Intels in serious trouble
Вставка
- Опубліковано 15 жов 2024
- As if things didn't seem bad enough with Intel CPU stability issues, things may get much worse! Watch more to find out!
Want to undervolt your intel CPU and be safe? Check out my undervolt video here - • You NEED to do this wi...
Sponsored Links
Falcon Northwest has been building high end systems for gamers since 1992! Learn more about what Falcon Northwest has to offer at www.falcon-nw....
Get your JayzTwoCents Merch Here! - www.jayztwocen...
○○○○○○ Items featured in this video available at Amazon ○○○○○○
► Amazon US - bit.ly/1meybOF
► Amazon UK - amzn.to/Zx813L
► Amazon Canada - amzn.to/1tl6vc6
••• Follow me on your favorite Social Media! •••
Facebook: / jayztwocents
Twitter: / jayztwocents
Instagram: / jayztwocents
SUBSCRIBE! bit.ly/sub2Jayz...
When you spend $300-$400+ on a cpu, you expect it to work as advertised out of the box for years. Not months.
@@DingleBerryschnappsBut also blame Intel for not enforcing any limits.
@@DingleBerryschnappsThe buck stops with Intel. They could have forced board partners to use correct specs out of the box on first boot.
Mine was crashing at stock or if the box, literally 1st boot it was crashing. So nice of Amazon not to even refund me.
@@DingleBerryschnapps how about we just blame all of them for combining their efforts to create this mess. There's never any reason to be "loyal" to a big tech company, especially when they are making faulty products. It will never benefit you in any way.
Details are still coming out but Mobo manufacturers aren't wholly to blame and if anything, only got caught with their hands in the cookie jar because of the underlying CPU issues. As more details come out it's looking more like this is a memory controller/voltage issue on the CPU side. Remember, this isn't just on consumer systems. This seems to be effecting systems in data centers running super conservative setting. @@DingleBerryschnapps
I'm going to give MicroCenter props here. Last November, I bought a 14900K and a 14700K. The rep said, "You don't need to buy extended warranty on anything else, but you should definitely buy it on the CPUs". I followed his advice, and I know from prior experience that MicroCenter will take it back, when it fails.
I TOTALLY wouldn't have taken it on a CPU too - we've been so spoiled by CPUs that have been pretty much the most bulletproof component in your system(or, on the rare occasion that you got a problem chip, it usually showed up pretty quickly out of the box) that I would have *never* believed that an extended warranty on a CPU would be anything but a waste of money.
Good call, LOL
So you now have 4 or 5 years instead of 3? LMAO totally wasted money.
The best advice MicroCenter could have given you would be to buy an AMD 😅
@@RyTrapp0 yep. In over a decade of tech support work across multiple large offices I only had 1 CPU failure (and you can guarantee I was shocked when I was all out of options and finally replaced the CPU, and bam the software stopped crashing!)
Only other times I had a CPU issue was killing it myself by overclocking it too far.
@@volvo09 In 40 years of working with computers/CPUs, I have only had 2 fail. One was a Cyrix, and it ran hot from the start, so the failure after about 3 years wasn't a surprise. The other was a Core 2 Quad Q9650, after about 5 years.
Intel trying to fight AMD so hard that they began punching themselves in their own face.
Yes
@@DingleBerryschnapps
The fight has been won by Intel a long time ago my friend. AMD fan boys need to wake up.
@@RuyGedares_GuyRedares Lies
AMD says, "Never interrupt Intel while they're busy making mistakes". Lol
@@thetheoryguy5544 Yes
14900k user here. I have had a faulty cpu for 8 months now. It has been in warranty for 4 times. They did not change anything in there so far. Always updated the bios and limit the voltages. And every time I get it back from there I am still hoping that it would work. Always intel user, never again. Disappointed. Whatever you guys do, don’t upgrade from 11-12 series. 🙄
wait you RMA'd your cpu and they just sent your same defective CPU back to you 4 times in a row?
@@mehck-gk9yn mine was a prebuild PC. They forgot to add correct bios settings and then 2-3 times they tried to add them but it did not solve anything. CPU is toast. Now the pc is in warranty for 3 weeks already. Maybe I will get a new cpu but I am out of hope already. Silicon lottery at its finest. 😀
AMD bot, no one that I personally know had their CPU die and they have been using it for a year
@@adelgubaishi You don’t know much about anything then. 😀
@@Normaalne You can def get RMA on your cpu, just have the prebuilt company handle the RMA for you. It sucks you'll be without a CPU for a couple weeks but it's better than having a dying CPU
Intel have been milking the crap out of consumers with their overpriced quad cores and constant platform changes for years while also stagnating the industry, I don't feel bad for them at all.
Finally someone who says the truth
The entire CPU industry stagnated before Ryzen, their quad cores (and dual cores) stagnated the industry from 2012-2017.
@@PandemicGameplay it's started at 2011 on sandybridge when AMD gave up to compete to intel
As an employee of Intel …things have been changing ..some of it is public , most is not . And things are changing for the better
100% agree
had my i9 14900kf for a solid 3 months.... died a couple of weeks ago.
mine also and microcenter staff was not nice about it
My last Intel CPU a quad core. Lasted me. 10 years. I operated last year to an AMD 7600. Price was good, low wattage and heat then Intel. Runs great!.
I never heard of so many Intels dying in 20 years. This is bad.
AMD bot, no one that I personally know had their CPU die and they have been using it for a year 👊
Intel should offer full refund or replacement (free upgrade or replacement), we all got scammed by intel promising performance that is now being downgraded... which sucks even more
Not to say performance will be worse, less voltage = less temp = higher and longer boosts
I get 41k+ op Cinebench R21 MC and my voltages peaked at 1.477v (didn't tune voltages). My PL limits are at 275w, Amps 400a, I have 3 cores that can go 6ghz.
