more of these comparisons for other similar games. Anyone else want to know how the input latency compares across all platforms on the earlier Mega Man Legacy Collection Games?
I don't believe this at beginning but after playing Mega man x1 on SNES classic and then X1 on the legacy collection, I can say Ps4 has a noticeable input lag you feel a little less response from the controls, noting to get angry but the lag is there
Yeah, the problem from a business standpoint is that everything works. It's not 100% optimized, but it works. They're not gonna waste time and money for a few less frames of input lag, which sucks but that's how the dice roll sometimes.
I’m confused a bit on the Switch part. You explained what the USB and Dongle are, but I’m not sure about the Default Switch and Switch Portable Joycon ones. The graph tells me that the Switch is docked for the default Switch test, but is it using the Joycons or the Pro Controller for that test? As for the Switch Portable Joycon test, is it using the Joycons while attached to the system or while wireless?
How about the GameCube version? Is its latency similar to the PS2? I've been searching for a while and I can't find anyone with numbers on the GCN release of the Mega Man X Collection.
While I doubt of Capcom using the PC version for X4, X7 and X8 ports in PC XLC, if they did then that would explain why the input lag is so low in that case.
You forgot the IOS version of X1!!1! In all seriousness, really glad you made this. No wonder X7 seemed easier on PC than on PS2. I spent a good half hour on Sigma 1 on PS2, but got it first try on PC. Don't play the IOS port of X1, it has Xover cooties all over it.
hey you dont need vsync to tackle screen tearing on PC anymore, so you should totally recommend those versions still, IMO. i'm not referring to adaptive sync/g-sync either. in 60hz games like these you can use any graphics card with RTSS. (Rivatuner Statistics Server) software. Enable scanline sync for the game and you will get tear-free image and no vsync. sounds like magic, but it really works, and more people should be talking about this as it is one of the best ways to play many older (and newer) games, especially 60hz content.
How do you hook up your original SNES to your HDTV? If you aren't using something like an OSSC you might be adding more lag. You should consider a Super NT.
I don’t remember being unclear about this. All TV Lag is subtracted, we tested the Super NT (the data is in there) and also the SNES via OSSC (it’s in there too.)
charvelgtrs I can’t watch it right now to timestamp but it it isn’t in here check the other episodes of Button2Pixel on the X Legacy Collection. One of them definitely discusses it.
I wonder if that's the PS4 Pro specifically. I have just the standard PS4 and I find X4 to 8 run just like the originals when it comes to button response.
Another thing I'm wondering - does the 8BitDo dongle exhibit similar lag with 8BitDo controllers, or is the additional lag only seen with PS4 controllers?
It seems like every time a DS4 was used, it was pretty bad. Even on PC, you observed this. Is it possible that the input lag is more divorced from the console's horsepower than we may expect, and maybe there is something weird going on with the controller or the API being used with the controller? Could explain some of the weird findings here. Maybe some tests on PC with different controllers could lead to further insight. Thanks for the video, keep it up!
On PC we were using input mapper which converts the dualshock 4into an Xinput standard. In the future we’re going to do more tests about different ways to hook up a dualshock 4 and using other controllers like the pro switch controller Xbox one and 360 controllers. I want to know what the best controller for PC is.
Thanks for the time and hard work you put into these videos, not nearly enough people are talking about this subject.
Agreed! I'm looking forward to explaining input lag and why it's really important in a future video because some people don't realize.
It's utterly pathetic that my damned $60 RetroPie setup can emulate these games with less input lag than an official Capcom port on a $400 system. This is pure laziness and there's no excuse.
Console emulation tends to typically suck. Dont even waste your money, just emulate on PC or better yet, get the real carts, hardware and a CRT if you can afford to. Nothing beats a real authentic setup!
@@NintenPizza It's not about the console. The Steam version, where you are literally paying to emulate it on your PC, has 120ms input lag. If you decide to drive while adding that level of input lag (by, say, getting drunk) you can be prosecuted. it's not good enough.
@@cakeisamadeupdrug6134 So just emulate the PS1 version.
As a huge Devil May Cry fan, that intro was just perfection.
