Watch our initial Oxidation coverage: th-cam.com/video/gTeubeCIwRw/w-d-xo.html And our video with Wendell: th-cam.com/video/oAE4NWoyMZk/w-d-xo.html Watch our AMD Zen 5 architectural coverage: th-cam.com/video/vNlPnruLfjM/w-d-xo.html Support our reporting and grab a GN Modmat! store.gamersnexus.net/products/large-modmat-gn15-anniversary
thank you my dear four xcelent analysis. how i buy four this amazing computer? i appreshiashin youre branes and four youre explanashin xcelent oxygen four the intel. my uncle work four this company many year and I am very very proud four be his friendly. i enjoi.
Will Snowflake be making an announcement about the announcement of the announcement of the Intel announcement for not being oxidation announcement that's about this actually being related to the via oxidation?
@@snowdoggieii exactly. the penalty compared to profits is dismal. they know it and all BIG COMPANIES take advantage of this and the customer keeps coming back. LOOP HOLES DAWG!!! Scumbag tactics, all corporations use.
Let me get this straight. They discovered a FAB level problem that caused VIA oxidation and just sold all the CPUs like they were normal and also refused RMAs when the CPUs later had issues? That should be a smoking gun for SERIOUS liability.
As someone who has worked in production. The bottom floor probably knew about it, or raised concerns about it but it takes a long time for it to be believed/reached at the top level
Also likely related to why they would only say "back in 2023"... it was likely an issue they knew about for a while. in a PR statement you want that window to be as narrow as you can confidently say. That's also likely related to why they aren't saying anything about dates or bad batches.
@@GoalOrientedLifting Also various levels of managers were likely ass covering holding off on reporting issues up the chain as long as possibble hoping it would go under the radar long enough for them to fix the issues without having to tell upper management something went wrong.
@@CMDKeenCZ Agreed, but I really hope "addressed" in this case means that it was "fully fixed" almost immediately. I hope they did not "begin to deal with it" sometime in 2023 and then "fix it" some ten months later. ☹
"No stability issues are related to the oxidation, but also, we know that some stability issues are oxidation related" Also, "please move along citizen", this is a historical issue and we'd like it to stay that way for the sake of our share price and the FTC
@@kitkat2407 I will still buy Intel. Why? I had so many issues with diverse AMD platforms and CPUs in the past 7 years, where AMD also swept everything under the rug. USB issues, issues with SATA ports dying where I had to RMA a motherboard twice, Memory woes, etc. My 13900KS is not affected. So yeah, will still buy Intel. I just hope they improve communication and liability in a crisis like that.
@@haziqtheunique Difference is Microsoft handled the RMA process fine. I remember them sending me a prepaid box to put my 360 in and they fixed it no questions asked. Very different from Intel's handling here.
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.
It didn't warrant a new CPU stepping code for Intel so that does lend some credence to it not being a particularly big issue. They're all a B0 stepping and so my guess would be it was a fairly narrow range possibly out of a single fab and yet they may not know exactly when it started. There are so many statistical process controls involved to try catching issues before they're even noticeable. Definitely hard for us to know how widespread it is but Intel does probably have a decent idea on when it got fixed and a bit of a guess on when it first started.
I had a board running a file server with that CPU die the final night of the warranty! Thankfully AsRock sent me a new board with the revised CPU that still works today.
They corrected but did they recall all of the faulty ones? Apparently not. So AMZ, Newegg, ect and builders like HP, Lenovo, ect continued selling\using the faulty CPUs they had until they were gone...It's bad enough that they give that one year window
As an immortal being, i can tell you. The span of one year is nothing on the big clock. Its barely a fart in the wind.... though Oxidation still smells bad.
Totally. But they have this argument : elitists that buy i9 know what they doing they know how to keep their processors cool and undervolted. They know their shit. It's the same bs these fanboys throw everytime . "Well If You know what u doing u can see that IT overvolts and u asjust in bios to not let that happen" yeah bro this îs how consumers are. Aha..sure. so much work wasted to buy these processors that won't last. These people can't comprehend this situation fully. In my opinion it's the same as stealing my Money. Thievery.
@@pR0ManiacS The thing is some people buying i9's only buy them in prebuilt systems and expect the company they bought it from to do all the work and then they just game on it or do work with it. Some of them don't go into the BIOS at all.
@@vigilant_1934 not some of them . The huge majority of i9 users won't tweak more than xmp profile. People buy high end usually cause there's nothing more expensive in a generation . Most of them are not overclockers. Tuners. Etc. They choose the Best and then they choose the extra option that îs PC assembly and first install. And not a single thirdparty when they build.they didnt tweak as it is supposed to be.
Right now I'm using an r9 7900 on asrock b650 pro rs downclocked 200 mhz or the system crashes. Although I used the cpu with default bios options, it crashed more than my unstable i9 system. I changed ram (which is in the qvl by the way), psu, gpu, even nvme and their slots but unless I activate PBO and downclock negative 200 mhz all games crashed, some under 1 minute. And this is the second cpu, first one was faulty from beginning, its iGpu wouldn't give output. I had to wait 1 month till its replaced, and I am ashamed to send the second one to RMA again. My experience with AMD didn't differ from intel... In my view, both companies need to create a pro level cpu which has all variables standardized with motherboard vendors ASAP. Speed may be a bit lower but they should sell it with a stability guarantee. They could even recommend a stability guaranteed MB and ram. For the last 2 years I'm tired of trying to fix their sgit.
On a platform that's becoming mostly sponsored content that prints out ad revenue, it makes channels like GN stand out. It's so refreshing to hear someone on the platform actually value integrity over sponsorship dollars. Please don't ever change
GN is a weird one, "word journalism" is excellent, testing procedures are great, but then we come to exact numbers and reviews and... noise normalized tests are set up at an unusably noisy level and extremely bad products are recommended all the time merely because they are slightly less bad than other similar ones if a graphic card requires external power it's junk if a chip runs way above the optimal clocks it's junk if you can hear your PC it's broken simple rules that should be respected in these reviews
@@SobeR999 as time passes information changes so they make another video with the update information. It also keeps companies feet to the fire to fix their problems instead of just hoping it goes away
This makes so much sense. I struggled with a PC that just kept crashing while playing just mildly graphically demanding games. All my troubleshooting kept coming back to two things: 1) power issues and 2) unknown GPU issues. Replaced both GPU and PSU. Continued having problems. Countless hours spent troubleshooting and even money taking it to a professional tech repair shop and still no fixes. Finally just rebuilt everything, new mobo, AMD CPU. Haven’t had ANY issues since getting rid of the Intel CPU.
Silicon lottery is not that much of thing as people think. People have same cpu but so many different combinations like psu, mobo and gpu that can affect a lot on cpu longevity undervolting and overclocking. Even the GPU can affect the voltage undervolting stability on CPU, I personally experienced it with different graphics on the same processor.
@@FoxSky-md1ul Not too sure about that one tbh. I had to swap quite a few ones on my simulation rig over the years. Even when I only swapped the CPU with another one of the same type, without touching other hardware, the results on how far you could push the new one were wildly different sometimes. The sample size here is still rather low though.
Idk about the US, most developed world mandates UNLIMITED warranties for manufacturing defects until reasonably expected life. A CPU can be reasonably expected to run for at least a decade
@@dotjaz That's true but in some regions like Germany after 12 months (or even earlier) the burden of proof reverses and the customer has to prove that the defect existed from the start, and good luck with that if the manufacturer is uncooperative like Intel. The real solution would simply be a recall and I hope Intel receives a really painful fine for not issuing one.
@@dotjaz That's not true. I know of no country that gives unlimited warranties. Normally it's around two years, not more. There are some companies which give their customers longer warranties up to lifetime, but that's their own business, not something that contries control.
"The oxidation issue is probably resolved"... For large corporations who got replacement CPU's maybe, but for every end user in the world who didn't even know there was an issue it absolutely isn't resolved because you can't resolve an issue if you are never been made aware there was an issue in the first place. So this means there are still potentially millions of CPU's out there that have the oxidation issue and none of the people who have the defective CPU even know it's defective because Intel refuses to tell anyone, including media outlets or board partners, which CPU's are affected. All they say is "13th gen CPU's made in 2023" have this defect, so that could be any and all produced in that entire year or just within a small window and since Intel refuses to tell anyone which range of serial numbers is affected, we must now assume ALL of the CPU's made in the entire year 2023 are defective.
In a class-action, this bullshit statement will bite them in the ass because any ambulance-chasing lawyer will latch on to it like a trash panda on acid.
And how do you inform everyone about issue even if you wanted to? 99% of people does not view any media about tech, shops absolutely will not care or should contact their clients. Intel can just do it like car manufacturers, put small note on their site, well hidden ofc that there is recoll. In real world people will just buy new cpu and never think about it.
@@wykydytron Way to miss the point there. They don't have to inform everyone individually, they just have to say something, anything, by any means and they didn't even do that. There was never any statement or press release or communication towards SI's or board partners or OEM manufacturers or retailers. NOTHING. The reason for this is simple... it's so Intel could claim they didn't know there was a problem so they can reject RMA's which is what they did and what this video and JayzTwoCents video shows.
yup. all those people who bought prebuilds with no detailed knowledge, swearing but living with intermitant crashes who are gonna end up with an expensive unrefundable paperweight in a year.
Risks of being an early adoptor, of course, it sucks. But you literally are putting yourself at a disadvantage if you don't hold off buying new hardware at release.
According to Germany law, every single buyer of 13th and 14th gen CPUs that are potentially affected are eligible to a full refund by intel for CPUs and Mainboards because they were persuaded into buying the product under the impression of buying a lasting and actually normally working product, so do retailers have a similar claim against intel. And if intel does not provide info on how to determine if a certain product is or might be affected, all count as affected
Most sane countries have laws like this. In NZ it's based on the price and type of product and how long you'd reasonably expect it to last before it breaks. Consumer guarantee. It should be a universal concept worldwide tbh.
You nailed the comment about "only a small number". That's a completely useless statement. Small in regards to what? For me, small would be that they dredged up 7 or 8. Small for them could be 10s of millions. Only a small number of the popular triple-bun fast food burgers were lethal...
@@tteqhu 14th gen seems to have even the locked CPUs affected. My 14500 lasted for days, when it started unaliving itself (high count of random reboots when doing ANYTHING, i´ve even seen green artifacting on the monitor plugged into iGPU. I managed to return it within returning period, which saved me a lot of headaches. Went with 13600K, assuming it will be okay. Random reboots continued 😢, although severely limited (only in idle & low loads) and i seem to have it solved now (replacing the surge protector did the trick - the old one was 7y/o and probably had issues powering up electronics with low voltages - i´ve also had occasional light fluctuations and some monitor issues).
Eh that's just damage control 101. Claim minimal impact. Numbers, or subjective language that should be ignored. Last week's outage was isolated to a subset of users and systems, a fix was rapidly identified and deployed by expert teams at Crowdstrike. We apologise to the limited number of people who suffered minor inconvenience as a result.
Feels to me like that's a reach. It is heavily implied that 14th gen is not affected by the oxidation... and if it is, seems like a blatant lie that's sue worthy
@@trivolous28Wendell's in depth breakdown, and Steve's summary in the prior news video that cited that breakdown, both mentioned that the 14900K and derivatives all suffer the same issues even when run on conservative power targets with datacentre grade motherboards
@@trivolous28 quite in the beginning they say the oxidation issue was just happening in 2023 and is now not happening anymore, and is not the reason. Since 13 and 14 gen CPUs are dying (see Wendells video aswell), this either was a blatant lie, or is actually the truth, and they have some even deeper lying issues.
Considering Intel normally has a 3 year warranty on their boxed CPUs and Raptor lake came out in 2022. A blanket warranty extension on all CPUs to a fixed date 2 years from when the microcode update comes out should be the bare minimum because all Raptorlake CPUs sold have been degraded to some extent. Extending the warranty in that way ensures that Raptor lake CPU owners who have had their CPU for more than a year or two and haven't yet seen instability don't get screwed by damage Intel did to their CPU.