My performance went up, my temps and powerusage down by a big margin. But you need to know what to do. Read my comment above. I describe how to.
@@Behemoth33
Why do you think they needed the higher voltages? Sure a golden sample might not need the extra volts but most chips arent golden samples. They either get volts or they dont hit clocks
Replace? With what? Another cpu that will die in 16 months? Full refunds and switch to amd is the only way.
this is why i go amd every time
14:32 - The amps and the volts both can't be too high at the same time with the watts staying the same. P (watts) = I (Amps) * E (Volts). So, if you are at 125W, if the volts go up, the amps must go down. It's the law.
Exactly what I came to say!
Inversely proportional.
Generally yes. But if you take a conductor and just increase the voltage, your current will rise too. P will not be constant.
Fixing P to a number requires the VRM to reduce the current when increasing the voltage.
@@juliusfucik4011 That is in theory, depending on the conductor. If the conductor can't handle the current, the resistance will rise, limiting the current. BUT, in either case, the power will also increase if the voltage and amperage increase. Again, it's the law.
Yup Ohms law
Actually Hardcore Overclocking released a video 5 h before, that 30% of a minecraft 14900K Server farm COMPLETELY died with a max Voltage of 1.539V on a Microblade server mainboard on 3600Mhz.
Yeah this is insane because these server does not have an intense workload on all their cores. Glad I bought my 7800x3D
@@lionhiid And that is the problem.
Low usage = low power = high volts (because both temp and power criteria are met, "TVB" goes to "OH it's 6GHz time !" mode)
Intel was pretty dumb to set high volts limit on VIDs (because that's why they are even an option for BIOS to use).
Greed was too big, and advantages too good to ignore. Well, now they have to deal with this.
Let's hope this is just a small Pentium 4's S.N.D.S. reminder before Core 2 rollout ?
Consumer market finding out the hard way that stability trumps performance.
Surprised to see server guys trapped in this, when I studied Computer Science at uni two decades ago, this was drilled into me. I haven’t kept up, I went a different direction, but seeing professional server operations on consumer kit is weird to me. Like, intrinsically wrong to my mind. That aside, chips so clearly up against a power draw and ghz limit are ones I would actively avoid. Not an Intel thing, not an AMD thing, just … let someone else suck that lemon first.
Not just him as before him Level1Tech did made great video with lot data to tell about servers being effected which has no near of these CPUs on normal boards settings from start or during degradation.
@@lionhiid me too just got one on prime day to finally upgrade from an 1800x I got in 2017.
RMAing itself is kind of terrible and I don't blame people who don't bother. It's easier to just buy a replacement and move on than to waste time trying to get an RMA approved, waiting for a replacement, making a trip to a post office to ship it out etc. When I was a broke student I could justify the time spent on it but now that I'm an adult with a job, it's easier to just eat the loss and switch to a competitor if a product fails on me.
Mine is crapping the bed. I submitted the rma today. I didn’t know this was the issue until Wendell’s video and other posts on Reddit. Thank you community for saving me. I was pulling my hair out over why my pc was just freezing and crashing for no reason all of a sudden
How long have you had your cpu?
@@RuRahKiller22 I got it May 19th of 2023 and I have been using it since then. I have been noticing issues for a few months now and did not realize this could be the issue.
Same question @@RuRahKiller22
Buy a i3-12100F lol, safe buy as a temporary chip.
1.63 VOLTS!?
That’s insane. Not too long ago I was using 1.1 volts on my high end Intel chips.
How Intel has fallen.
I think the 8086 ran at 5 volts. Intel 7 can run at 2 volts with a specialty node.
It doesn’t actually need that. 1.27-1.33ish is usually enough. Which is why they are degrading, they aren’t meant to run with that much voltage or validated for it.
I had a flashback to Athlon XPs running 1.65V.
@@billiam247 I don’t remember it being that high stock, but those things OCd like crazy. I mean like 30-40% with good cooling, which back then was hard to find. I think I had an old system lying around and I just kept increasing the clocks and voltage until it popped in a very smelly plume of smoke. Nowadays you are close to that out of the box it seems like, no headroom for CPUs or GPUs.
They had to do it just to compete with X3D. But, failed miserably. 😂
switched out my 13900K and went to AM5, bought a 7900X3D on Amazon when it was sale for $327 before tax and a used 'Like New' X670E-E motherboard, my first AMD CPU and it's been stellar
An updated undervolting / bios video would absolutely be welcome!
Intel can compete with Boeing trying to get back the trust they once had.
As well as Ford.
@@theplaintechfords pretty bad but id say bmw is arguably worse
@@usymmij Maybe, but I just spent $3,000 on my Ford Escape to have it scrapped because they couldn't fix my brakes because of an unavailable part and couldn't make it road worthy. Moreover, they have a whole line of models that just came out sitting around they can't sell because of a defect. I know BMW is bad, but from my perspective Ford's become the new Boeing of automotive.
@@theplaintech i had an unavilable part problem for a different car (abs sensor loom) every shop said unobtainium, i looked on ebay and found a brand new genuine 1 in seconds 😉
Intel didn't kill people. While it sucks, it's not nearly on the same level as Boeing.
Thanks to you, GN, Hardware Unboxed, Wendell, etc. You guys are on our side, and if you weren't around I bet Intel would try to sweep it under the rug, ignore and delay.
I'm so happy I chose the 7800X3D...