The amount of work you put into this is amazing.
Thank you!
In the past, I was never really aware of input latency. I first played Mega Man X on the Legacy Collection, and I noticed so many times where the input did not feel responsive at all. Out of curiosity, I tried Mega Man X on my SNES classic edition, and the difference was night and day. Amazing job on this video, I truly admire how thorough you were in investigating this, you guys are true scientists!
Thank you so much for testing on G sync. Glad you included it. It's currently on sale so I'm researching the input lag :D
Input lag should be something game devs should take into account, because they clearly don't. Console versions should be on par with PC, heck even original 25 year old hardware, for an emulation collection. The fact Capcom can't even make 20+ year old games that responsive is appalling. People that playtested this collection were either incompetent testers or unskilled enough to tell how sloppy the games control.
I wouldn't blame the testers. Publishers often ignore the notes testers make in order to get a game out on a specific date.
Western QA would've likely got it. Noodal's sent me a Post Mortem on Black Ops 3 where they said the QA department complained about input lag. It's an interesting watch.
So... NVidia is better than AMD for lag? - I am very interested in this series, but not so interested in Megaman, since I would obviously play it on a Analogue Super NT. It is interesting that the 8bitDo is so slow, thats very disappointing. - Very keen to see how a non-emulated game like Sonic mania performs. Since the main confusion in this series is the difference between a baseline measurement for the tech (native game) VS Mega Man, which is specifically crap. IE - the difference between the platform being slow, and the programmers doing a bad job
Yea, I haven't owned an AMD card in a while and no one I know has a Free Sync display.
We will definitely be doing Sonic Mania soon as a part of our tests.
Input Lag is actually hardware based, not something a game dev can just magically fix via the implementation of their game. You see, there are drivers for controllers that interface with the console and computer you are using. Those drivers along with the hardware itself is set to send input frequencies for the OS to read and do something with.
In short, don’t blame game devs for the short coming of console hardware.
The DMC intro was pretty creative! That already earned you a like fam
I don’t know how it’s possible but this is even better than what I expected. Glad to see this video come out!
To see X1 being terrible on all platforms hurts so badly. It's my favorite game of all time and like many people was beyond excited for this port.
Same here.
You're doing excellent work here. Input latency is an important factor for me when deciding which version of a game to pick up and for years I've wished I could just go on UA-cam and find a video like this to help me make my decision. This shit is just as important as Digital Foundry's framerate graphs.
I know it's a huge amount of work, but I hope you guys can continue to produce videos like this.
I hope we can too. I was a little disappointed when Digital Foundry only did 2 videos with Noodall's input latency kit. We're definitely going to push to do more videos like this, with Sonic Mania likely being the next one we do.
I have never seen your videos before, but I do very much watch a lot of Digital Foundry videos, so that first 1:50 was the best introduction to your Charisma, Channel and Humor.
I was absolutely dying!
Glad you liked it! I’m a patron of DF, really appreciate the work they do.
Y'all are really hating on Dan for saying the word SNES?
Do you people have lives?
This is what you criticize in videos? I'll never understand the internet.
T Money Fun fact a lot of people in UK and Australia say it SNES.
If you say NES add an S to the front then pronounce that combination is how and why it's said that way.
Plus the quirks of accents.
Fun times.
I say all of this as an Australian.
Pronouncing each letter fully is only a North American thing from what I've seen. I have never heard anyone who is not from NA call it by each individual letter. It is always a single word, SNES. Living in the UK everyone pronounces it that way so you could argue the NA way is wrong to the rest of the world, but it is simply different dialects.
EstharFalcus I'm an American and I've always called it a SNES, or just a Super NES.
This like the C° and F° sit. Literally the same
I wish i could say I don't care, but it actually triggered me to hear that.
It's videos like this that remind me why I like your channel so much. So much time and dedication into something that many people don't even think about, but is the kind of thing I care about immensely.
I'm so flattered! Thanks! :D
I knew the PC version felt good now I got this video to prove it isn’t just in my head, thanks for doing the video, man! :D
Absolutely! Always great to have information. I really do think G-Sync is where it's at for PC gaming. Wish I could jump on the bandwagon but it's too costly for now.