We got like $30 per drive for the WD red settlement a couple years back, but of course if we had known about the issue ahead of time we wouldn't have spent an extra $1k on drives to replace the faulty ones, which were also bad.
@@IAmPattycakes "We got like $30 per drive for the WD red settlement a couple years back" - Have I been lucky with my drives? I never heard of any WD red lawsuit, and I own about 13 8TB Red drives, purchased in 2017 and 2018.
1980: my parents buy a microwave that lasts 23 years before they finally replaced it, and it still worked when they got rid of it. 2024: as the years go buy products are made worse and worse.
@@bernds6587 If intel didn't keep increasing clock speed and voltages to keep it running, their CPUs would not catch up to more efficient architecture of Ryzen. Intel use to make fun of AMD for using "glued CPU chips" and they literally are doing just that.
I need to be honest, I was building my last workstation a few days ago. Something big that is supposed to follow me for a few years to come in my work (3D artist). I was finally done with all the part buying stuff, going to bed with the sense of completed duty, and I just wanted to watch a tiny TH-cam video before bed. That video was yours with Wendell uncovering the shitstorm that was ongoing for 13th and 14th gen. Without that video I would have not canceled my CPU order, and I would have not saved my rig and my work for tedious problems, client service calls, and hair pulled of my head, thank you, the timing was miraculous.
my story is not that wild, but he was reporting on an upcoming new AMD platform called AM5, so I was waiting for it to drop. Now with that 7950X I might even be able to upgrade to the last AM5 CPU gen in like.. 2027? 2029?
Man, I have been pulling my hair out for almost a year now chasing instability issues. I RMA'd my RAM, RMA'd motherboard, reinstalled OS, changed SSD's. Turns out issue is my 13700k. I bought it right when it came out so I'm sure mine is affected by the manufacturing issue. So frustrating!!!
@@Shijaru64 Mostely because I've alwyas had Intel+nVidia builds and because Intel has much higher market share, they tend to be more stable and supported by software. Devs will pay more attention to making sure their stuff works with Intel first before AMD because it's the majority of their customers. Same reason I switched to iPhone from android. Because apps tend to work better on iPhone not due to anything inherent with he iPhone, just that devs pay more attention to it because it's higher market share (in the US at least).
@@intriguingfacts5434 People keep touting the 7800X3D, and sure it's a great CPU, but most people would be better off getting a 7600 or 7600X, as that's likely still going to give more than enough performance, and it costs about half as much as the 7800X3D... And you can always upgrade later on if/when you need more performance.
LTT on his WAN show said something like the "TH-camrs niche" was a tiny group and doesn't "impact" intel that much. They forgot, that most DevOps and infrastructure nerds who works for the service providers and big techs are the main public of these so-called "TH-camrs niche" which makes the tech influencers a pretty respected crowd.
Unsubbed from all their spam years ago, best thing I've ever done. Their "knowledge" of things is just about enough to cause damage and hurt themselves...
TBF to LTT, Intel still maintains overwhelming PC market share, dominating prebuilts and laptops. He's right that regular joes and boomers don't know or care about this issue, they only want Intel Inside.
Well that is true. Have you seen how dominant amd gpus are in youtube comments and influencers? But in the real world, no one buys them. Market share will tell you that. Yes youtube is a niche.
I had to get an RMA on a 14900k, and it took a month. It involved Intel telling me to change multiple BIOS settings while asking me to contact the motherboard manufacturer if I didn't know how to make them myself and repeated BIOS flash updates from ASUS. At one point, they also suggested I had an Engineering Sample (it wasn't) and requested I take a photo of the CPU scan code which meant time spent removing the AOI cooler, taking the photo, etc. for no reason. The problem, from my experience, was that it felt Intel wanted me to jump through hoops rather than accept that I simply wanted a plug and play (not overclocked, just wanted the ability to do so someday) experience that was made impossible by the CPU. The return was completed, and I replaced it with a 14900 (non-K series) that works flawlessly. I can accept a failure in design that wasn't exposed until enough were sold to create a bigger sample size, but the customer service process to have that issue addressed and get a refund or replacement was inexcusable.
@richardw412 Like a lot of Fortune 100 companies the even have external help developing their RMA processes. The goal is to make you go away, give up, etc. Congratulations on making it through the gauntlet.
The issue is turbo mode, essentially. So you are stuck with a stable chip that is now vastly slower than an AMD equivalent for the price. If you manually hard limit the voltage and wattage into the CPU, it's effectively a 2.7 or 3.0 Ghz processor with no turbo mode at all, ever. And that's the ONLY fix - never running it to where it gets past 80C. Such a mess. lol.
And that should be grounds for a lawsuit IMO. One that sees the manufacturer paying you for your time, at the rate of an expert trouble-shooter, plus legal costs and a standard punitive fine based on the price of the product. Your time being wasted by their failure should be something they have to compensate you for.
This is the problem with modern customer support services, they make you jump through many hoops in the hope that you'll just go away. That's their actual business model. I've had this happen before and even when I present evidence that the product is blatantly obviously not working as intended, they give the "well that's normal behavior" excuse and slam the door in your face. Shoutouts to Ifi Audio and Noctua for doing that. They will only start making an effort to fix things if they can see there would be significant legal consequences if they don't.
I don't know if you are in the US, but is it actually necessary to do the RMA via Intel? I am in Europe and had to replace an 14900k, but I sent it back to my online retailer. They told me beforehand that they would then have to send it back to Intel and wait for Intel testing it and see if it was "my fault" before sending me a new one. But how it actually went was that they received my RMA and sent me a new one the same day. Don't know for what reason. I mean I did write them that it wasn't my fault and that all other components are fine. So, anyway, sorry you lost all this time. Maybe next time (hope there wont be) try it with your retailer!?
from what I've been told, the flaw that causes oxidation issues are affected in almost every single CPU, so if you have a 13th or 14th gen CPU from intel, it's just a matter of time
@@aliaksei.kartashou lucky me I got a 12th gen, so I might have a little bias, I heard of some gaming company gathering relevant stats about this issue
Todays News: Intel says the oxidation and voltage issues were not true, but instead the problems were caused by oxidation and voltage issues. Tonight at 10: How breathing could be damaging your brain. Back to you, Steve!
I'm so tired of companies making massive mistakes, only to brush it under the rug for as long as possible and have their customers face the consequences instead. They're so worried about their brand's image that they actively soil it.
They've known of these problems for a long time now. I don't believe they've suddenly just managed to "fix" the microcode right after the problems get made public.
@@zrider100zThey most likely dont want to update until AMD Launch is through and Every Reviewer says „Intel isnt that bad compared to AMD but we cant really recommend Intel right now since they also may be loosing some Performance when the Update will be pushed on 13th and 14th Gen CPUs“. Something like that. Anyways. Point is… Intel WILL have maximum Performance still for the Reviews. I could now put on my Thin Foil Hat and say they always planned this to happen but didnt think it would happen so fast but… I mean who knows. Right. We dont know for sure. But they are very dodgy for a very long Time.
@@flimermithrandir at the other hand, the 9000 Ryzens will be tested against potential defective and failing intel CPUs, so the recommendation will surely fall towards AMD. (my guess will be, no recommendation to upgrade from 7000 to 9000 anyways)
I just bought a 13th gen for my server and feel defrauded. Intel knew about these problems a year ago but still sold the cpus and didn’t tell anyone until they were caught. They’re going to loose a lot of money in lawsuits.
Conspiracy time: OR!!, something huge is about to happen; hence the DEI corps are purposely losing money because all the mega companies probably know economically what's about to happen.
If u in uk check consumer law: you are able to return to your retailer up to 6 years for repairs or replacement where a manufacturing defect is identified- some retailers better at it than others but the law is on your side.
@@scottfarley6867As he said it's a contract with the retailer not intel so that's irrelevant. Though the retailers are going to cause hell for intel if they have to accept a huge amount of returns.
Reason why i want to wait over for the update for BIOS, and see if it fixes issues, atm just runing CPU with Intel Turbo Boost disabled and on intel power settings. At least i can use PC with no issues, yet it's so underclocked it never exceeded 50C on 3Dmark CPU test :D
@@Beastlordius Question is if your CPU was already damaged prior to you doing such, and while it might not show now, will it in the future? The resell price might essentially be nill, cause who knows when it bites the dust.
"Short answer: Oxidation is not related to the instability. Long answer: Oxidation is only sometimes connected to the instability." That's not the same thing. That's not the same at all.
Its simple. Oxidation is an issue only when CPU is under voltages as designed. If the voltages are lowered(and performance), the rusty chips have no problem surviving the warranty period. So it's not the oxidation that's the issue per se, it's the voltages they applied to achieve good benchmarks on release. Understand now?
@@Six_Gorillion They are defective damaged goods. Not working as designed. That is an issue. Lowering performance significantly to maybe have around the same lifespan as expected is unacceptable. Simple as that.
Intel is really struggling - the only thing holding them up is their reliable reputation, which is now crumbling. They've been falling behind other chip manufacturers for years now, despite huge amounts of R+D funding (which is dwindling as investors are pulling out). This "voltage" and "oxidation" issue is a symptom of a horribly managed corporation.
From the first computer with an i486 processor I worked on in high school to the current Ryzen, I have never heard of a processor oxidizing. What a time to be alive.
Microcode patch is just asking for a class action lawsuit since it seems like they won’t be able to fix the stability issues without throttling the boost clocks. Hope they can get this house in order, this is a huge problem.
It also makes me think of Bay Trail coming out half-baked, where it would corrupt SD cards if the speeds got over about 25 MB/s. Did they fix it? Hell no, they put a speed limit on the SD reader.
One that probably slipped the consumer radar but not enterprise is Silvermont based processors suffering LPC, USB and SD card circuitry issue. Plus there's also Airmont's SD card and LPC degradation, which made buying Bay Trail tablets a bit of a risky affair... both of the issues did set me back on considering the HP Pro Slate 10 EE G1 as a portable Chromecast TV, primary for streaming my PC screen while I'm on my bed. (I won't mind the outdated Android version bcus of this) so that stings
@@MagicSwordKingthe microcode issue is that according to Intel the CPU’s microcode is bugged and requesting too much voltage. For example if the cpu needs 1.4v it might be asking for 1.6v and degrading itself in the process. This is very fixable and doesn’t need a frequency downgrade. This also explains why i5s and mobile processors are less affected because their lower clocks means they have actual voltage headroom and if the CPU wants more voltage it might not be high enough to actually be damaging. There’s probably a lawsuit coming over oxidation and RMA denial, but the microcode fix should be enough to avoid legal action (assuming that Intel isn’t lying).
@@GamersNexus This all feels a bit weird. Is the claim this massive quality issue is out in the field and OEMs with these unrealistically large number of time 0 failures are in contact with Intel and they're doing nothing about it? Usually, these issues are worked out between OEMs and manufacturers WRT RMAs and issues found and this whole situation seems to be sensationalized and doesn't add up to me.
Heads up to all Intel owners trying to RMA their cpu. I have been in contact with Intel for over 2 weeks now. Today I accepted the Terms and Conditions for a replacement however my tech support on average only responds every 48-72 hours. Good luck my fellow Intel customers who got burned by Intel.
What I admire most is that Steve does not hesitate to expose serious problems with products of major companies in the computer industry very effectively. Keep up the great work Steve! 👍
I watched an interview der8auer had with an intel engineer, and that dude PROUDLY downplayed power draw issues stating that these cpus are meant to be pushed to the limit
The question I have is, if Intel knew there were oxidation problems in manufacturing, why did they still ship those affected CPUs out to the market? The loss in dies (or even full packages) is tiny compared the impact to their reputation for a few dollars on their quarterly report.
Totally misleading and click bait. Not worhty of this channel. Just 13th gen and just a tiny little fracton. Steve doesnt need click bait openers like this.
So, they’re claiming that all 14th gen instability issues are not caused by a manufacturing defect that resulted in oxidation, since they’re saying the manufacturing problem was solved with 13th gen. Interesting…
Maybe they're being generous with causality - the 13th gen chips that were rebadged Alder Lake parts weren't susceptible so it was kind of solved in that sense lol
@@bosstowndynamics5488 No, 13th gen is just as affected by instability as 14th gen is, which both happen to be made using the exact same manufacturing process, on the same lines and tooling probably.