Same! dont miss Intel at all
Same here
my dual AMD 5000 tower is solid as a mountain. 5600x driving my sim rig on one half of the tower, 5950x driving my FPS shooter and adobe rig on the other half. dead stable since i built them right after 5000 released. love my 120 pound desktop tower haha
Yeah I bought my x670 board last week
5800x3d here, and yes very happy I stick with Team Red (both cpu and gpu) Knock on wood I've never had any problems with AMD
I was kind of mad when I built my 12th gen system because 13th gen came out like 4 months later but I've never been so happy to have a 12th gen Intel CPU than at this moment.
I bet. Talk about catching a lucky break and a blessing in disguise!
Didn't GN, Level 1, and Moore's Law all just recently have stories about Intel's issues and said that this was more than just a voltage issue? In fact I think Moore's Law commented on this clearly being an attempt to cover up exactly how bad their problems actually are yesterday...
Yep oxidisation of the via
they are stalling for 15th gen, trying to keep consumer confidence up for the initial sales rush on release. then, they'll turn around after the rush and say "whoopsie! it's actually a hardware defect! 15th gen isn't affected, haha!"
@@flynntaggart8549 This is 100% what they are doing. Stalling for 15th gen.
Did you not watch the video? Jay said as much, as did everyone else.
Lowering the voltage most likely only prolongs the inevitable.
@@brettgrieves which is just speculation at this point because no one has done a full third party FA on one of these. I suspect GN wants a dead/damaged one so they can send it off to their FA lab they have used before to actually get more details. the thing is, is that it doesn't really matter at this point, intel clearly rushed things to try and get wins in the "benchmarks and charts", they should offer refunds or replacements to affected customers, the issue is that you can't trust any of their CPUs at this point, they are all likely damaged, and in the end, their "fix" is to reduce performance, which should entitle all customers to a refund.
Never been so happy to have decided to go AM5 this time... I feel bad for the people affected by these issues though... 😢😢😢
Ditto
@@DingleBerryschnapps So you basically did, what every UA-cam Tech Influencer recommends, set down the voltage manually and you're fine.
And also, people should be able to expect a stable, long-lasting system out of the box, if they don't fiddle with it, a system that breaks down on it's standard settings within MONTHS is just horribly bad.
Not everyone is a PC enthusiast crack building 4 GPUs and 3 Watercooling systems into his minecraft PC, most PC users are probably regular guys, who know how to turn on the PC and that is it, they got no clue about all the background shit.
They just come home one day, like many days, hit ON and the day before the CPU was still working fine and then it is grilled and they suddenly got huge problems and are innocent of these and don't know what happened.
AM4 user and 5950x AMD Fanboy who switched to intel for the "combined gaming + productivity" performance. My 14900KS has less of a multi-core score than a Ryzen 7 7950x3d.
Been using amd for 15 years and it’s only getting better.
My 13900k finally died a few days ago after months of crashes while gaming, video editing, web-browsing and watching streaming services. I am in communication with Intel and should hear about what my options are soon. So far, it's been 5-6 emails about troubleshooting steps and methodology even though I have stated multiple times that I have tried everything short of a lab setting to troubleshoot. I have told them that the issue is 100% a dead cpu but I keep getting boilerplate/copy and paste troubleshooting steps and things to try. I assume it's just the typical C.M.A. (Cover My A$$) hoops that I have to jump through because not everyone knows how to troubleshoot and diagnose issues.
EXACTLY , A lot of people just know to turn on and turn off their system ,many of then don't even know how to get in to the Bios
When is it going out of the warranty period? Maybe they are stalling.
Problem is the replacement will have the same problems. They can't exactly just make new chips right away. My 13900K is still 100% stable in normal use but it is failing some synthetic tests, checked cuz of this whole thing. It's been running 19 months now but I didn't overclock it (for long) and it's been on water with c-states turned off which stops the boosting behavior.
I did run it with ASUSes horrible OC for a few weeks/month at the start before I noticed it was shoving 1.5+ through the core so it might have degraded from that. Hopefully it stays stable for normal use.
@@qzy-179SanTzxkW still has a year and a half left
@@JathraDH I just want a replacement and I'll either sell it for cheap or use it until 9000 series comes out. I switched to my 5800x/3070Ti system for the mean time.
Bought my Alienware in November 2023. Started crashing in game or exiting game into a BSOD after 3-4 months. Computer would restart on its own, wouldn’t wake from sleep. I had a couple in home visits to replace hard drive then motherboard, had same instability. Finally sent it back to Alienware, they replaced cpu and sent it back in middle of July. It’s working good as new for now, have to hope it be ok until a permanent fix. Everything was covered under warranty.
Glad you mentioned manually going in to set your voltage to a certain limits, instead of depending on motherboard bios to automatically set it. I have been doing this for years now.
That thumbnail is a work of art.
It really is.
Your racoon finger is a work of art
all aboard the class action train
@@DingleBerryschnapps Have you watched the video? It's Intel CPUs ''REQUESTING'' more volts. If I apply pressure to the gas pedal and rev the engine to the redline and it blows up, do I blame the engine or myself? I requested more power, not the engine.
@@DingleBerryschnapps are they overvolting amd's chips too?
@@DingleBerryschnapps its intels' problem.
I almost went intel thank God I didn't, loving my AM5 build and loving my 7800x3d CPU. Most efficient and best CPU ever existed. Thank you, AMD.
SIs are responsible for warranty/RMA for preconfigured systems, even if Intel is responsible the CPU issue. The SIs owe warranty service to their customers, it's on *SI* to go after Intel after they make their customer whole. That's why SIs get the preferential access, pricing, and service relationship with the manufacturers. Intel is not wrong deflecting OEM CPU owners to their SI.