The input lag was so bad on x2 on the ps4 that I had to purposefully plan my jumps early to compensate for it
Quality Content as always gigaboys
Thank you!
Amazing video! Thank you for making this and the work that brought it to life
The effort put into this video alone earns it a like in my book.
Some of the numbers here are staggering. I swear the SNES Classic felt very similar to an actual SNES.
It's always inspiring to see this much dedication aimed at such an important aspect of gaming (retro gaming collections as well in this case). Thank you Dan Videogames.
While this is really comprehensive and I appreciate how this validates my SNES Classic purchase, this is totally your "1.5 A Press Rolling Rocks" video.
This made me laugh and flattered me at the same time!
Amazing work guys! It was totally worth it!
Awesome work, GigaBoots! Having produced thousands of test samples for my own input lag testing, I know how much work it is to do this kind of thing. It's exhausting (and kind of addictive).
Would have been great to see an optimized RetroArch setup in there, though. I know quite well where it would land from my own testing (around 5 ms higher than the original SNES), but still. :) Maybe for a future video!
Absolutely in a future video. Emulation vs Collections vs retro emulation hardware.
Awesome.
By the way, it was also nice to see that your figures for the SNES Classic vs original lined up with my measurements. I had the SNES Classic at ~1.3 frames higher input lag than the original, which is within the margin of error compared to your result.
Also interesting to see is that all of the pre-MMX LC releases of MMX 1-3 have decent input lag, while all of the new collections venture into the "kind of distracting" area. After lots of testing (of input lag in general), I'd personally say ~5 frames of total input lag or less is fine, while input lag above those levels starts to become increasingly distracting.
One thing I've personally thought about is creating a video illustrating different levels of input lag in slow-motion. For example, showing how far in front of a ledge you need to press the jump button in order to not just run into the pit. Would be nice to see what that looks like with, for example, 3, 8, 12 frames of input lag. The problem is of course worse the faster the character can move. I've never gotten around to doing a video like this though, in large part due to a major lack of time recently.
That is precisely the next video we’re doing on this topic because people simply don’t understand and need it explained and illustrated. It’s really important that we make people care, IMO.
Does your 5 frame estimate include display lag?
That will be a really awesome video. Can't wait!
Yep, 5 frames of total input lag from button to pixel. Some people are certainly more sensitive than I am, but I personally feel that 5 frames is fine unless you run around and deliberately look for lag. It's obviously always better with less, though, since any lag will make you play objectively worse.
This video is awesome! Sorry this has been such a hassle, but thanks so much!
I’m really glad you enjoyed it!
Capcom still hasn't fixed the lag issues on Switch. I recently got it for Switch and noticed the lag right away. I even did my own tests like you guys did with my SNES and other systems. A friend of mine swears im crazy and the issue is just my cart, but seeing your video is the proof i need to prove him wrong. Thanks for addressing this issue. I hope they patch it soon.
I wanted it on Switch, but I guess i'll get the PC port afterall.
@@shivasthong4924 It has lag too
I completely lost it at "This input lag's gettin crazy." Seriously, Bob, Dan, which of you did that, it was beautiful.
RetroNutcase Let's Rock!
Why is no one else talking about that great DMC3 joke??? Thanks for the informative video!
it was a good joke, but i think they missed the timing of the joke...
Marcos Danilo no, I think their timing is correct. Dante presses a button to make X jump (not start the music), and since there's so much lag he gets pissed and destroys the jukebox.
I think the reason the joke was missed is because the majority of this audience are megaman fans, and not DMC fans.
something something dante and vergil are parallel universe x and zero
TotallyNotThatGuyYouJustBanned na I was looking at the joke as a DMC fan first and missed it till I looked at it with a broader perspective. As a DMC fan I expected the music to play once the button was presed. That's just me though
I knew I wasn't crazy when I thought the Xbox One X version seemed more responsive than the rest. Thanks for the info!