@@celeriumlerium8266 Re-read what I said, I'm not defending Intel here, I'm taking a jab at the circuitous market speak by pointing out that the only way it could actually make sense is if they point at Alder Lake and say "see, we fixed it" (the issue at hand affects 13th and 14th gen parts but only the high end stuff, both the low end Meteor and Raptor Lake parts, and of course the older stuff using Alder Lake that just got relabeled and sold as 13th gen, aren't affected as far as we know)
@@bernds6587I mean Raptor lake is pushed so hard out of the gate that if excessive voltage is applied the CPU would be cooking itself. 14th gen is pushed even harder and would cook itself faster with improperly high voltages. It also explains why lower tier CPUs like the x600K are mostly unaffected because their much lower clocks need lower voltages and improperly high voltages may not be high enough to be damaging. I am inclined to believe that most instability on slower Raptor Lake CPUs like the i5 and mobile chips is oxidation (should be fixed) and the higher tier CPUs are just degrading themselves with voltage requests being improperly high. Intels response is pretty bad, but the problem seems very fixable now as they are promising a microcode update in August.
AMD - it never was alive in first place AMD - for long winter nights you no longer need heating AMD - what's compatibility? AMD - 2x price of Intel outside of USA because f u. And so on.
I just want to point out their current CEO was their CTO (2001-2012) during their long standing anti-trust issues (2002-2006).. given that, I find this behavior unsurprisingly on-brand.
Intel lying about whatever is "on brand" - they have a whole department dedicated to nothing else (marketing). Sad to see that as of late AMD is (once again) going down the same drain.
@@dudebroguymate Apart from the very misleading RDNA3 launch graphs - these you could attribute to them genuinely believing in being able to still pull it off somehow - their marketing as of late has devolved into some very dodgy claims. The worst being: - For the latest AM4 launch (R7 5800XT), they claimed to match the i7 13700K in some gaming benchmarks. Testing was done with an old, very low end RDNA2 card (6600 ot 6600XT, can't remember which one exactly), carefully avoiding a CPU limit. That's just bullshit, they could've also claimed that an R5 3600 matches the Intel part (or the "new" 5800XT for that matter). That's not lying per se, because that's certainly what happens when you pair these CPUs with an RX 6600 - but it's still dodgy af because it says nothing at all. - During their recent tech day they also made some weird comparisons - R7 9700X vs. Intel i7 vs. R7 5800X3D, across various slides, and no matter how you sum up the results, they just don't match up. Worst of all - I don't get *why* they are doing this. It's utterly unneccessary to weirdly hype up their CPUs that way. The 5800XT is what it is - a bleed off for Zen3 CCDs they still have to make for Epyc Milan. It'll sell anyway, no matter how. And it's certainly not meant to compete with RL i7s, so why the dodgy marketing. Just releasing it and taking the good press for their stellar platform longevity would've been more than enough. And as to Zen5? It already looks to turn out pretty good all by itself. Fast, efficient, cost effective and on a well established, future proof platform. With little to no competition by its launch. So why wiggle around like that? Just present the graphs, show the R9 9950X completely destroy the i9 and be done with it. Hammer home the TDP reductions all day, tbh. Satisfy the enthusiasts with Curve Shaper (awesome!) and on the fly memory tuning (stellar!). It all looks exciting enough. Maybe tease X3D for late Q3 / early Q4. There's no need to overhype anything and open yourself up to criticism.
Major companies have been getting away with these wishy-washy statements and actions for too long. Thanks to people like you(and your team), we're finally starting to get some pushback on the dystopian doublespeak. Bull shit is bull shit and should be pointed out to be such in no uncertain terms.
I just want to say "Thank you" for keeping up the pressure on these big companies. If you didn't "threaten" the failure analysis lab, who knows, Intel might have never admitted to the oxidation thing.
@@Savethepandabears So if no one complained, researched and broke the news, Intel would have still come forward admitting their fault, admitting that they had known about manufacturing problems in 2023 and thus KNOWINGLY SOLD DEFECTIVE PRODUCTS to customers? Just how much do you love multibillion companies to be so naive?
@Savethepandabears it's not just this channel, but plenty of others are looking to GN because who else is spending 5 figures to do Intel's homework for our benefit?
my last intel cpu, thats for sure. bought 13900k last year and got about 20-25 different bsod, swapped every component twice(!) without suspecting the cpu. with no other component left, i contacted intel and they replaced it, not having any issues since. still might have a timebomb in my rig now... nice feeling!
Agreed. Can't see me buying intel ever again, the trust is gone. my 13900k still works, but I need it for very, very hard workloads and now I just feel like sitting on a timebomb. Not using it to preserve it? Wasted money. Using it and potentially destroying it? Also wasted money. End consumers are left with a lose lose type of situation. And i wanted to get a 7000 series setup at first! Damn, why didn't I.
@@nikethunner2732 You have to limit and lock the frequency. Based on Framechaser and Buildzoid videos, the issue is that the 2 best cores request voltage above 1.5 V under light workload to boost to the max frequency and that degrades the CPU. The only solution is to lock a lower frequency, limiting power or temperature cannot help.
Maybe you could add a modifier to the performance charts to reflect 0% performance for 50% of Intel CPUs 50% of the time they work every time, except for when they don't.
Microcode updates on this issue are band-aids. They must be buying time to stop the bleeding of RMA’s, until they address the inherit issue of electromigration from a foundational level
I wonder if without the reporting from GN Intel would have been public about their oxidation issues, especially considering how they made their statements.
I think there was enough other coverage and pressure they would have eventually, but with Steve and GN sounding the alarm I suspect they stepped up the timeline significantly.
Intel woulda probably tried burying it. Aside that, I have said it here before, what Steve and GN's team as a whole does could well be dangerous work. Some of these companies get really, really nasty and outright hostile when people expose them. This is especially true for mega corps like Intel. The fact that Steve and team puts their neck out there to help us regular folk is so worthy of admiration. Takes some serious bravery.
I doubt it. The words “failure analysis lab” coming from Steve though… they knew they’d be caught. Maybe that’s cynical, but it’s not my fault I’ve been trained to expect the worst, most dishonest, and corrupt decision making from corporations.
Class action lawsuits always make money for PR Firms and lawyers… I have been included in 2 1. Ms fixing price issues spent $625 got $68 2. Ford DCT transmission $ 84 credit… but I still might have to get transmission replaced at some point in time!
So my take on this is that Intel saw that they had a good thing with the 12th gen, and instead of investing in new designs to keep up with AMD they decided to just keep dialing up the 12th gen architecture by slapping in more cores and pumping voltage and clock speeds for those sweet benchmarks. It's unsurprising that the 13th/14th gen chips are failing, the 12th gens were already pushing it. The whole thing is like a car maker introducing a turbo'd sportscar that's a great success, and then keeping up with the competition by simply adding more boost every year. And then when engines start blowing up they just say oops sorry, we can't help you if your engine is dead but if not come in for recall service. And then they "fix" the cars by removing the turbos, and then say "all better now the engines should run fine." But people paid extra for the turbo, right? So where's the compensation?
@@ShlomoWalfish I don't know that mobile chipsets are even affected by this, I've seen conflicting reports on that. Laptops with 12th gen chipsets are generally discontinued at this point, but there are plenty of refurbs and you can probably find new-in-box if you look hard.
@@Xnoob545 Yeah especially since AMD isn't doing so great either, and I really don't want one gaming CPU maker to dominate and jack prices up like has happened with Nvidia GPUs.
until intel releases theirs which uses tsmc n3b, which means they have the nm advantage compared to zen 5 which uses tsmc n4, which is just enhanced n5 oh, and amd themselves also admitted that their regular zen 5 chips aren't even worth it compared to a 7800x3d, which means that regular zen 5 is DOA i'm not defending them but both fanbases are a disease to both companies
I think the math bug had very little real world impact though, unless you were running very specific calculations with certain optimizations. It certainly didn't result in broken CPUs that have to be replaced.
@@gorak9000 I agree, a useless CPU is definitely worse. They did do a recall on the Pentiums, I'm just not sure of the financial impact it had on Intel, or if this will end up being more expensive for the company.
Even worse if you tell me, if I remember right a microcode change fixed the issue 100%… it rarely caused instability issues.. And back then might have been 3 SKUs… now it’s at least 9….
99% sure Intel discovered the oxidation issue and said to ship the CPUs anywas as they didn't want to spend the $ to fix or take the Stock price hit, in hopes that only a very few CPUs would have any amount of issues from it and if they did it'd be well out of the warranty.
Of course they did, that's how Corporate America rolls. We don't do anything that doesn't serve the banks or stock markets. Quality products made by workers who are financially secure with a decent work/life balance? Ha haha ha ha ha ha. At least this defect doesn't result in fatalities. GM killed 124 people due to a known defect with the ignition (en.wikipedia.org/wiki/General_Motors_ignition_switch_recalls). Do we even have to talk about Boeing 737 Max's 346 deaths due to MCAS (en.wikipedia.org/wiki/Boeing_737_MAX_groundings)?
Underclocked since day one. Water cooled. Two p-cores completely unusable to the point I had to disable them in the bios after days of mind numbing testing and bsod's. I will not be jumping through hoops to get an RMA. Intel, you lost a customer of 20 years. Great job.
@@user-to7ds6sc3p You're average PC user doesn't give a fuck nor knows anything about this. I'm pretty sure intel will still dominate the market in the future.
Wait; so they knew about the oxidation issue last year and are only just now admitting it may have something to do with the instability issues on their CPUs?
Yes and no. They are saying that the oxidation issue is NOT related to the instability issues, but they are also saying that the oxidation issue affected only a small number of instability-affected CPUs. Don't think about it for more than three seconds or your head might explode.
They knew of an anomaly. They did not know it was an issue until later. I'm sure they know of many things. They're not always aware of the significance of any of it though.
@@mardus_ee what difference does it make? They were made so far removed from common reality that it doesn't matter. In clean rooms never touched by human hands.
My 13600kf just failed last week. I had no idea this was even a thing. Weirdest thing is no BSOD, no crashes, no boot problems. Was just playing Black ops 6 and suddenly no video output and no peripherals.
"Only a small number" - yeah, same verbiage used by Apple during the butterfly keyboard fiasco, display cable tears and such. Which we all know, was looking nothing like a small number.
Thought my 13600K was immune since it had been on almost 24/7 since Oct 2022 (launch) but then started seeing symptoms in the last couple months such as odd lag when switching between Tabs in chrome and sometimes tabs would hang and crash, also started seeing really long Nvidia driver update times (no crashes tho), normally would take a few minutes but was now taking 10 minutes, even tried DDU but it didnt help, went into BIOS and bumped up the CPU Lite Load setting from 10 to 11 and this solved all the problems and system in general feels faster & more responsive especially when dealing with Chrome, but of course it still means degradation if bumping up voltage solved it, I've now updated to the latest BIOS & 0x12B microcode, I was on 0x129 before which is concerning but perhaps 0x12B really will stop further degradation if Intel's claims are true. EDIT: Intel baseline profile defaults to mode 18 CPU Lite Load so my setting of 11 is technically undervolted from stock still, I'm not having to run the CPU over stock voltages to keep it stable.
Working in Unreal Engine is when mine stated to display issues (memory address issues, crashing the engine when saving) The replacement one has been good however.
@@Ravenx217 Jayz2cents had a short video - even on the intel "failsafe" (or however that was called) setting it pushed the voltage to over 1.6V. insane. Better check your voltages, like with hwinfo to display the max.
@@bernds6587 yeah jay doesn't know shit- intel's fail safe uses the highest voltage to brute force stability. i set my svid to typical until that stopped working, now it's back to auto (LLC referral) with 1.2 SA and 5.6Ghz lock on all cores. This keeps me at 1.31 vcore but my games still crash sometimes. It's borked.
Thank you for covering this issue, do you know if any Laptop CPU are also impacted by this issue? We have a bunch of i9-13950HX on laptop workstation and would like to know if they may be impacted
I bet it is worse than 50% failure rate. I suspect that is the failure rate under warranty. How many of us replace our machines just because the warranty ran out? And unlike the usual "bathtub curve" for tech failures, this one is probably going to have an exponential failure rate. This is a failure curve that looks more like the failure curve for automotive tires...