Imagine what would happen if Ford or GM told customers to contact tire manufacturers for warranty claims if they sold cars with a bad batch of tires... oh wait that has happened more than once and each time the car manufacturer covered the warranty through the dealership, replaced the tires, then got reimbursed by the tire manufacturer. As if warranty laws are a thing.
Ironic how people recommend intel because of stability and plug and playability, look at where we are now
12th gen is different because Intel changed the cache and e-cores.
Indeed - the process node (10ESF) was improved which meant they could add more cache. As AMD found with 3D VCache, that drives up the power consumed. 12th Gen seems a lot more well-balanced
Because 13th and 14th gen are refresh from that with more e cores and more IPC only from gen 13
No. The oxidation process was changed
A 30% ipc uplift is insane, let's be honest. Pumping excess voltage down these chips with a stock cooler is asking for a degrading chip. Same happened with ryzen 5 3600 chips due to 7nm.
This is really why i love your channel you explain complex stuff very clearly... A cut above the rest 💪💪💪
At this day and age I honestly question why would anyone go Intel instead of AMD.
Long term users. Back in the day AMD was a laughing stock across the board. Barely scraping by in the budget end. Then Ryzen hit, but it only changed the mindset of those of us who actually pay attention to the changes in technology. 🤷 Most people don't spend their time on it.
lol why not, it’s like the world has forgotten about exploding ryzen and 7900xtx GPUs. The only reason people forgot is because only 15% of people own AMD. If 80% owned AMD their issues would have been much higher on the radar.
Because they are fanboys. The percentage of AMD chips that fail nowadays is much lower than Intel's.
@@DingleBerryschnapps "I was able to get a 13700k a few days ago for only $200 because people are returning them out of fear. 😂" Until voltages have already chewed a bit of those cpu's.
the chipsets on intel have been generally more stable than AMDs, CPU itself not with standing,
at 11 minutes, this actually happened to my old 12700k, it ran at 1.5-1.6 volts for around 8 months, when 12th gen 1st launched, and there was a voltage and vrm issue that got fixed by a bios update.
as soon as the CPU hit 90 degrees C, it crashed or BSOD, also some funky ram issues, so now on a 14700k and new ram, there is no longer any issues
Back when I got my 14900K an uneasy feeling came over me seeing it hit over 1.52v with all the heat that came with it. Not an electrical engineer or Buildzoid type expert, but it just didn't feel right no more right than if one, despite not being a mechanic, saw their engine hit over 350 degrees. People online was like "It can hit 1.7v and be okay" and it still just didn't feel right. Feel somewhat vindicated, lol.
1.52 volt is suicide on the CPU in long term! Change the Turbo mode cpu core voltage to minus 0.25 volt or if possible with your motherboars start out with smaller steps. My Asus does not allow to set anything less then 0.25 volt on the turbo mode cpu core voltage. My I7-14700KF went from 1.45 volt to 1.3 volt Max!
@@karlos1060 Thank You. Good advice, indeed. The chip I mentioned above was RMA'd just before all this news started to hit a few months back (It was hitting me with blue screens harder than CrowdStrike). I got my new chip and just turned a lot of stuff down. Don't know the exact numbers and BIOS settings now since I am not on that rig RN, but VID is somewhere in the ballpark of 1.42 or so. Still somewhat high, but it has been more or less stable. I've gotten like 2 or 3 crashes since early May, and turned even more down. Naturally, performance fell off as well, which pisses me off, and rightly so. It's not the performance I paid for and it's not even like I can eat the cost and get a new, better chip since 1700 is pretty much eol. Adding to that, I have a nagging feeling the chip is just not reliable. Last thing I want is to be running some big file transfer, be on an important video conference, or have a boss at 1% and crash, because that's just my luck lol. Eagerly awaiting the supposed fix coming mid-August. Sorry for that massive writeup, but yeah... :P
Yeah, 1.5 V sustained is too high. My i7-13700K on stock is at 1.35 V at idle, and drops to 1.1 V under load. I do wonder what makes some chips pull crazy voltages and others not...
@@peterwstacey Yes, I am far happier with the 13700K. The rig I am on right now is running that chip on an Asus TUF Z790 Plus. Just to give y'all some numbers compared to the 14900K, and this is good because as of this writing this machine has an uptime of 188:03:54 (that's over a week of uptime). Voltage numbers as recorded over that time by HWiNFO, which I have set to boot upon startup and keep running 100% of the time:
Core VIDs = Min: 0.679 V; Max: 1.338 V; Avg: 0.850 V
Vcore = Min: 0.258 V; Max: 1.332 V; Avg: 0.829 V
I am not undervolting or anything, chip is running stock and I am on an older BIOS from about 2 or three months back. Don't have specifics like BIOS ver as I don't wanna restart right now, lol (but will restart if someone really needs). Not super precise or scientific, just ballpark numbers. Anyhow I trust this chip far more than the 14900K. Been running it since July of last year without a single hiccup.
@@Moboproc The latest updated bios of my Asus Rog Strix Z790-H gaming motherboard did not solve anything. It made it even worse! Because there are degraded cpu's they want to shove more volt even in to get it running again without crashing. Just check the Turbo mode cpu core voltage on your motherboard. Set global core svid voltage to adaptive mode. Ofset mode sign needs to be minus otherwise more voltage is inserted. and then the turbo mode cpu core voltage. My Asus board allowes only 0.25 as a minimum. I tried like 0.05 and it sets to 0.25v value. I am lucky to get mine running wich helped a lot in terms of vcore and temps on idle.
My 2 preffered cores are not injected with crazy high voltage anymore and that is the sollution. For sure some more teaking is needed by Intel or Asus as a lot of components relate to eachother. I am just using it for over 24 hours but so far all good.