Its very noticeable in Ps4 many times i swear to god i press jump mega man just stands there. Still beat all 4 games but still im not getting the second half for this reason well unless they fix it
"This input lag is getting crazy" LOL
Im glad someone other than me enjoyed it! Hahaha
6000 views? This needs atleast a 100,000 views with how much hard work has been put into it.
Thanks! We can hope, right?
I loved this video, I've been waiting for this for a while now, Congrats on getting this video out
Thank you!
This is really excellent. Thank you so much.
You’re welcome.
Any improvement or update since this video?
liked and subbed, thank you for doing all this testing, i guess i will just play the games on snes classic. Quite insane the results....
Thanks for the like & sub. You'll want to follow us over on GigaBoots Tech Lab where we're going to put our more technical videos going forward. It may take a bit before they start coming out though because of current production problems. ua-cam.com/channels/J7eptoy8sqcSI_3g8ofKNw.html
Glad to see someone validate what I noticed with these emulated games. Newer developers don't have an eye for detail like us old games do.
Imagine if game Devs had Dan as a consultant, I'm sure the latency would be as low as physically possible.
I'm really impressed about the X8 on PC, I wish I had the means to playing that, twice as much responsiveness must be insane!
It did feel really nice. Keep in mind, I was using a CM Quick Fire 1000Hz gaming keyboard so getting something as responsive is difficult.
@@GigaBoots Do you recommend I get the PC versions of the Legacy Collections over the console versions?
I really appreciate this video man
That would explain everything why I suddenly jumping most of the times into my death instead of jumping correctly out of danger . My favourite was x3 of the collection but what baffles me is that many times when I wanna jump from the wall ,while pressing the dash button , it doesnt register that im going to make a long jump.
Oh and I forgot the charge-shots that I always shoot out after im getting hit by a boss to get the last hit or make huge damage in a fight .This infuriates me everytime because everytime I see the frame where x is getting out of the damage animation I always release the shoot button but the charge shot is suddenly gone . WTF?! Good that the ps1 games are more responsive but still they need to fix X1-3 from the frame latancey and the screen tearing as fast as possible . Thats a no go.
I noticed right away when I played MMX on the Legacy Collection (on a PS4 Pro) just how bad the input lag is. It is playable, it just took me awhile to adjust to. Especially after playing MMX on the Super NT, which really is a great system with next to zero input latency. It's my go to for playing my SNES games in hdmi 👌.
The super NT is a really great solution. I recommend it highly, especially now that they fixed the colors.
But is the Input Lag as bad on a Original PS4? I have the original PS4 and dispite being one of those that rarely ever notices input lags in games, I have the feeling that if the PS4 is as bad as the PS4 PRO, then I should definately notice it (especially since I played X1, 4 and 8 to death and should notice!).
Weird how they include Original XBONE, and the XBOX (X Box One X, Clever Microsoft), they should have also included the original PS4 as well.
As cool as the Super NT is, Retroarch actually lets you play with less input lag than on original hardware, and it doesn't get much cooler than that.
Is it that bad when you play on PS4 Pro? Like I'm planing to buy this collection but now I'm thinking about it… and I don't have another system…
That Dante input lag intro XD Glorious!
This is a fantastic video. I farted around with this on Switch last night and I was pretty sure that something felt off. It's good to know I wasn't just imagining an issue that doesn't exist. I hope that they get this patched. If not, I'll likely go back to the PS2 collection. Having this on Switch is brilliant, but I'm very sensitive to input lag, and its presence here would really ruin the experience.
Great video. Thank you for all of the work!
I was back-and-forth on whether I wanted to get it on PS4 or Switch. Eventually decided to go for Switch since the graphics don't really suffer from playing them portable. Looks like I happened to make the right choice for latency as well, though the Switch still has input lag that seems noticeable even to me (or maybe its just because I've seen videos like this saying it does...).
Awesome stuff to see all of this tested. Also, please just keep saying SNES, somebody that thin skinned deserves to really have it driven home.