What drives me wild is they had the issue with the 13th gen processor, so they went ahead and made the same errors with its successor. That and the lack of communication, of course. I spent countless hours trying to pin down this issue so real glad we're finally talking about it.
This is where the 10nm "Intel 7" is hitting its limits since 13th gen. They have to crank the power otherwise they can't compete in single threaded workloads with AMD.
It’s because 14th gen is really a minor revision of 13th gen, not a new architecture. Calling those chips “14th gen” was a marketing decision. From a technology perspective, they’re really just additional chips in the 13th gen range.
They pushed “innovation” over stabilization - it’s a bad situation and we shouldn’t be satisfied with their explanation. At this rate they may be subject to litigation!
the solution lingers in their fabrication. unless they are aiming for more altercations and worse evaluation of their stock...option? next one must make a haiku
Back when I was running an IT department, I stopped buying Intel based servers and laptops because AMD was making better processors and they were cheaper. Most PCs I have bought since then for personal use have also been AMD. When my son wanted to build a gaming PC, he had already decided on AMD over intel so I didn't have to give him "the speach" on why AMD was better. This whole situation makes me feel even better about my decision to support AMD over Intel.
They're getting their strategy from car dealerships that have been protecting profits over taking responsibility for shoddy products. Just put some sawdust in there to quiet it down until the warranty expires. Then it's their problem.
If they identified the problem at some point in 2023 but did not do a recall, not inform effected customers and not even acknowledge the issue in RMA claims, can't they be held accountable in court?
@@Six_Gorillion lmao. Intel going to get freight trained as AMD Builds boost. I'm also amazed jtube allowed your handle. Iv'e also found handles the best way around censorship on heavily moderated sites. it's like they didn't think about it. I used to havea pic of uncle adi with 'in your heart you know he's right' and it lasted on here for at least a year till someone reported it lmao. 2024, the year 'they' named themselves to all the normalfogs. Enjoy it bud ;D
@@diamondlion47Yep. Intel is a scumbag of a company. Which is why I stopped buying them since 2006. My last Intel CPU is a Merom on a laptop. When I heard their illegal tactics against AMD (circa 2006), I vowed to vote with my money to never ever bought their products ever again, and since then I've owned AMD's Sempron, Phenom II X2, Phenom II X6, Ryzen 1400, Ryzen 4300U, and my next one will be a Zen 5 for sure. I'm not a fanboy. I'm just giving my money to a company that's not an asshole.
They massively upgraded their testing after that and a couple of years ago the new CEO said why are we wasting so much money on testing and stopped testing. And here we are.
Watch our initial Oxidation coverage: th-cam.com/video/gTeubeCIwRw/w-d-xo.html
And our video with Wendell: th-cam.com/video/oAE4NWoyMZk/w-d-xo.html
Watch our AMD Zen 5 architectural coverage: th-cam.com/video/vNlPnruLfjM/w-d-xo.html
Support our reporting and grab a GN Modmat! store.gamersnexus.net/products/large-modmat-gn15-anniversary
Ye gawdz, we need data on what lots, etc. are at risk and how to obtain replacements.
thank you my dear four xcelent analysis. how i buy four this amazing computer? i appreshiashin youre branes and four youre explanashin xcelent oxygen four the intel. my uncle work four this company many year and I am very very proud four be his friendly. i enjoi.
Will Snowflake be making an announcement about the announcement of the announcement of the Intel announcement for not being oxidation announcement that's about this actually being related to the via oxidation?
@@WarkWarbly i have wife.
Shave one side of your head to look like Skrillex before it's too late.
Intel trying to avoid the most expensive CPU recall since CPU’s were invented
Ever since their last recall (yeah, that happened once)
If you own a 13 or 14 gen find your receipt so you can get in on the class action
@@futurepastnow No way! you waive liability for a 10$ check?
@@snowdoggieii exactly. the penalty compared to profits is dismal. they know it and all BIG COMPANIES take advantage of this and the customer keeps coming back. LOOP HOLES DAWG!!! Scumbag tactics, all corporations use.
@@tablettablete186 The market was so much smaller back then. This would be far larger.
Intel is doing incredible marketing for the 9800X3D
For gamers only.
Their doing excellent marketing zen5 in general and a minor endorsement for previous zen cpu's on the side.
AMD put a mole inside Intel, and they fucked up the chips.
@@LiberatedMind1 i bet intel wishes that was the case, but no this is just intel catastrophically failing
Like amd doesn't have problems, as example less than 1 y ago, burning chips and melting motherboards
Let me get this straight. They discovered a FAB level problem that caused VIA oxidation and just sold all the CPUs like they were normal and also refused RMAs when the CPUs later had issues? That should be a smoking gun for SERIOUS liability.
As someone who has worked in production. The bottom floor probably knew about it, or raised concerns about it but it takes a long time for it to be believed/reached at the top level
Boeing style
Also likely related to why they would only say "back in 2023"... it was likely an issue they knew about for a while. in a PR statement you want that window to be as narrow as you can confidently say. That's also likely related to why they aren't saying anything about dates or bad batches.
@@GoalOrientedLifting Also various levels of managers were likely ass covering holding off on reporting issues up the chain as long as possibble hoping it would go under the radar long enough for them to fix the issues without having to tell upper management something went wrong.
@darthkarl99 that feel when youre send a suggestion, but it has to go through 11 layers of managers 😂
Unreal Engine supervisor at ModelFarm blasts 50% failure rate with Intel chips - he just went public with this couple days ago as well
If Intel knew about the oxidisation problem in 2023, they should have announced it in 2023. There's no excuse and they need to fix their mess.
But line need go up!!! Who cares about next quarter 😊
@@firstname565 understanding material conditions goes a loooong way into truly understanding line goes up
@@odjsjaks They do say it was "addressed back in 2023" though, so they must have known about it back then.
@@CMDKeenCZ Agreed, but I really hope "addressed" in this case means that it was "fully fixed" almost immediately. I hope they did not "begin to deal with it" sometime in 2023 and then "fix it" some ten months later. ☹
Is it possible that contemporary AMD processors were also affected by fabrication defects, or is it only Intel's 13th and 14th Gen Core processors?
"winning the silicon lottery" has taken on a whole new meaning
If its says AMD on it, congrats you've won.
winning the silicon lottery on Intel means it can run at stock.
Want to win the silicon lottery? Don't buy Intel, then.
@FrancisFjordCupola hindsight is great
@@WalterEKurtz-kp2jf Hindsight? People should have stopped buying Intel altogether when Zen 2 launched. That was 5 years ago.
"No stability issues are related to the oxidation, but also, we know that some stability issues are oxidation related"
Also, "please move along citizen", this is a historical issue and we'd like it to stay that way for the sake of our share price and the FTC
Processor with a DOA core might have been due to oxidation.
Don't forget to buy the upcoming Intel CPUs they will surely Tell you about Problems with it soon enough so maybe in 4 years or so.
@@soundsparkI can see that. I just don't think of DOA when I hear stability issues. I think it runs, but not properly.
The big customers should sue intel, lets see how that goes for them.
@@kitkat2407 I will still buy Intel. Why? I had so many issues with diverse AMD platforms and CPUs in the past 7 years, where AMD also swept everything under the rug. USB issues, issues with SATA ports dying where I had to RMA a motherboard twice, Memory woes, etc. My 13900KS is not affected. So yeah, will still buy Intel. I just hope they improve communication and liability in a crisis like that.
Oof. 50% failure rate?
Imagine if literally anything else you bought had a 50/50 chance of self destructing out of the box.
Launch Xbox 360 type shit tbh.
@@haziqtheunique Difference is Microsoft handled the RMA process fine. I remember them sending me a prepaid box to put my 360 in and they fixed it no questions asked. Very different from Intel's handling here.
you're car just breaks down every other day you go to work 😆NEVER GO INTEL GUYS
50/50, oh boy u did think u playing and gambling on a gacha game, except when ur 50 missed, u are completely fked.
@@haziqtheunique At least Microsoft admitted they fkd up and fixed it, Intel is just pretending like they didn't notice there was a problem.
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.
Intel: uhh, iunno it was WAY BACK in 2023.
I am sure Intel did a FA on all of the 13th gen RMAs in 2023 to ensure that large customers were made whole. It’s called sarcasm, people!😂
@@GamersNexus heh, January 1st or December 31st, what difference does it make at this point??
@@GamersNexus I think the survey might help answer that
It didn't warrant a new CPU stepping code for Intel so that does lend some credence to it not being a particularly big issue. They're all a B0 stepping and so my guess would be it was a fairly narrow range possibly out of a single fab and yet they may not know exactly when it started. There are so many statistical process controls involved to try catching issues before they're even noticeable. Definitely hard for us to know how widespread it is but Intel does probably have a decent idea on when it got fixed and a bit of a guess on when it first started.
amount of network routers I replaced due to Intel atom clock failure is probably far beyond this fuckup. Intel had quite a few now
I forgot about that one! That was a good one too!
Don't forget the Puma based cable modem latency issues too - intel is circling the drain
The NAS servers using those affected Atoms too
The Atom C2000 / C3000 issue where they die...
I had a board running a file server with that CPU die the final night of the warranty! Thankfully AsRock sent me a new board with the revised CPU that still works today.
So, Intel corrected the oxidation problem somewhere between Jan 2023, and Dec 2023? Good thing we got specifics during such a short period of time.
They corrected but did they recall all of the faulty ones? Apparently not. So AMZ, Newegg, ect and builders like HP, Lenovo, ect continued selling\using the faulty CPUs they had until they were gone...It's bad enough that they give that one year window
That means they can expect even more RMAs I guess
Actually, it could be sometime in Jan 01, 2023 12:00:00.001AM and Dec 31, 2023 11:59:59.999PM
@@vikramsinhdantkale8525 consider timezones. an extra day due to time differences!
As an immortal being, i can tell you. The span of one year is nothing on the big clock. Its barely a fart in the wind.... though Oxidation still smells bad.
I can fully see them denying RMA's in the future because the user failed to update the microcode and the CPU became degraded.
This
Totally. But they have this argument : elitists that buy i9 know what they doing they know how to keep their processors cool and undervolted. They know their shit. It's the same bs these fanboys throw everytime . "Well If You know what u doing u can see that IT overvolts and u asjust in bios to not let that happen" yeah bro this îs how consumers are. Aha..sure. so much work wasted to buy these processors that won't last. These people can't comprehend this situation fully. In my opinion it's the same as stealing my Money. Thievery.
@@pR0ManiacSr
@@pR0ManiacS The thing is some people buying i9's only buy them in prebuilt systems and expect the company they bought it from to do all the work and then they just game on it or do work with it. Some of them don't go into the BIOS at all.
@@vigilant_1934 not some of them . The huge majority of i9 users won't tweak more than xmp profile. People buy high end usually cause there's nothing more expensive in a generation . Most of them are not overclockers. Tuners. Etc. They choose the Best and then they choose the extra option that îs PC assembly and first install. And not a single thirdparty when they build.they didnt tweak as it is supposed to be.
INTEL: "AMD is in our rearview mirror."
GN STEVE: "That's because you're going the wrong way!"
INTEL: "Thanks Steve!"
baahahahahaa epic quote. AMD about to see a HUGE spike in desktop cpu sales
Yea all trust for Intel, whatever was left, is gone at this point
rofl, omg. Made my day
Right now I'm using an r9 7900 on asrock b650 pro rs downclocked 200 mhz or the system crashes. Although I used the cpu with default bios options, it crashed more than my unstable i9 system. I changed ram (which is in the qvl by the way), psu, gpu, even nvme and their slots but unless I activate PBO and downclock negative 200 mhz all games crashed, some under 1 minute. And this is the second cpu, first one was faulty from beginning, its iGpu wouldn't give output. I had to wait 1 month till its replaced, and I am ashamed to send the second one to RMA again. My experience with AMD didn't differ from intel...