I did not have a bluescreen sofar in 9 months.
Also: your 3 points of power: watts amps and volts…. Doesn’t work that way…. Two points will give you the 3rd (no matter which two you use)
9:08 where as my 5800x3d is pulling 1.1v
I smell a class action lawsuit
The only people that really win in a class action lawsuit are the lawyers involved in the lawsuit.
@@miker6570
When any company gets hit with that much of a lawsuit; shareholders tend to notice when they are losing on dividends. Insurance and underwriting also notice. This will cost Intel tens of millions, and will send volumes of people on over to AMD - and even Apple
you and steve are the only two people i listen to about these matters
lol wendell and jay first. then steve jump in.
I highly suggest u follow Frame Chasers. This guy single handily solved the gen 13-14 CPUs problem a year ago.
im going to make this easy for everybody giving me their unwarranted opinion on who I WANT TO WATCH... SHADDUP
@@DingleBerryschnapps and you, to call my opinion sad really pisses me off, you know whats sad? responding to comments on youtube to tell them they are sad! get over yourself
@@kuttsukukoff Can you give me a link? :o
One thing I love about the PC community is the amount of people genuinely looking to help others to get things fixed.
I game on PC and it has been a great experience for people willing to help when problems do occur.
This was a simple case of Intel wanting to beat AMD, and the motherboard companies wanting to have the highest benchmarks. So between them they poured far too much power through these chips. And as with the core i9 having 24 cores, it will always going to pull a lot of power.
@@DingleBerryschnapps both
Only 8 P-Cores, the rest are 16 E-cores.
@@DingleBerryschnapps Both. Intel is at fault for pushing the chips too hard trying to keep up with AMD when they know for sure it shortens lifespan. Motherboard manufacturers are at fault for putting out expensive motherboards that half the BIOS options aren't explained at all or they are mislabeled or they just don't work right.
Oh man, I'm happy that I didn't get a 13th gen CPU and went for a 12th Gen I9 when I upgraded 2 years ago.
Jay can dust off his AM5 platform again xD
It will be interesting to see what the 7950X is like, or maybe even the new 9000 series.
Crawling back lmao
yeah i just watched his vid "Why I switched back to Intel..." LOL.. ages like milk.
Well, even if IPC uplift for Ryzen 9000 averages ~15% across the board, if we assume the Intel microcode patch might knock off 15% of performance off of the 13th and 14th gen K SKU's (that's less than what game server datacenter orgs are actually seeing when they limit the 13/14th gen 1x900K chips), for all we know the 9950X could end up being 25-30% faster than a 14900K, maybe even more, all while still while consuming half the power.
@@The1astGuardian *CRAAAAWLING INNN MY SKINNNN, THIS WOUNDS, THEYYYY WILL NOT HEEEAALLL...!*
13:12 that's actually a great idea.
Waiting for video on this 👍👍
The Star Wars reference in the thumbnail was great.
Agreed. That's probably the biggest reason I clicked haha
(As embarrassing as that is to admit)
@@Austn8r Hey at least some people got the reference hehe
Problem seems to be oxidation ... So microcode might delay the slow death a little bit till out of warranty ...
All you have to do is sync all cores on the bios. If your motherboard allows set it to 56 for 5.6 GHz and no problems. Once you do that I would also recommend undervolting for efficiency as well.
sync all cores for p core only or even for e cores to 42 ?
crying with my 13th gen mobile i9 that crashes hang over a year now and the rma 2 times could´nt find anything. with the new knowledge i´m bringing it in again. because now mobile chips a suspected too.
Intel has almost the same problem with its early NetBurst processors. Willamettes and Northwoods melted down in numbers.
These kids don't remember that! We called 'em Dieburst.
Willamette and Northwood had no issues with overheating (real or imaginary).
Prescott had "issues" with running at a high nominal temperature but within the temperature spec they were designed for (AKA imaginary overheating).
@@cmkeelDIM Who did? There is no record or supporting evidence for any Netburst CPU having a high failure rate.
@@tim3172 Who talked about overheating? There were hotspots inside the chips, where thanks to the higher than nominal voltages and currents resulted in practical melting of the conductor, and sometimes even the silicon layers. Well, even in that case Asus motherboards had a higher than others contribution factor. You had to modify your board to provide that much voltage and current. Except Asus boards. Those were able to pump 3x the current and 2x the voltage compared to the nominal. Asus was always an asshole company. It is not a new developement. At that time automatic overclocking was't a thing. You have to do it yourself. All motherboards started with Optimised Defaults, and in those ages that meant compiance to the CPU manufacturers specifications. The only thing motherboard manufacturers handled "lazily" was the External Clock. That was the only thing OOB can gave them advantage. There were almost no motherboards where the external clock were set to the standard 100 MHz value. It was always at least 101 MHz, and in some cases (again, Asus) as high as 103 MHz. And thanks to the PCI bus speed's dependancy on the EC the whole thing easily ended up in instability. And then came the need for the higher voltages. Some people set 2 V VID insted of the nominal 1.75 V max for the Willamettes, and even the same 2 V for Northwoods instead of the nominal 1.525 V max. Nowadays the UEFI itself does this by default, without your consent. Almost all LGA 1700 motherboards start killing your processor from the first power on cycle.
Prescott was a failed design in a failed fab node. Intel's 90 nm node was a joke. Never befor and after a node has so high current leakage. That made those processors a farce. And that was the reason intel never achieved its original 6 GHz goal in clock speed. When they (somewhat) fixed the issue with the 65 nm node, nobody cared about the P4.
10 bucks says intel knows about this issue at release, they just wished no one would notice.