John Tortorich Even tho i say it like s n e s, i don’t see why people get that upset over it.
w0ng™ yeah man, i feel you. same people who get pissed when someone says "wally" instead of "waldo". Here in Brazil, the sega "genesis" is know as "mega drive". And thats how i say it, and yet some people make fun because "genesis is the correct name". Lol
I heard so many people say SNES, I'm honestly shocked he got attacked for it.
@@ricmagus Joke's on them, "Where's Wally?" and "Sega Mega Drive" are the original names before their American localisations. So they can stick that in their pipes and smoke it.
Great video
Thank you very much. It was a lot of work. I basically had to cut myself off from more tests so it'd get out ever.
GigaBoots it's a really great video. I saw the first video and subscribed and actually kept waiting for this video to come
They better fix this for the mmz-zx legacy collection
Exactly what i was thinking. Those Zero games will be damn unplayable at these levels of lag.
Thankfuly, it's plays just fine on Mega Man Zero/ZX Legacy Collection. At least on PC, where I played, I do not feel any input lag like MMXLC1. It's smooth. :)
Or they should fix it on all collections.
Thank you for this video. I was waiting for it before buying it on Steam. I don't think I will buy it now. Input lag is something I have little tolerance for!
Awesome channel! Keep up the good work! :D
Anyone have any idea how the lag compares to the Megaman (Classic) Legacy Collection? I was able to get through all of those games on Switch, and found it pretty responsive in portable mode, even though I'm usually pretty sensitive to lag.
PogueSquadron megaman legacy collection classic, part one has fairly noticeable input lag on the switch. I own the pc and switch version and I could instantly tell the difference. However part 2 of the collection is fine and much more responsive.
Just picked this collection up on PS4 and immediately noticed the input lag, and stumbled upon this video. I’m glad you made it so I know I’m not crazy 😅 sad to have wasted $20 though
It's really that bad? Damn
Yes it is sadly@@moister3727
You know fellas, I thought you guys were crying over something silly. I never had an issue with input lag. Until I started really watching you guys, then I started noticing it.
Good job. This shits stupid.
Keep it up guys! 👍
Sounds like the whole 30fps/60fps nonsense.
Nobody noticed it till some UA-camr cries about it, and suddenly, it's the worst thing ever.
Sounds more like an excuse to suck at videogames.
You think UA-camRS invented caring about FRAMERATE?!? That is literally the dumbest comment I’ve read this month.
the month is still young, Dan. :p
GigaBoots Damn, All I wanted to do was show my support and post a intentionally humorous comment. Forgot you can't run from the hate and cancer in UA-cam. Sorry for starting a war Dan!
Next time a guy at Whataburger interrupts your conversation you better pull this video up and make him watch the whole thing.
Great job though, this exhaustive amount of effort and detail in your videos are why so many of us love you guys.
Yea, it'll really save me time! Then he'll just run away screaming!
Input lag is the reason I use ZSNES over Snes9x to play games like the Mega Man X series (inb4 muh accuracy). ZSNES with v-sync enabled is less than half of the delay that other emulators like Snes9x produce. Input lag should be the number one priority when developers make/re-release games. It's probably the most infuriating problem you can have with games, especially platformers.
It's a bit bewildering how much of a mixed bag this collection is. On one hand, sure it's a great value, far better than Nintendo's VC where each of these games would cost $8+ cool high res scans, X-Challenge is ...alright and all that
But the games have slowdown ...where the X-Collection on PS2 didn't, the save system is also worse than that collection.
How are we having trouble with emulation on hardware THIS POWERFUL, I can emulate these games better on my old Pentium 4 laptop than an officially done product. That's not good.
Amazing video, great job!
Yes finally new content. I can stop watching years old videos.
For these brief 15 minutes!
GigaBoots 15 minutes I don't have to think about the existance of the 64 Castlevania games
Two odd questions
#1. Did you test the PS4 and XB1X with the cord physically connected? I heard that solves some latency.
#2. At the time, the games didn't feel much different to me, but I just tried to play through the legacy collection again, just for fun, and everything felt.. off.
MMX I've never had an issue with, but all of a sudden, I was dying to Sigma 1's tower climbs. I could just be getting old (44)
So the test with DS4 controllers + vsync on pc seems to point at the controller itself causing the input lag. I actually wonder if that holds true across other cross platform games.