In my view, both companies need to create a pro level cpu which has all variables standardized with motherboard vendors ASAP. Speed may be a bit lower but they should sell it with a stability guarantee. They could even recommend a stability guaranteed MB and ram. For the last 2 years I'm tired of trying to fix their sgit.
lmao
On a platform that's becoming mostly sponsored content that prints out ad revenue, it makes channels like GN stand out. It's so refreshing to hear someone on the platform actually value integrity over sponsorship dollars. Please don't ever change
Actually this channel is pathetic how much it milks the same topic every single day
GN is a weird one, "word journalism" is excellent, testing procedures are great, but then we come to exact numbers and reviews and... noise normalized tests are set up at an unusably noisy level and extremely bad products are recommended all the time merely because they are slightly less bad than other similar ones
if a graphic card requires external power it's junk
if a chip runs way above the optimal clocks it's junk
if you can hear your PC it's broken
simple rules that should be respected in these reviews
Yes. Buy from them! Support the cause!
@@SobeR999 as time passes information changes so they make another video with the update information. It also keeps companies feet to the fire to fix their problems instead of just hoping it goes away
@@Z4KIUS They provide enough data to make your own decisions
This makes so much sense. I struggled with a PC that just kept crashing while playing just mildly graphically demanding games. All my troubleshooting kept coming back to two things: 1) power issues and 2) unknown GPU issues. Replaced both GPU and PSU. Continued having problems. Countless hours spent troubleshooting and even money taking it to a professional tech repair shop and still no fixes. Finally just rebuilt everything, new mobo, AMD CPU. Haven’t had ANY issues since getting rid of the Intel CPU.
2 words.....CLASS ACTION.
This isn´t silicon lottery anymore, this is the silicon hunger games.
lol
May the odds be forever in your favor.
@@manitobasky Morgan Freeman: "they weren't..."
Silicon lottery is not that much of thing as people think. People have same cpu but so many different combinations like psu, mobo and gpu that can affect a lot on cpu longevity undervolting and overclocking. Even the GPU can affect the voltage undervolting stability on CPU, I personally experienced it with different graphics on the same processor.
@@FoxSky-md1ul Not too sure about that one tbh. I had to swap quite a few ones on my simulation rig over the years. Even when I only swapped the CPU with another one of the same type, without touching other hardware, the results on how far you could push the new one were wildly different sometimes.
The sample size here is still rather low though.
So Intel just wanted to sit this one out, in the hopes that problem shows up after the guarantee period.
Idk about the US, most developed world mandates UNLIMITED warranties for manufacturing defects until reasonably expected life. A CPU can be reasonably expected to run for at least a decade
@@dotjaz That's true but in some regions like Germany after 12 months (or even earlier) the burden of proof reverses and the customer has to prove that the defect existed from the start, and good luck with that if the manufacturer is uncooperative like Intel. The real solution would simply be a recall and I hope Intel receives a really painful fine for not issuing one.
@@SaHaRaSquad doubt they get any trouble. To big to fail for the US, like Boeing where people died.
@@ploed Yeah the only chance would be a fine issued by the EU, though I don't think enough people are affected for them to care.
@@dotjaz That's not true. I know of no country that gives unlimited warranties. Normally it's around two years, not more. There are some companies which give their customers longer warranties up to lifetime, but that's their own business, not something that contries control.
"The oxidation issue is probably resolved"... For large corporations who got replacement CPU's maybe, but for every end user in the world who didn't even know there was an issue it absolutely isn't resolved because you can't resolve an issue if you are never been made aware there was an issue in the first place.
So this means there are still potentially millions of CPU's out there that have the oxidation issue and none of the people who have the defective CPU even know it's defective because Intel refuses to tell anyone, including media outlets or board partners, which CPU's are affected.
All they say is "13th gen CPU's made in 2023" have this defect, so that could be any and all produced in that entire year or just within a small window and since Intel refuses to tell anyone which range of serial numbers is affected, we must now assume ALL of the CPU's made in the entire year 2023 are defective.
In a class-action, this bullshit statement will bite them in the ass because any ambulance-chasing lawyer will latch on to it like a trash panda on acid.
And how do you inform everyone about issue even if you wanted to? 99% of people does not view any media about tech, shops absolutely will not care or should contact their clients. Intel can just do it like car manufacturers, put small note on their site, well hidden ofc that there is recoll. In real world people will just buy new cpu and never think about it.
@@wykydytron Way to miss the point there.
They don't have to inform everyone individually, they just have to say something, anything, by any means and they didn't even do that.
There was never any statement or press release or communication towards SI's or board partners or OEM manufacturers or retailers. NOTHING.
The reason for this is simple... it's so Intel could claim they didn't know there was a problem so they can reject RMA's which is what they did and what this video and JayzTwoCents video shows.
yup. all those people who bought prebuilds with no detailed knowledge, swearing but living with intermitant crashes who are gonna end up with an expensive unrefundable paperweight in a year.
Risks of being an early adoptor, of course, it sucks. But you literally are putting yourself at a disadvantage if you don't hold off buying new hardware at release.
According to Germany law, every single buyer of 13th and 14th gen CPUs that are potentially affected are eligible to a full refund by intel for CPUs and Mainboards because they were persuaded into buying the product under the impression of buying a lasting and actually normally working product, so do retailers have a similar claim against intel. And if intel does not provide info on how to determine if a certain product is or might be affected, all count as affected
Most sane countries have laws like this. In NZ it's based on the price and type of product and how long you'd reasonably expect it to last before it breaks. Consumer guarantee. It should be a universal concept worldwide tbh.
You nailed the comment about "only a small number". That's a completely useless statement. Small in regards to what? For me, small would be that they dredged up 7 or 8. Small for them could be 10s of millions.
Only a small number of the popular triple-bun fast food burgers were lethal...
Microsoft used the same wording when they got hacked in January
Small in relation to instability issues that are reported lately.
Some breakdowns show you they 14th series isn't significantly less affected.
Apple's damage control moment.
@@tteqhu 14th gen seems to have even the locked CPUs affected. My 14500 lasted for days, when it started unaliving itself (high count of random reboots when doing ANYTHING, i´ve even seen green artifacting on the monitor plugged into iGPU. I managed to return it within returning period, which saved me a lot of headaches. Went with 13600K, assuming it will be okay. Random reboots continued 😢, although severely limited (only in idle & low loads) and i seem to have it solved now (replacing the surge protector did the trick - the old one was 7y/o and probably had issues powering up electronics with low voltages - i´ve also had occasional light fluctuations and some monitor issues).
Eh that's just damage control 101. Claim minimal impact.
Numbers, or subjective language that should be ignored.
Last week's outage was isolated to a subset of users and systems, a fix was rapidly identified and deployed by expert teams at Crowdstrike. We apologise to the limited number of people who suffered minor inconvenience as a result.
Notice how they say "addressed" and not fixed.
Good eye.
Feels to me like that's a reach. It is heavily implied that 14th gen is not affected by the oxidation... and if it is, seems like a blatant lie that's sue worthy
@@trivolous28Wendell's in depth breakdown, and Steve's summary in the prior news video that cited that breakdown, both mentioned that the 14900K and derivatives all suffer the same issues even when run on conservative power targets with datacentre grade motherboards
Notice how they say "which is A key element", so that means, is not only that what's causing issues
@@trivolous28 quite in the beginning they say the oxidation issue was just happening in 2023 and is now not happening anymore, and is not the reason.
Since 13 and 14 gen CPUs are dying (see Wendells video aswell), this either was a blatant lie, or is actually the truth, and they have some even deeper lying issues.
Thanks [Tech outlet presenter], and everybody at [Tech outlet]
You're welcome, [tech viewer]!
@@GamersNexus I LOVE THIS [tech outlet] comment :D
All praise [tech Jesus] for constantly having our backs
[redacted] 😂
Intel shipped the cpu's at medium rare, and one's who bought it are cooking it to well done.
Considering Intel normally has a 3 year warranty on their boxed CPUs and Raptor lake came out in 2022. A blanket warranty extension on all CPUs to a fixed date 2 years from when the microcode update comes out should be the bare minimum because all Raptorlake CPUs sold have been degraded to some extent. Extending the warranty in that way ensures that Raptor lake CPU owners who have had their CPU for more than a year or two and haven't yet seen instability don't get screwed by damage Intel did to their CPU.
this is what needs to happen at minimum
Keep dreaming brother...
This sounds more than fair.
This year's disappointment build is going to be top shelf
KF probably ^^
requirements: having it run (optional)
More coming buddy
13700T 35W that fails will be EPIC
13700T
Someone better start that class action so I can get my $12 settlement.
$12 a person would be the largest CAL payout I've ever seen! "Too bad" I went Ryzen for my current build.
"Someone better start that class action so I can get my $12 settlement" - $12 settlement? Someone's optimistic.
This is basically true... years ago I got a $15 settlement for my IBM Deskstar HD's ..... YAY?
We got like $30 per drive for the WD red settlement a couple years back, but of course if we had known about the issue ahead of time we wouldn't have spent an extra $1k on drives to replace the faulty ones, which were also bad.
@@IAmPattycakes "We got like $30 per drive for the WD red settlement a couple years back" - Have I been lucky with my drives? I never heard of any WD red lawsuit, and I own about 13 8TB Red drives, purchased in 2017 and 2018.
1980: my parents buy a microwave that lasts 23 years before they finally replaced it, and it still worked when they got rid of it.
2024: as the years go buy products are made worse and worse.
He used the "thanks Steve"! Thanks Steve!😁
I regretted not using it in the last one!
@@GamersNexusI'm glad to see my advice helped lol
@@GamersNexus back to you steve
@@GamersNexus i love steve
@@GamersNexusthanks Steve!
Attorney Vincent: "Ah shit, here we go again..."
If there's ever a class action filed over this one, he will definitely be on the show to help explain it!
all you had to do was follow the damn train, CJ!
aka: all you had to do was not crank the throttle up to 150%, intel!
@@bernds6587 If intel didn't keep increasing clock speed and voltages to keep it running, their CPUs would not catch up to more efficient architecture of Ryzen. Intel use to make fun of AMD for using "glued CPU chips" and they literally are doing just that.
Intel is AMD best hype man. WWE style.
Class action lawsuits are garbage and you lose your rights after it. Companies and lawyers love them since it's a win win.
I need to be honest, I was building my last workstation a few days ago.
Something big that is supposed to follow me for a few years to come in my work (3D artist).
I was finally done with all the part buying stuff, going to bed with the sense of completed duty, and I just wanted to watch a tiny TH-cam video before bed.
That video was yours with Wendell uncovering the shitstorm that was ongoing for 13th and 14th gen.
Without that video I would have not canceled my CPU order, and I would have not saved my rig and my work for tedious problems, client service calls, and hair pulled of my head, thank you, the timing was miraculous.
GN has saved me from "unwise purchase decision" multiple times in the past. On top of my head, the Enermax cooler saga and the NZXT riser thingy..
my story is not that wild, but he was reporting on an upcoming new AMD platform called AM5, so I was waiting for it to drop. Now with that 7950X I might even be able to upgrade to the last AM5 CPU gen in like.. 2027? 2029?
Cancel motherboard too you're buying a dead end platform where you either buy an Intel or throw it out.
Are Steves videos ever tiny?
Tech Jesus strikes again
Thanks Steve
Man, I have been pulling my hair out for almost a year now chasing instability issues. I RMA'd my RAM, RMA'd motherboard, reinstalled OS, changed SSD's. Turns out issue is my 13700k. I bought it right when it came out so I'm sure mine is affected by the manufacturing issue. So frustrating!!!
Curious, why didn't you switch to AMD years ago?
@@Shijaru64 Mostely because I've alwyas had Intel+nVidia builds and because Intel has much higher market share, they tend to be more stable and supported by software. Devs will pay more attention to making sure their stuff works with Intel first before AMD because it's the majority of their customers.
Same reason I switched to iPhone from android. Because apps tend to work better on iPhone not due to anything inherent with he iPhone, just that devs pay more attention to it because it's higher market share (in the US at least).
Dodged a bullet by not saving enough to upgrade. My laziness saved me again.
most of the time, we don't really need an upgrade.
or you can buy AMD instead
Dodged a bullet by waiting to upgrade to the highest end sku! 14900KS!