My cpu has had the issue, 13700K, but I live in Thailand so trying to get it RMA’ed will be impossible.
Strange how the country they MAKE the dang CPUs in is the country you can't even return them in, when you could probably DRIVE to where they make the CPUs.
Crazy how globalization fucks things over.
@@DingleBerryschnapps right, so Jay is wrong and Gamer’s Nexus is wrong, but you, a Internet person who showed no reason to be trusted, expect me to believe you?
@@GompMC "He" is a well known intel bot.
I have spent months trying to figure out what was wrong with my system. I replaced my memory, then my motherboard. Multiple reinstalls of windows. Used a different SSD. Even swapped out my PSU for a different one. I didn't bother with my 13900K because it has always been stable and I had the limits enforced.
After hearing about all of this I replaced it with a 7950X3D and the x670 version of the board I already had. Zero issues since. No crash, no instability. Just smooth running. I hated doing it with 9000 series just around the corner but at this point I am glad I did!
I agree it’s very questionable timing that Steve puts out his video and then they put out this statement immediately after claiming they have a solution, but we don’t get to look at it for like more than a month.
How much is Intel paying you bro? Trollolol
Jay, any chance that even older Processors be affected? I have a i7 10700k and started having power issues since 4/18/2024. Ive had the computer for 3 years and its working fine. My sons i5 13700k is doing and giving the same Kernel-Power issues.
Thank goodness I have my Steve Burke "embarrassing"-approved 11900k
11900K is a god for PS3 emulation with the AVX-512
What is avx 512 bro@@Syping
@@alikf8017 an instruction set in the CPU.
I believe modern CPU's dropped it.
@@volvo09 ok that's sad
@@volvo09no Intel dropped it, it's in AMD CPUs and fully supported in 9000 series
Fantastic video Jay! Please do a video where you show us how to manually set the limits to the different cpu voltages to save our CPUs.
They let the market stagnate for a whole decade and allowed AMD to catch up and surpass them- they could have been WAY ahead if they'd not gotten comfortable with their laziness in the 2010s and now they get to see what all that relaxation means- they could have a total Monopoly on consumer CPUs but they mucked that up pretty hard.
AM4 was too good. No one is running out to upgrade to AM5
14:50 - 15:05 the comparison of intel CPU with that Terminator 2 scene was funny XD
I suspect you need to edit your video description:
> Want to undervolt your intel CPU and be safe?
"Safer" is probably a better choice of words.
@@DingleBerryschnapps "Be safe" suggests this will work for everyone, even when the CPU is already degraded. And in the context of the video, "be safe" suggests that it will prevent any crashes etc.
"Be safer" suggests that it will reduce the likelihood of crashes, not that it will prevent it.
Should I build a gaming/multimedia PC with the Intel i9 15900k and a Nvidia RTX 4080 Super/5080 or the Ryzen 9 9950X and a 4080 Super/5080?
What are some good brands for motherboards for both Intel and AMD CPUs? I have heard ASUS and Gigabyte are not great in terms of quality.
Can I use G.Skills 7200mt RGB 32gb (16x2) XMP 3.0 compatible ram with an AMD CPU and AM5 motherboard?
Finally! Once this patch is out and tested, i can stop having to talk customers out of intel chips! When they do insist on them, I've been undervolting my customers' builds. This was after the first time I received a service call on a 14th Gen I9 that crashed constantly in games. Fortunately, Intel accepted my RMA on that chip.
@DingleBerryschnapps Clearly, that's why, after selling a further 30ish builds (undervolted) based on Intels 14th gen. none of them have had the issue the first 3 I sold had. moronic intel fanboy.
@@DingleBerryschnapps you only think you do
@@flofixer1 sarcasm
Is operating @ 1.4V too much for 14700K? Set the power limits as soon as I bought it but did not undervolt it. Struggled to find that info in the first place for my motherboard... so I kinda gave up undervolting.
Will there be any bench to check for damage?
I am glad all my builds are amd
i was thinking of buying 13500hx laptop in flikart or amazon sale under 1000$ which usually cost around 2000-3000$,is the affect only for desktop cpus and not laptop mobile gpus or are they both?like laptops cpus 13th-14th gen are also affected with this?
glad I went AMD
What a bummer... After many years I finally replaced my good old 2600k with 14700k and I build my new beautiful white PC case about 3 weeks ago and now I'm seeing this... It looks like my old 2600k(overclocked since day one) with outlive my new 14700k. Well I guess I better lower 14700k clock speeds and wait for that bios update and pray for the best outcome.
Fingers crossed for my I7 13700k
Same, friend. I got an i7 13700f with my prebuilt back in May. Here's hoping we're good.
A bit of info that I've been able to gleam from intels vague posts. Do your processors have a base tdp of over 65w? If so then your processor is at high risk of suffering from degradation.
One way to determine if your chip is currently at risk is to put your chip under 100% load. Is it running at 1.5v or higher?
If so, you need to take action immediately. Lower your processors pl2 (I have a 14900Hx, and I lowered the vlid max to 1.4v and 105w.) I'm currently running a 100% workload for 12 hours hoping I've got the right settings in place now. But it hasn't crashed once and seems about as stable as I could hope for.
Act quick guys before your cpu ends up becoming permanently damaged.
Hi Jay, I've been into PCs since the first PC Clones arrived. This problem reminds me of that early 1960s movie FAIL SAFE. The punch line was: "We didn't expect this to happen. Our computers must have failed us."
more evidence that the FX 9590 is a GOATED cpu. Literally eats voltage for breakfast.
Lol :)
Haha yes it does until the motherboard VRMs die 😂😅
It also doubled as a furnace so you can stay warm in a cold morning while you enjoyed your own breakfast.
that's not how you use literally!