As I recall, the DS4 controller has more lag wired than it does with bluetooth, but it's still only like 4ms, which is a fraction of a frame.
The results of your tests with involving the DualShock 4 controllers made me wonder if there was some sort of input latency that was just endemic to the controller's design itself considering the almost identical results between the PS4 Pro and PC w/ DS4 results.
I'm really interested in the development of this collection, mostly because while X1-X3 seem to have been proven as ROMs, there's too much that suggests that X4-X8 may be all ports to MT Framework.
X7 & X8 are definitely native ports. I didn’t realize there was a theory about x4-x6, that’s really interesting!
As for the dualshock 4, it likely has a slower polling speed (250Hz Wired, ~500Hz wireless) but on PC we’re using an input wrapper called input mapper. That layer is probably adding latency, we will do some tests later this year on what the ideal way to play is and if that is adding latency.
I think some of the settings that Sony requires from developers is accountable for the additional input lag over switch. Triple buffered V-Sync, and all.
It's a little less than theory, I've been gathering information from so many sources, and doing my own testing as well.
For one, X4-X6 all universally have the faster loading times, it just seems that the fade ins for the "Now Loading" screens are hardcoded to appear on all platforms, so the benefit isn't as immediately apparent. Being familiar with the three game's file structures due to doing graphics rips from them, I've also found that the files for X4 have had some of their names changed. For example, the Ultimate Armor version of X is a separate character file, named PL00_U.arc in the PS1 original and the PS2 collection, but in XLC it's named PL02.arc.
As well, the individual files for X5 and X6 are all in a container format in their PS1 originals, named ROCK_X5.bin and ROCK_X6.bin respectively. In XLC2, that isn't the case and they're all in individual files like X4 was natively, with short little nicknames for all the bosses. I tried to investigate these files using my usual set of tools (GGD, PSound) but everything's been converted into a new format altogether.
As well, there's also the differences that are in practice itself. X4 has a lot of sounds that are completely different between the PS1 and XLC2 versions, most notably the Ride Chaser's blaster or whatever. Or, in X5, Zero's saber has a random pitch variation to it that the PS1 original does not have at all, and some audio effects that are present in XLC2 aren't in the PS1 version of X5.
Not to mention that either X7 was more broken than we remember or a lot of port-specific issues popped up.
Late to the party. Got X Legacy on my PS5 and boy that X1 latency issue is serious. It took a while to get used to be at least I was able to complete all secrets just adjusting my reaction speed.
Just started X2 and getting that capsule on the crocodile stage is infuriating. I simply can't get the timing right
It's a shame that Capcom doesn't seem to have any plans to release a patch at all. If they had, they would already have done so, and the fact that there is not a single word from Capcom on the Steam forums about any of the problems people are having really just tells me that this is a situation that isn't ever getting better, just staying the same.
This was a really interesting video, and I appreciate all the work you guys did to make this happen.
This has actually got me wondering if it is just this instance on PS4 where the latency is bad or if this is happening to other titles as well on the console.
We'll find out..!
Love the DMC Intro lmao
Then the work was worth it. Glad you enjoyed it.
It really is unfortunate that if you want to avoid all the input lag you have to go with the console originals or go with emulation. You can't even pay to have a near console perfect collection of games that are over 20 years's old. I highly doubt this will be fixed with a patch because as of release Capcom has said nothing and nothing will be said. It's...sad.
Analogue Super Nt is the answer for X1 - X3 at least.
Ok, not-native english speaker here so the language barrier is preventing me from understanding: so how should his graphics be interpreted? Bigger latency means it's less responsive? Is that correct?
Yep, it's the time between pressing a button and stuff happening on screen.
Latency = time between BUTTON and ACTION
Thanks for uncovering this scam. I bought both collections on a new Switch not knowing and immediately felt something was very crappy. Guess now I can at least emulate the OG games on PC guilt-free.
So~
Capcom has some things to work out.