@@intriguingfacts5434 People keep touting the 7800X3D, and sure it's a great CPU, but most people would be better off getting a 7600 or 7600X, as that's likely still going to give more than enough performance, and it costs about half as much as the 7800X3D... And you can always upgrade later on if/when you need more performance.
@@Kanev00 Ah yes, the greatest CPU to ever exist: the 14900KYS
LTT on his WAN show said something like the "TH-camrs niche" was a tiny group and doesn't "impact" intel that much.
They forgot, that most DevOps and infrastructure nerds who works for the service providers and big techs are the main public of these so-called "TH-camrs niche" which makes the tech influencers a pretty respected crowd.
Linus has really been just making worse and worse takes of late hasn't he?
It's also a niche group that lets sexual harassment and assault happen at their company without knowing about it.
Unsubbed from all their spam years ago, best thing I've ever done. Their "knowledge" of things is just about enough to cause damage and hurt themselves...
TBF to LTT, Intel still maintains overwhelming PC market share, dominating prebuilts and laptops. He's right that regular joes and boomers don't know or care about this issue, they only want Intel Inside.
Well that is true. Have you seen how dominant amd gpus are in youtube comments and influencers? But in the real world, no one buys them. Market share will tell you that. Yes youtube is a niche.
I had to get an RMA on a 14900k, and it took a month. It involved Intel telling me to change multiple BIOS settings while asking me to contact the motherboard manufacturer if I didn't know how to make them myself and repeated BIOS flash updates from ASUS. At one point, they also suggested I had an Engineering Sample (it wasn't) and requested I take a photo of the CPU scan code which meant time spent removing the AOI cooler, taking the photo, etc. for no reason. The problem, from my experience, was that it felt Intel wanted me to jump through hoops rather than accept that I simply wanted a plug and play (not overclocked, just wanted the ability to do so someday) experience that was made impossible by the CPU. The return was completed, and I replaced it with a 14900 (non-K series) that works flawlessly. I can accept a failure in design that wasn't exposed until enough were sold to create a bigger sample size, but the customer service process to have that issue addressed and get a refund or replacement was inexcusable.
@richardw412 Like a lot of Fortune 100 companies the even have external help developing their RMA processes. The goal is to make you go away, give up, etc. Congratulations on making it through the gauntlet.
The issue is turbo mode, essentially. So you are stuck with a stable chip that is now vastly slower than an AMD equivalent for the price.
If you manually hard limit the voltage and wattage into the CPU, it's effectively a 2.7 or 3.0 Ghz processor with no turbo mode at all, ever. And that's the ONLY fix - never running it to where it gets past 80C. Such a mess. lol.
And that should be grounds for a lawsuit IMO. One that sees the manufacturer paying you for your time, at the rate of an expert trouble-shooter, plus legal costs and a standard punitive fine based on the price of the product. Your time being wasted by their failure should be something they have to compensate you for.
This is the problem with modern customer support services, they make you jump through many hoops in the hope that you'll just go away. That's their actual business model. I've had this happen before and even when I present evidence that the product is blatantly obviously not working as intended, they give the "well that's normal behavior" excuse and slam the door in your face. Shoutouts to Ifi Audio and Noctua for doing that. They will only start making an effort to fix things if they can see there would be significant legal consequences if they don't.
I don't know if you are in the US, but is it actually necessary to do the RMA via Intel? I am in Europe and had to replace an 14900k, but I sent it back to my online retailer. They told me beforehand that they would then have to send it back to Intel and wait for Intel testing it and see if it was "my fault" before sending me a new one. But how it actually went was that they received my RMA and sent me a new one the same day. Don't know for what reason. I mean I did write them that it wasn't my fault and that all other components are fine. So, anyway, sorry you lost all this time. Maybe next time (hope there wont be) try it with your retailer!?
from what I've been told, the flaw that causes oxidation issues are affected in almost every single CPU, so if you have a 13th or 14th gen CPU from intel, it's just a matter of time
@@aliaksei.kartashou lucky me I got a 12th gen, so I might have a little bias, I heard of some gaming company gathering relevant stats about this issue
13 unlucky for some. The Chinese wouldn't have released a 13th gen lol
@@gungagalunga9040 14 you mean
Todays News:
Intel says the oxidation and voltage issues were not true, but instead the problems were caused by oxidation and voltage issues.
Tonight at 10:
How breathing could be damaging your brain.
Back to you, Steve!
Fighting the frizzies at 11
bamboozled
My brain oxidated while reading this.
inhale
I'm not wearing pants. Film at 11. 😂
@@myfavoriteviewer306
Hi-Five!!!
Glad to see someone who gets the reference
🤣🤣🤣🤣
I'm so tired of companies making massive mistakes, only to brush it under the rug for as long as possible and have their customers face the consequences instead. They're so worried about their brand's image that they actively soil it.
As long as there is no penalty for it it will keep happening
Stock price*
Capitalism goes brrrrrrr.
They don't give a shit about consumers, only about share holders.
@@Aaron-zl5gq Easy to make mistake: The penalty should be shrinking market share -- but we all know people will just keep on buying intels.
@@egnappahz You cant loose market share in a duopoly of an essential product.
They've known of these problems for a long time now. I don't believe they've suddenly just managed to "fix" the microcode right after the problems get made public.
especially when they say, the manufacturing issue is not the reason, and it was fixed in 2023 already. certainly feels weird.
Intel did not learn from their past mistake on the pentium FDIV bug. They just want to conceal it, hoping no one knows...
The microcode probably affects performance so they didn't provide it because everyone would notice instantly.
@@zrider100zThey most likely dont want to update until AMD Launch is through and Every Reviewer says „Intel isnt that bad compared to AMD but we cant really recommend Intel right now since they also may be loosing some Performance when the Update will be pushed on 13th and 14th Gen CPUs“. Something like that. Anyways. Point is… Intel WILL have maximum Performance still for the Reviews.
I could now put on my Thin Foil Hat and say they always planned this to happen but didnt think it would happen so fast but…
I mean who knows. Right. We dont know for sure. But they are very dodgy for a very long Time.
@@flimermithrandir at the other hand, the 9000 Ryzens will be tested against potential defective and failing intel CPUs, so the recommendation will surely fall towards AMD. (my guess will be, no recommendation to upgrade from 7000 to 9000 anyways)
I just bought a 13th gen for my server and feel defrauded. Intel knew about these problems a year ago but still sold the cpus and didn’t tell anyone until they were caught. They’re going to loose a lot of money in lawsuits.
I wish they do!
if i were you i would go back to a 12th gen cpu untill they can prove they fixed these problems and the 13th and 14th gen cpu's wont burn out on us.
Intel doesn't seem to realize that a full recall is less expensive than losing the trust of the entire world.
Conspiracy time: OR!!, something huge is about to happen; hence the DEI corps are purposely losing money because all the mega companies probably know economically what's about to happen.
Good point
You don't seem to realize they are a lot smarter than you.
I never trusted Intel.
So they will definitely not do that.
If u in uk check consumer law: you are able to return to your retailer up to 6 years for repairs or replacement where a manufacturing defect is identified- some retailers better at it than others but the law is on your side.
I doubt the law will protect us from INTEL.
@@scottfarley6867As he said it's a contract with the retailer not intel so that's irrelevant. Though the retailers are going to cause hell for intel if they have to accept a huge amount of returns.
@scottfarley6867
Then the logical choice would be to not buy Intel in the first place
Reason why i want to wait over for the update for BIOS, and see if it fixes issues, atm just runing CPU with Intel Turbo Boost disabled and on intel power settings. At least i can use PC with no issues, yet it's so underclocked it never exceeded 50C on 3Dmark CPU test :D
@@Beastlordius Question is if your CPU was already damaged prior to you doing such, and while it might not show now, will it in the future?
The resell price might essentially be nill, cause who knows when it bites the dust.
"Short answer: Oxidation is not related to the instability. Long answer: Oxidation is only sometimes connected to the instability."
That's not the same thing. That's not the same at all.
Its simple. Oxidation is an issue only when CPU is under voltages as designed. If the voltages are lowered(and performance), the rusty chips have no problem surviving the warranty period.
So it's not the oxidation that's the issue per se, it's the voltages they applied to achieve good benchmarks on release.
Understand now?
@@Six_Gorillion You're missing the point of what I said.
@@Six_Gorillion They are defective damaged goods. Not working as designed. That is an issue. Lowering performance significantly to maybe have around the same lifespan as expected is unacceptable. Simple as that.
If you say all the possibilities, you have to be right
@@TheHopefulOneMyBroA2C4 facts !!
Intel is really struggling - the only thing holding them up is their reliable reputation, which is now crumbling. They've been falling behind other chip manufacturers for years now, despite huge amounts of R+D funding (which is dwindling as investors are pulling out). This "voltage" and "oxidation" issue is a symptom of a horribly managed corporation.
From the first computer with an i486 processor I worked on in high school to the current Ryzen, I have never heard of a processor oxidizing. What a time to be alive.
I mean I knew very very old systems could have tin creap - where the solder could grow whiskers and short out. But that was pre-486 lol
Thank you 2 for making me feel old… 8088 kid here😂
@@teezee1000 my first computer was a vic 20 - it had 5 KB ram. I bought the 8 KB ram expansion - it loaded programs (and saved) to audio cassette.
American work ethic is slowly derailing in recent years for some reason
@@xThemadpoet Lol when I was a kid I'd play games on a dual tapedeck amstrad. We were popular amongst friends as we could dub audio tapes xDD
A 'microcode patch' just seems like they're trying to avoid an en-masse recall (ala P67 chipset USB issues)
Microcode patch is just asking for a class action lawsuit since it seems like they won’t be able to fix the stability issues without throttling the boost clocks. Hope they can get this house in order, this is a huge problem.
It also makes me think of Bay Trail coming out half-baked, where it would corrupt SD cards if the speeds got over about 25 MB/s. Did they fix it? Hell no, they put a speed limit on the SD reader.
One that probably slipped the consumer radar but not enterprise is Silvermont based processors suffering LPC, USB and SD card circuitry issue. Plus there's also Airmont's SD card and LPC degradation, which made buying Bay Trail tablets a bit of a risky affair...
both of the issues did set me back on considering the HP Pro Slate 10 EE G1 as a portable Chromecast TV, primary for streaming my PC screen while I'm on my bed. (I won't mind the outdated Android version bcus of this) so that stings
Oooh class action. Enjoy your 20 bucks.
@@MagicSwordKingthe microcode issue is that according to Intel the CPU’s microcode is bugged and requesting too much voltage. For example if the cpu needs 1.4v it might be asking for 1.6v and degrading itself in the process. This is very fixable and doesn’t need a frequency downgrade. This also explains why i5s and mobile processors are less affected because their lower clocks means they have actual voltage headroom and if the CPU wants more voltage it might not be high enough to actually be damaging. There’s probably a lawsuit coming over oxidation and RMA denial, but the microcode fix should be enough to avoid legal action (assuming that Intel isn’t lying).
I hereby call for an emergency "is intel screwed?" meeting
I love this. I'll message Gordon immediately.
@@GamersNexus +1 on that please!
I hope Wendell can make a guest appearance
@@GamersNexus This needs the full squad, Avengers Assemble time, call everyone in! :)
@@GamersNexus This all feels a bit weird. Is the claim this massive quality issue is out in the field and OEMs with these unrealistically large number of time 0 failures are in contact with Intel and they're doing nothing about it? Usually, these issues are worked out between OEMs and manufacturers WRT RMAs and issues found and this whole situation seems to be sensationalized and doesn't add up to me.
Heads up to all Intel owners trying to RMA their cpu. I have been in contact with Intel for over 2 weeks now. Today I accepted the Terms and Conditions for a replacement however my tech support on average only responds every 48-72 hours. Good luck my fellow Intel customers who got burned by Intel.
What I admire most is that Steve does not hesitate to expose serious problems with products of major companies in the computer industry very effectively.
Keep up the great work Steve! 👍
Yep, thanks to Steve we know to avoid Intel at all costs! Really saving us a headache there.
Everyone is crapping on Intel for this and has been for several days. Not like it is just these guys.