I adjust 4 things in the BIOS/UEFI: XMP enabled, IccMax 307 Amps, PL1/PL2 253 Watts, and Lite Load 9, which I believe limits the voltage to Intel defaults. I guess it saved my 14900K because I run Handbrake x265 for 4K video a lot. My CPU can spend a week working through a queue operating at 93-96C without a break. I had one blue screen when I first installed it, which I think it was an XMP memory error, but I restarted, and it has been fine since.
You undervolt Ryzen to make it faster. You undervolt Intel so it doesn’t brick itself 😂
I feel like if I buy a pre-built, I shouldn't have to call Intel for an RMA. IMO, someone who buys a pre-built system shouldn't have to take it apart at all. I'm gonna send the system back to the SI, and they should fix it. That's the main advantage of buying a pre-built, the SI provides a warranty.
im still happy with my i9-9900k and 2080 ti combo. i would love to upgrade but the whole system is water cooled so its a mess to swap parts. but i think i can survive another 3-4 years with what i have
I bought I9 13900k in March 2023 and hadn't any issues ever with it. I also did Bios update couple times included last one with Microcode update. Should I still be worried even I don't have any issues? My Voltage before Bios update was between 1.4-1.5V maybe while I was gaming was peaking over 1.5V but just for very short time. Now it runs on idle between 1.1- 1,4V and while gaming 1.3-1.5V Is this still too high?
i wanted a 13900K for a Min. REALLY Glad i just got the 12700K Tho now lol
I got a laptop from digital storm back in march. It has a 14900hx. Has there been reports of laptop cpus going bad?
I had a blue screen for the first occur for the firsr time since i got my computer. Im not sure what caused the blue screen.
Why do America type "month/day/year"? Doesn't make a whole lot of sense to me.
The same reason we use imperial units, routinely circumcise babies, and have guns for sale in Walmart and ammo vending machines - insanity.
Makes sense to me
@@carlpanco7417 To me its like counting 1,3,2.
@Voodoodrul no its actually not like that at all.
@Awesome1914 its just language. Its written how its spoken. Its not about counting, its just communicating. Here we typically convey the month than date than year. Its not about anything deeper than that. It aint about counting.
Do what I did, use the option that reduces the voltage by -0.10 volts or whatever works for you.
Then when you get the bios update, you can go back to defaults.
Does this apply to the i9 13900hx? The reason I ask is because over the past couple months I've noticed CPU temps jumping up on my Acer Helios, and I only have a few tabs opened on chrome. I don't game much on my laptop, it's my daily driver for work and school. And I code a few times a week. I don't think I'm pushing the limits on my CPU, and with a few chrome tabs opened the CPU jumps to over 92c and the fans go to full speed.
yes it does
however idk if this is exclusively the issue you are experiencing. i would look at probably replacing the thermal paste or looking into what thermal paste was used first before going forward
ok i don’t think you can link but i searched duckduckgo for “acer predator helios thermal paste 13900hx” and it looks like it might just be the laptop honestly but you make your own decision about what to do
Like right now my idle temp ranges from 74c to 101c with the average being 85c and max 101c. I have 10 chrome tabs opened.
That's cooking
Friends don't let friends buy Intel.
@@DingleBerryschnapps Then, just try common sense. Lol
I got influenced into buying intel by a friend who works in tech at the pc store when i was dead set on getting amd a week ago.
Nice 😅 I feel for ya. I've learned to not by first gen tech. Think of 13th gen as a rev. 1.0 or 2.0 or 1.1 or 2.2 etc. of 12th gen for example. Same with 14th gen is a new release on a super similar design. (13th gen). I watched the original introduction on Gen 1 (zen) AMD Ryzen chips and I decided to stick it out until zen 3 to upgrade my FX 6300. When I bought my car I waited until it was on the fourth year of its 4th gen design because nearly all bugs were fixed by then.@@Aliibrahim-yx4pl
@@Aliibrahim-yx4pl RIP
@@Aliibrahim-yx4pl same bro
Okay convinced that I should now undervolt my i7 14700KF. What should then be the max voltage be for the CPU?
I'm very glad i ended up going for a 12th gen for my recent first custom build.
Should have got 5800X3D or 7800X3D. Now you just have a hot, slow chip.
@@rustler08 uh, do you know how much extra those chips cost? Literal hundreds of dollars more than mine. I did a budget build with the 12100f. It does me just fine for almost all my gaming needs.
@@grindstone2512100F is probably a banger build for the price especially with a DDR4 setup i think, but maybe not very future proof, but still have a good upgrade path after everything have being fixed
@@rustler08expensive asf
The crazy part of all of this is I have been "red flagging" on how much power draw from the wall is happening for YEARS and just now all of this is crashing down. We should be sipping power from the wall. More power draw means that you are more likely that you're going to burn out components, not to mention get a huge power bill from the power company. Circuitry is NOT resilient and we need to be much more responsible with our electricity usage.
I hope they get class action sued into oblivion for this.
Not ev n gonna happen
I'm one of the people who wouldn't have RMA'd by busted CPU. despite only using defaults and not over clocking anything, I assumed Intel would just reject an RMA from an individual hobbyist.
My CPU fails if I enable AVX2 support. Hadn't realized the chip itself could just be damaged.
I think I'm going to try the RMA process now. Thanks Jat (and team) ❤
AMD all the way.
Intel was fine back in 2014.
I have a Intel i 5 14600K and a Asus b760 plus d4
Before last bios with change on the TVB microcode, with the intel baseline profile (and before it), I was having 1,288 v without XMP activate, with XMP active it raise to 1,45v+- if you desactivate your XMP profile, this v down to 1,288v again
Now, core voltage is always 1,288v without XMP active or with it active.