Hopefully a patch will come out to improve/fix issues like:
- flickering
- screen tearing
- lag
- input lag
In MMXC for the PS2, X1 and X2 are really not so badly emulated, since the emu only ads to them one more frame of lag. From X4 to X6, they are simply the PSX original releases thrown at one disc for the PS2, since it can run most PSX games with varying levels of audio and graphical incompatibilities.
Odd question, where do X2, 3, 5, and 6 land on the latency?
Like what are they closest too?
The "sness" is literally the official way to say it according to Nintendo, so no one should really give you shit for it.
I know. The comments on some of these videos have been really mean and gross in the shittiest ways.
@@GigaBoots I came 2 years later to say that you can just delete those stupid comments
Damn. I really wanted to buy this. I loved these games so much and they are something I associate with my childhood. Sigh. Thanks for your hard work.
Thank you for the comprehensive testing.
I wish I saw this video before I just bought the PS4 copy.
What a sham, X1 feels like ass and I am playing on PS5.
Never would have thought a collection like this would have input lag. I just tried X1 on my SNES on a CRT and it is butter smooth, I wish I could get a refund.
This video belongs in a museum
Do the new-ish NVIDIA Control Panel "Low Latency Mode" inspire another round of testing? This isn't just vsync on/off or gsync. GPUs often queue up frames (PS4 does this a lot). I know how subjective this can be (without your objective testing method) so take this lightly but setting Low Latency Mode to Ultra (which effectively means 0 frame queue) it felt like the input latency in X1 (and others) improved.
Thanks for the hard work you put into this. Do the unmentioned games (X2, 3, 5 and 6) have relatively similar lag to their nearest sequels that you did test? Or does anyone else know?
bump this
I'm impressed by the scope of this, even doing the PC games. All that's missing are emulators like Snes9X.
Man of Truth were doing a future video on emulators. Wanna be comprehensive.
Damn, you guys are great, keep up the good work.
Tried playing X2 on switch recently and felt like I was taking crazy pills. Glad to see I wasn’t
was the lag on the original snes version tested using an ossc?
Of course.
Glad I am not crazy. The Xbox version does have a noticeably better feel to it. Thanks for putting together this video! Very well done!
I have a question.
When you talk about these frames of input lag, is that tied to 60fps or the 1200fps mentioned?
All charts are referring to whole 60FPS frames. 1200fps is so fast that the whole screen doesn't even have time to change. If you watch the OLED footage then you can see the refresh going down the screen over and over.
Ah, well in that case, I'm having second thoughts about picking up this collection.
Thanks for responding, I was honestly expecting nothing.
You're doing god's work here. Could you also do some frame rate analysis? That's a trickier test setup because to compare frame time profiles you need a way to automate inputs from game start.
willis936 I’ve been talking with people about how to make a framerate analysis tool. It’s definitely not easy but I’ve read some ideas.
I've done some of my own work on it out of curiosity. The repeatable input is by far the hardest part. Data collection and analysis is trivial.
In the first link the imgur album at the top is the most important bit. The rest of the post is just details and analysis.
In the second link the most important part is the edit in the middle. I measured a difference when doing manual runs but not when using a low framerate case that didn't use human input. It's difficult to draw actual conclusions from that and repeatable inputs would help greatly.
www.reddit.com/r/3dshacks/comments/4hgwg3/performance_and_battery_life_test_of_luma_3ds/
www.reddit.com/r/NintendoSwitch/comments/87b47g/nintendo_switch_seems_to_have_lower_frame_rates/
I've noticed the input lag in the switch version, it made playing x1 and 4 more frustrating, and judging from what I hear, x6 is nearly unplayable.
In a game like megaman X, you don't want alot of input lag, and capcom should know and address this.
Fascinating stuff. And here I thought I was weird for getting this on the Xbox One!
What about MMX1 on a SNES Emulator? like Retroarch's SNES Core?
By the way, what kind of game company releases a collection that is actually worse in many aspects? and they don't even managed to get rid of Slowdowns (MMLC have a option now atleast)
I appreciate this information very much and would love to see more of these comparisons for other similar games. Anyone else want to know how the input latency compares across all platforms on the earlier Mega Man Legacy Collection Games?