Intel didnt know what caused the problem, refuse RMA wanna keep customers money. Yep sounds like americans.
The cuts to "Thanks Steve" still crack me up.
It's a classic at this point
Never gets old
Whats the meme here? Is that an old Intel presentation?
I watched an interview der8auer had with an intel engineer, and that dude PROUDLY downplayed power draw issues stating that these cpus are meant to be pushed to the limit
Yeah, I'd love to see a follow-up interview with him in light of mass failures of their CPUs.
I'm not sure they are engineers..
@@932viper engineers can be stupid too, you know. Colleges shove everyone through now if you're of a certain group, lets just say.
@@932viper yeah, an absolute joke.
but that's the problem- they are being pushed over the limit
The question I have is, if Intel knew there were oxidation problems in manufacturing, why did they still ship those affected CPUs out to the market? The loss in dies (or even full packages) is tiny compared the impact to their reputation for a few dollars on their quarterly report.
Under poorly regulated capitalism, short term profit matters more than long-term
@@gintozlato1880
I wish you weren't right. But you are.
They can't test every CPU made. They batch test a certain amount. This is not new and industry standard.
They are out of money and simply cannot afford a large recall.
@@vamwolf LOL, they admit they knew about the problem... This has to be the dumbest response to my question possible.
"Intel just admitted oxidation issues..."
Well, that started off positive.
Totally misleading and click bait. Not worhty of this channel. Just 13th gen and just a tiny little fracton. Steve doesnt need click bait openers like this.
@@bestonyoutube Yeesh, you buy into Intel stock when it was high or what?
Intel 13/14th gen dying at a 30% rate...
@bestonyoutube : "Oh, must have been the wind"
Not sure if it was intentional, but I read this comment as a joke about the chemistry of oxidation 😂
@@brandyballoon your sense of humour might have corroded sir. 😁
So, they’re claiming that all 14th gen instability issues are not caused by a manufacturing defect that resulted in oxidation, since they’re saying the manufacturing problem was solved with 13th gen. Interesting…
Maybe they're being generous with causality - the 13th gen chips that were rebadged Alder Lake parts weren't susceptible so it was kind of solved in that sense lol
@@bosstowndynamics5488 No, 13th gen is just as affected by instability as 14th gen is, which both happen to be made using the exact same manufacturing process, on the same lines and tooling probably.
@@celeriumlerium8266 Re-read what I said, I'm not defending Intel here, I'm taking a jab at the circuitous market speak by pointing out that the only way it could actually make sense is if they point at Alder Lake and say "see, we fixed it" (the issue at hand affects 13th and 14th gen parts but only the high end stuff, both the low end Meteor and Raptor Lake parts, and of course the older stuff using Alder Lake that just got relabeled and sold as 13th gen, aren't affected as far as we know)
yes, either they lie, or it's really "just" the voltage... or some deeper lying problem.
@@bernds6587I mean Raptor lake is pushed so hard out of the gate that if excessive voltage is applied the CPU would be cooking itself. 14th gen is pushed even harder and would cook itself faster with improperly high voltages. It also explains why lower tier CPUs like the x600K are mostly unaffected because their much lower clocks need lower voltages and improperly high voltages may not be high enough to be damaging. I am inclined to believe that most instability on slower Raptor Lake CPUs like the i5 and mobile chips is oxidation (should be fixed) and the higher tier CPUs are just degrading themselves with voltage requests being improperly high. Intels response is pretty bad, but the problem seems very fixable now as they are promising a microcode update in August.
Intel - It's dead inside
Intel - I$reali backdoor inside.
Intel - Oxidation inside
Amd -( it works) would be enough 😂😂
AMD - it never was alive in first place
AMD - for long winter nights you no longer need heating
AMD - what's compatibility?
AMD - 2x price of Intel outside of USA because f u.
And so on.
@@tao4124 I'll take an Israeli backdoor over a Chinese one any day :) And, tbh, issues in a CPU obviously prevent the backdoor from being used, so...
I just want to point out their current CEO was their CTO (2001-2012) during their long standing anti-trust issues (2002-2006).. given that, I find this behavior unsurprisingly on-brand.
Intel lying about whatever is "on brand" - they have a whole department dedicated to nothing else (marketing).
Sad to see that as of late AMD is (once again) going down the same drain.
@@danieloberhofer9035 What's AMD done? Genuine question.
@@dudebroguymate they did Ryzen
@@erkinalp Obviously, but that's not what I was asking.
@@dudebroguymate Apart from the very misleading RDNA3 launch graphs - these you could attribute to them genuinely believing in being able to still pull it off somehow - their marketing as of late has devolved into some very dodgy claims. The worst being:
- For the latest AM4 launch (R7 5800XT), they claimed to match the i7 13700K in some gaming benchmarks. Testing was done with an old, very low end RDNA2 card (6600 ot 6600XT, can't remember which one exactly), carefully avoiding a CPU limit. That's just bullshit, they could've also claimed that an R5 3600 matches the Intel part (or the "new" 5800XT for that matter). That's not lying per se, because that's certainly what happens when you pair these CPUs with an RX 6600 - but it's still dodgy af because it says nothing at all.
- During their recent tech day they also made some weird comparisons - R7 9700X vs. Intel i7 vs. R7 5800X3D, across various slides, and no matter how you sum up the results, they just don't match up.
Worst of all - I don't get *why* they are doing this. It's utterly unneccessary to weirdly hype up their CPUs that way. The 5800XT is what it is - a bleed off for Zen3 CCDs they still have to make for Epyc Milan. It'll sell anyway, no matter how. And it's certainly not meant to compete with RL i7s, so why the dodgy marketing. Just releasing it and taking the good press for their stellar platform longevity would've been more than enough.
And as to Zen5? It already looks to turn out pretty good all by itself. Fast, efficient, cost effective and on a well established, future proof platform. With little to no competition by its launch. So why wiggle around like that? Just present the graphs, show the R9 9950X completely destroy the i9 and be done with it. Hammer home the TDP reductions all day, tbh. Satisfy the enthusiasts with Curve Shaper (awesome!) and on the fly memory tuning (stellar!). It all looks exciting enough. Maybe tease X3D for late Q3 / early Q4. There's no need to overhype anything and open yourself up to criticism.
If Intel is struggling so hard with logic I'm not surprised that they cannot make CPUs properly.
Intel and making some bs shenanigans when is about taking accountability and benefiting the consumer is old news.
I just noticed the list of 13 th gen cpus affected by oxidation is literally EVERY raptor lake cpu (since everything else was rebadged alderlake)
Fancy way to say they let their processors rust and then sold them.
Major companies have been getting away with these wishy-washy statements and actions for too long. Thanks to people like you(and your team), we're finally starting to get some pushback on the dystopian doublespeak. Bull shit is bull shit and should be pointed out to be such in no uncertain terms.
Userbenchmark has to do some pretty intense 3d chess to explain this.
They'll just leave up all the OG runs before the CPU had to be nerfed and start blocking new scores.
Userbench aside, the performance is there... for a half a year.
Amds team of advanced sabatuers have illegally infiltrated the only good cpu maker
@konev13thebeast What??
Userbenchmark rigs its benchmarks, we have known this for well over 4 years
I just want to say "Thank you" for keeping up the pressure on these big companies. If you didn't "threaten" the failure analysis lab, who knows, Intel might have never admitted to the oxidation thing.
😂 Intel is a 54 billion Revenue company if you think for one second this little TH-camr has any pull over what they do you're naive
@@Savethepandabears So if no one complained, researched and broke the news, Intel would have still come forward admitting their fault, admitting that they had known about manufacturing problems in 2023 and thus KNOWINGLY SOLD DEFECTIVE PRODUCTS to customers? Just how much do you love multibillion companies to be so naive?
@Savethepandabears it's not just this channel, but plenty of others are looking to GN because who else is spending 5 figures to do Intel's homework for our benefit?
Intel pushed voltages so hard trying to keep up with AMD that they degraded the silicon.
No, the voltage seems to be a compensation to their own manufacturing flaw. The chips probably wouldn't have worked well while slow either.
i think im going back to AMD
my last intel cpu, thats for sure. bought 13900k last year and got about 20-25 different bsod, swapped every component twice(!) without suspecting the cpu. with no other component left, i contacted intel and they replaced it, not having any issues since. still might have a timebomb in my rig now... nice feeling!
Agreed. Can't see me buying intel ever again, the trust is gone. my 13900k still works, but I need it for very, very hard workloads and now I just feel like sitting on a timebomb. Not using it to preserve it? Wasted money. Using it and potentially destroying it? Also wasted money. End consumers are left with a lose lose type of situation. And i wanted to get a 7000 series setup at first! Damn, why didn't I.
@@nikethunner2732 sell it and go to AMD before the plebs figure this out and the value plummets
@@nikethunner2732the trust is gone until AMD screw up, then you'll be back to square one.
@@nikethunner2732
You have to limit and lock the frequency. Based on Framechaser and Buildzoid videos, the issue is that the 2 best cores request voltage above 1.5 V under light workload to boost to the max frequency and that degrades the CPU. The only solution is to lock a lower frequency, limiting power or temperature cannot help.
@@nikethunner2732because we thought "Intel is better" 😔
Maybe you could add a modifier to the performance charts to reflect 0% performance for 50% of Intel CPUs
50% of the time they work every time, except for when they don't.
Just don't put any load on it, like running 3 copies of notepad at once, or it'll crash.
@@solenoidnull9542 i can't even go hard in chrome without crashes and oh snaps
Microcode updates on this issue are band-aids. They must be buying time to stop the bleeding of RMA’s, until they address the inherit issue of electromigration from a foundational level
Perhaps with Bartlett Ridge?
I always knew that having 1.72 volts as maximum core voltage in their documentations is insane. Even 1.5 volts is insane.
So is 1.21 gigawatts
frankly, my 14700KF does peak out just under 1.5 V... while idling around on the desktop...this is really sh...
I wonder if without the reporting from GN Intel would have been public about their oxidation issues, especially considering how they made their statements.
I think there was enough other coverage and pressure they would have eventually, but with Steve and GN sounding the alarm I suspect they stepped up the timeline significantly.
Intel woulda probably tried burying it. Aside that, I have said it here before, what Steve and GN's team as a whole does could well be dangerous work. Some of these companies get really, really nasty and outright hostile when people expose them. This is especially true for mega corps like Intel. The fact that Steve and team puts their neck out there to help us regular folk is so worthy of admiration. Takes some serious bravery.
I think to be honest here.. Level1Tech did drop the bomb here not GN this time.
I doubt it. The words “failure analysis lab” coming from Steve though… they knew they’d be caught.
Maybe that’s cynical, but it’s not my fault I’ve been trained to expect the worst, most dishonest, and corrupt decision making from corporations.
Answer is always no. Companies as big as Intel don't cave to anything other than financial pressure, or the threat of it.
Class action incoming for Intel
Attorney Vincent: "Ah shit, here we go again..."
Nah, they don't want to cost MORE money.
Class action lawsuits always make money for PR Firms and lawyers…
I have been included in 2 1. Ms fixing price issues spent $625 got $68
2. Ford DCT transmission $ 84 credit… but I still might have to get transmission replaced at some point in time!
So my take on this is that Intel saw that they had a good thing with the 12th gen, and instead of investing in new designs to keep up with AMD they decided to just keep dialing up the 12th gen architecture by slapping in more cores and pumping voltage and clock speeds for those sweet benchmarks. It's unsurprising that the 13th/14th gen chips are failing, the 12th gens were already pushing it.
The whole thing is like a car maker introducing a turbo'd sportscar that's a great success, and then keeping up with the competition by simply adding more boost every year. And then when engines start blowing up they just say oops sorry, we can't help you if your engine is dead but if not come in for recall service. And then they "fix" the cars by removing the turbos, and then say "all better now the engines should run fine."
But people paid extra for the turbo, right? So where's the compensation?
15th gen should be actually good...
Let's hope
can you buy 12th gen laptops now?
@@ShlomoWalfish I don't know that mobile chipsets are even affected by this, I've seen conflicting reports on that.