Everyone going AMD and they will undervolt, and they will enjoy it
Wrooong... since i moved to amd 7800x3d from i5 12600kf...nothing but regrets...intel was awesome...since i 5 6600k i only had amd nothing but problems....moved to intel 12600kf. Maaan i wqs reborn....kid wanted new pc so i gave him mine..and i was a big idiot to get talk me into going back to amd with 7800x3d... regret it in first 2 weeks..thinking now to go on 14700kf or wait for 15th gen intel
You ain't wrong.
Wrong...im on amd 7800x3d..and nothing but regrets. Moved from i5 12600kf... i was fool to get yalk into going back to amd....had only amd cpus since i5 6600k,nothing but problems
@@koky179skill issue and bad luck. Look around.. no one is having your experience bud.
@@christophermullins7163 😂😂 bro i build so many pcs that last to the day.. even all amd sistem ..and no you are wrong there are alot of people having problems with amd too..but since y all amd fan boys keeps yelling like they are perfext nobody talks about it
Imagine beeing a technician who has to flash 5.000 EFIs in a Datacentre on entry level servers (which are mostly gaming related servers), because Intel is very desperate to write "the fastest gaming CPU" on the box. On top of this: Usually EFI Updates on Servers take way longer, because they include all sorts of stuff like "Power Supply Firmware" etc.
Given Intel has shipped radioactive cermic packaged CPUs, then the Pentium FPU bug, then the last batch of the original Atom series CPUs failing after 18 months, just anther mistake in manufacturing.
And will forever be that way
AMD is forced to be better by their underdog position, however they don't have a perfect record. When AM4 came out soc voltage were being allowed to exceed safe limits and destroying (with visible damage) their chips. That said they quickly narrowed down the problem and issued a new ageasa version that lock SOC voltage to 1.3v or less.
I am not in the least anti-amd here. My three main systems are all am4, a 3900x, a 5600, and a 3600. I had 1600AF as my first ryzen system and before ryzen I had an 8350 8(ish) core system my last Intel system was the original core2 quads. Not sure what I did with that, so the platform is probably in box in a back room somewhere along with the 8350 and the 6350 a budy had before I replaced it with the 1600af I was upgrading from. I'm not an AMD fanboy either, I'm well aware the minute they get enough market share vs Intel they'll relax on being the 'good guy underdog'. Though it's been a few decades they've been stuck in that position( sometimes their fault, sometimes dirty dealing from their competition) and have to be good and a good value.
So how worried do I need to be with an i9-14900k system that I got back in May? Is my cpu just gonna crap itself in 6 months? I'm not savvy enough to do all these wild fixes and things like that so, am I just screwed? lol
If you lack the interest to research all these fail safes and take the time to learn how to best protect your cpu I'd just refrain from using it until the bios update comes put that changes the microcode, you could always bring your pc to a repair shop and have them flash if for you as well you don't feel comfortable about it
Makes me thing of the Samsung 990 Pro debacle
Are there issues with the 990 Pro? I haven't done a 'new build' in years and plan to do something in the coming months and the 990 Pro's were what's on my build list.
@@BeastMortThe They are all good now, solid drives. Originally when they were released, there was an issue with them that caused rapid degradation of drive health. This was fixed via a software update, but any drives that suffered from the degradation could not be returned to normal because it was physical degradation.
not 990 pro. 980 pro.
So i was running max 1.447v on VCORE - set bios to “Disabled - enforce all limits” and “intel fail safe”
Voltages immediately jumped over 1.6
Not sure that is a good suggestion there Jay??
i9 14900k on a msi z760-p ddr4, am i cooked.
I got an i9 14900k in a msi z790 w/ ddr5 6000MT/s , xmp enabled . even though i havent had any issues yet i still feel like its going to happen eventually
@@IPLAYMCPE2 I just got my stuff too 😭
@@IPLAYMCPE2 Apply the baseline profile and you going to be OK.-
This whole fiasco Is likely to hit Intel in the pocketbook for many years to come. I think anyone building a new computer today who is even considering a 13th or 14th gen i7 or i9 is absolutely out of their minds. I had already switched to AMD before this problem started and now I probably will never leave.
AMD can sieze the consumer and high end market for the first time in 30 + years, provided they don't repeat past mistakes and cut corners, themselves.
So im currently RMA'ing my 13700k. Will I be at risk if i install my new cpu prior to the microcode correction? your comment about how quickly a high voltage draw can permanently damage my cpu has me worried about this window of time before the micro-code is released/installed on my system. I have the latest MSI bios update from april on my pc atm.
When was the oxidisation issue identified? How long did it take to fix it? What CPUs were shipped with the oxidisation problem? When in 2023 was it fixed? There are 365.25 days in a year, Intel, get a grip and stop taking your customers for fools!! Thanks Steve, great work as usual.
A friend of mine just bought the parts for a new 14900k system but hasn't assembled it yet. What BIOS setting should he use to keep his CPU safe until the microcode patch?
"Can we all just have some shit that works, please?" is the most relatable sentence in tech history, tbh 😂
Just started an RMA process through NZXT for my player 3 prime. Hopefully it comes back with a working PC
Really refreshing to see a video that lays the blame on intel (and board manufacturers) without going ranty brand crusader or making misleading clickbait accusations.
Awesome having a common language explanation of exactly what's happening.
What are the symptoms? My computer on i7-14700kf started to turn off while in idle mode, i.e. I only have a browser open and suddenly the computer turns off and on by itself. The problem went away after I fixed the CPU watts to 125 from 'auto'. During stress it could heat up to 100+ degrees, now it maintains