I don't believe this at beginning but after playing Mega man x1 on SNES classic and then X1 on the legacy collection, I can say Ps4 has a noticeable input lag you feel a little less response from the controls, noting to get angry but the lag is there
This is insane. Is it possible that Capcom could fix this via patch?
In theory but it seems incredibly unlikely.
Yeah, the problem from a business standpoint is that everything works. It's not 100% optimized, but it works. They're not gonna waste time and money for a few less frames of input lag, which sucks but that's how the dice roll sometimes.
So G-Sync fixes everything except for X1? Man that is stupid.
I’m confused a bit on the Switch part. You explained what the USB and Dongle are, but I’m not sure about the Default Switch and Switch Portable Joycon ones. The graph tells me that the Switch is docked for the default Switch test, but is it using the Joycons or the Pro Controller for that test? As for the Switch Portable Joycon test, is it using the Joycons while attached to the system or while wireless?
Joycon communication is wireless either way.
The default is docked with pro controller.
How about the GameCube version? Is its latency similar to the PS2? I've been searching for a while and I can't find anyone with numbers on the GCN release of the Mega Man X Collection.
It’s worse than PS2 is what I’ve been told. Also it’s pixels aren’t sharp at all. It’s not worth me picking up to test b
You get my comment, like and subscription. I hope it was worth it.
Thank you! Hope you enjoy our videos going forward.
While I doubt of Capcom using the PC version for X4, X7 and X8 ports in PC XLC, if they did then that would explain why the input lag is so low in that case.
Thanks saved me money
You forgot the IOS version of X1!!1!
In all seriousness, really glad you made this. No wonder X7 seemed easier on PC than on PS2. I spent a good half hour on Sigma 1 on PS2, but got it first try on PC.
Don't play the IOS port of X1, it has Xover cooties all over it.
and the pc port lol (Please don't kill me for mentioning that one)
hey you dont need vsync to tackle screen tearing on PC anymore, so you should totally recommend those versions still, IMO.
i'm not referring to adaptive sync/g-sync either. in 60hz games like these you can use any graphics card with RTSS. (Rivatuner Statistics Server) software. Enable scanline sync for the game and you will get tear-free image and no vsync.
sounds like magic, but it really works, and more people should be talking about this as it is one of the best ways to play many older (and newer) games, especially 60hz content.
Interesting. I'm aware of RTSS but this if my first time hearing about Scanline sync. Very intriguing.
How do you hook up your original SNES to your HDTV? If you aren't using something like an OSSC you might be adding more lag. You should consider a Super NT.
I don’t remember being unclear about this.
All TV Lag is subtracted, we tested the Super NT (the data is in there) and also the SNES via OSSC (it’s in there too.)
@@GigaBoots The Super NT data is in here? I must have missed it.
charvelgtrs I can’t watch it right now to timestamp but it it isn’t in here check the other episodes of Button2Pixel on the X Legacy Collection. One of them definitely discusses it.
Thanks, what about megaman classic legacy collection, any idea?
I wonder if that's the PS4 Pro specifically. I have just the standard PS4 and I find X4 to 8 run just like the originals when it comes to button response.
Another thing I'm wondering - does the 8BitDo dongle exhibit similar lag with 8BitDo controllers, or is the additional lag only seen with PS4 controllers?
Is the PC version the original release or the legacy collection ones?
It seems like every time a DS4 was used, it was pretty bad. Even on PC, you observed this. Is it possible that the input lag is more divorced from the console's horsepower than we may expect, and maybe there is something weird going on with the controller or the API being used with the controller? Could explain some of the weird findings here. Maybe some tests on PC with different controllers could lead to further insight. Thanks for the video, keep it up!
On PC we were using input mapper which converts the dualshock 4into an Xinput standard. In the future we’re going to do more tests about different ways to hook up a dualshock 4 and using other controllers like the pro switch controller Xbox one and 360 controllers. I want to know what the best controller for PC is.
Soooo... all we need now are tests for PS4 and XB1 Versions on PS5 and Xbox Series X XD