Laptops with 12th gen chipsets are generally discontinued at this point, but there are plenty of refurbs and you can probably find new-in-box if you look hard.
@@Xnoob545 Yeah especially since AMD isn't doing so great either, and I really don't want one gaming CPU maker to dominate and jack prices up like has happened with Nvidia GPUs.
AMD and Lisa Su is probably impressed that Intel has done the best marketing campaign in history, for a new launch of AMD cpus....
Imagine if snapdragon had actually lived up to the hype
until intel releases theirs which uses tsmc n3b, which means they have the nm advantage compared to zen 5 which uses tsmc n4, which is just enhanced n5
oh, and amd themselves also admitted that their regular zen 5 chips aren't even worth it compared to a 7800x3d, which means that regular zen 5 is DOA
i'm not defending them but both fanbases are a disease to both companies
@@rarinth atleast their cpu doesnt break
@@rarinth at this point I am assuming the pricing for the new Zen5 chips are going to cost a ton with how the competition is behaving.
@@rmo9808 It is doing just fine.
Great reporting on this issue. This may be worse than the math bug Intel had in their Pentium 1 chips back in 1994. Yes, I'm old.
I feel this comment in my knees, or maybe it's going to rain.
I think the math bug had very little real world impact though, unless you were running very specific calculations with certain optimizations. It certainly didn't result in broken CPUs that have to be replaced.
@@gorak9000 I agree, a useless CPU is definitely worse. They did do a recall on the Pentiums, I'm just not sure of the financial impact it had on Intel, or if this will end up being more expensive for the company.
Even worse if you tell me, if I remember right a microcode change fixed the issue 100%… it rarely caused instability issues..
And back then might have been 3 SKUs… now it’s at least 9….
@@gorak9000 very specific calculations like... Division
Intel ended up replacing all the defective CPUs, unless you just couldn't be bothered.
Thank you for your integrity
Thank you so much for the direct support!
GN doing Gods work again!
99% sure Intel discovered the oxidation issue and said to ship the CPUs anywas as they didn't want to spend the $ to fix or take the Stock price hit, in hopes that only a very few CPUs would have any amount of issues from it and if they did it'd be well out of the warranty.
Bingo...
Exactly, screw clients, shareholders first…
Of course they did, that's how Corporate America rolls. We don't do anything that doesn't serve the banks or stock markets. Quality products made by workers who are financially secure with a decent work/life balance? Ha haha ha ha ha ha. At least this defect doesn't result in fatalities.
GM killed 124 people due to a known defect with the ignition (en.wikipedia.org/wiki/General_Motors_ignition_switch_recalls).
Do we even have to talk about Boeing 737 Max's 346 deaths due to MCAS (en.wikipedia.org/wiki/Boeing_737_MAX_groundings)?
Nope I just know don't ask me how
@@rimurutempest2130Then nope for you.
"a small number" Louis Rossmann echoing in my head
Underclocked since day one. Water cooled. Two p-cores completely unusable to the point I had to disable them in the bios after days of mind numbing testing and bsod's. I will not be jumping through hoops to get an RMA. Intel, you lost a customer of 20 years. Great job.
Thanks, Steve.
Thanks, Steve.
Thanks, Steve
Thanks, Steve.
Thanks, Wendell.
Intel: It's fine... This is fine.
🔥🔥🔥🔥🔥🔥🔥🔥
🔥🔥This is fine🔥🔥
🔥🔥🔥🔥🔥🔥🔥🔥
They still have majority PC CPU market share, dominating laptops and prebuilts.
@@Aggrofoolthis debacle will take some time to reflect in the market share, as it will only influence future purchases.
@@user-to7ds6sc3p You're average PC user doesn't give a fuck nor knows anything about this. I'm pretty sure intel will still dominate the market in the future.
Wait; so they knew about the oxidation issue last year and are only just now admitting it may have something to do with the instability issues on their CPUs?
Standard corporate practice. Fix the problem going forward, deal with any defective products when warranty claims come in.
Yes and no. They are saying that the oxidation issue is NOT related to the instability issues, but they are also saying that the oxidation issue affected only a small number of instability-affected CPUs. Don't think about it for more than three seconds or your head might explode.
Yeahh, don't worry bout it, nothing to see here, move along folks.
Knowing there is a manufacturing issue (oxidation vias) and keepign quiet about it for over a yr while selling them....should be straight up illegal.
They knew of an anomaly. They did not know it was an issue until later. I'm sure they know of many things. They're not always aware of the significance of any of it though.
I'd like to know which country those defective CPUs were manufactured in.
@@mardus_ee what difference does it make? They were made so far removed from common reality that it doesn't matter. In clean rooms never touched by human hands.
I am sure they where hoping to get out of the warranty time before the CPUs failed at the customer.
@@1pcfredIt would be interesting to see if it was a specific fab that had an issue.
My 13600kf just failed last week. I had no idea this was even a thing. Weirdest thing is no BSOD, no crashes, no boot problems. Was just playing Black ops 6 and suddenly no video output and no peripherals.
"Only a small number" - yeah, same verbiage used by Apple during the butterfly keyboard fiasco, display cable tears and such. Which we all know, was looking nothing like a small number.
Intel pulling the “trust me, bro”
Never did, never will.
Also don't need to, I'm on AM4/AM5 since the R7 1800X.
Steve and GN team. We need you more than ever! Please don't let Intel get away with this! Intel fanboys are in full force!
Thought my 13600K was immune since it had been on almost 24/7 since Oct 2022 (launch) but then started seeing symptoms in the last couple months such as odd lag when switching between Tabs in chrome and sometimes tabs would hang and crash, also started seeing really long Nvidia driver update times (no crashes tho), normally would take a few minutes but was now taking 10 minutes, even tried DDU but it didnt help, went into BIOS and bumped up the CPU Lite Load setting from 10 to 11 and this solved all the problems and system in general feels faster & more responsive especially when dealing with Chrome, but of course it still means degradation if bumping up voltage solved it, I've now updated to the latest BIOS & 0x12B microcode, I was on 0x129 before which is concerning but perhaps 0x12B really will stop further degradation if Intel's claims are true. EDIT: Intel baseline profile defaults to mode 18 CPU Lite Load so my setting of 11 is technically undervolted from stock still, I'm not having to run the CPU over stock voltages to keep it stable.
Working in Unreal Engine is when mine stated to display issues (memory address issues, crashing the engine when saving) The replacement one has been good however.
good for now..
im on my second 14900KS and it's having issues again- despite this one being within intel spec from day 1.
@@Ravenx217 Jayz2cents had a short video - even on the intel "failsafe" (or however that was called) setting it pushed the voltage to over 1.6V. insane.
Better check your voltages, like with hwinfo to display the max.
@@bernds6587 yeah jay doesn't know shit- intel's fail safe uses the highest voltage to brute force stability.
i set my svid to typical until that stopped working, now it's back to auto (LLC referral) with 1.2 SA and 5.6Ghz lock on all cores.
This keeps me at 1.31 vcore but my games still crash sometimes. It's borked.
@@Ravenx217 sure, but 1.6V? that's insane.
Thank you for covering this issue, do you know if any Laptop CPU are also impacted by this issue? We have a bunch of i9-13950HX on laptop workstation and would like to know if they may be impacted
From reports I hear they are
RIP used market for 13th and 14th gen i7 and i9 😢
This is so bad I decided not to sell my 13600k. Although it has no issues but I couldn't possibly make someone else miserable.
@@mhosseinafz1869 you should have bought AMD cpu, by the way I feel bad for you hope you CPU don't die on you
@@mhosseinafz1869I am still on my 13600kf. I never asked myself why so many crashes and gpc latency issues... Fuck me!
@@Hell-__-Sign X86 sucks. I'll stick with RISC-V thanks
@@purpleduggy7680 Take the POWER pill, the last modern CPUs without hardware backdoors you can get. Until they introduce them on those too.
I bet it is worse than 50% failure rate. I suspect that is the failure rate under warranty. How many of us replace our machines just because the warranty ran out? And unlike the usual "bathtub curve" for tech failures, this one is probably going to have an exponential failure rate. This is a failure curve that looks more like the failure curve for automotive tires...
What drives me wild is they had the issue with the 13th gen processor, so they went ahead and made the same errors with its successor. That and the lack of communication, of course. I spent countless hours trying to pin down this issue so real glad we're finally talking about it.
This is where the 10nm "Intel 7" is hitting its limits since 13th gen. They have to crank the power otherwise they can't compete in single threaded workloads with AMD.
They didn't create the 14th gen from scratch, it'll be the same exact issue.
It’s because 14th gen is really a minor revision of 13th gen, not a new architecture. Calling those chips “14th gen” was a marketing decision. From a technology perspective, they’re really just additional chips in the 13th gen range.
Dark chocolate is a good source of antioxidants. Use it as thermal paste on your Intel CPU and you should be fine.
waste of good chocolate, just saying
Definitely @@TomTschritter
This has to be bait, right?
bro was caught
UNDERRATED XD
A revelation of oxidation in this nation on this station? What in tarnation?
Let’s hope the corporation can achieve amelioration.
They pushed “innovation” over stabilization - it’s a bad situation and we shouldn’t be satisfied with their explanation. At this rate they may be subject to litigation!
I take great fascination in Steve's narration of this situation.
Intels procrastination has blown my workstation halting all masterb*tion
the solution lingers in their fabrication. unless they are aiming for more altercations and worse evaluation of their stock...option?
next one must make a haiku
Back when I was running an IT department, I stopped buying Intel based servers and laptops because AMD was making better processors and they were cheaper. Most PCs I have bought since then for personal use have also been AMD. When my son wanted to build a gaming PC, he had already decided on AMD over intel so I didn't have to give him "the speach" on why AMD was better. This whole situation makes me feel even better about my decision to support AMD over Intel.
Even if it’s purely a voltage related issue, no microcode update will fix the degradation on those chips previously exposed to high voltages lol.
They're getting their strategy from car dealerships that have been protecting profits over taking responsibility for shoddy products. Just put some sawdust in there to quiet it down until the warranty expires. Then it's their problem.
You run the voltage in the opposite polarity. This turns back time and puts the magic smoke back into the traces.
Obviously, so it will limit the damage. 🤷
If they identified the problem at some point in 2023 but did not do a recall, not inform effected customers and not even acknowledge the issue in RMA claims, can't they be held accountable in court?
Sadly that's the sort of thing that you have to find out in court (at least in the US).
En la UE sí que se puede
yeah and they should, intel need to be held responsible for its many failures. personally advising everyone I know to avoid Intel at all costs.
Seems like Pat Gelsinger’s famous rear-view mirror is so heavily oxidized itself that it’s impossible to have a clear vision of the situation.
That rear-view mirror is giving him a nice view of AMD destroying his tailpipe. lmao.
@@Six_Gorillion lmao. Intel going to get freight trained as AMD Builds boost. I'm also amazed jtube allowed your handle. Iv'e also found handles the best way around censorship on heavily moderated sites. it's like they didn't think about it. I used to havea pic of uncle adi with 'in your heart you know he's right' and it lasted on here for at least a year till someone reported it lmao.
2024, the year 'they' named themselves to all the normalfogs. Enjoy it bud ;D
Intel needs to pull like a total recall on these chips right now!
a long ago intel had to recall a cpu because it did floating point maths wrong. they didn’t handle that great either.
Typical scumbag behavior from intel unfortunately, why I'm telling everyone to avoid them for the forseeable future
@@diamondlion47Yep. Intel is a scumbag of a company. Which is why I stopped buying them since 2006. My last Intel CPU is a Merom on a laptop. When I heard their illegal tactics against AMD (circa 2006), I vowed to vote with my money to never ever bought their products ever again, and since then I've owned AMD's Sempron, Phenom II X2, Phenom II X6, Ryzen 1400, Ryzen 4300U, and my next one will be a Zen 5 for sure. I'm not a fanboy. I'm just giving my money to a company that's not an asshole.
They massively upgraded their testing after that and a couple of years ago the new CEO said why are we wasting so much money on testing and stopped testing. And here we are.
@@thewhitefalcon8539 sounds like time for a new CEO.
Intel bringing out the "trust me bro" warranty
LTT liked this 👍.