First post, by Jasin Natael
- Rank
- Oldbie
What is everyone's thoughts on Intel's recent problems regarding defective chips, as well as the plethora of other issues they are facing?
I just don't this flying under the radar for long.
What is everyone's thoughts on Intel's recent problems regarding defective chips, as well as the plethora of other issues they are facing?
I just don't this flying under the radar for long.
Doesn't affect me since I don't jump on the latest and greatest (at least not for new tech). Waited 3yrs to switch from Intel to AMD (5950x) back in 2020 since stability both at the CPU and chipset level is always the foremost concern for me. For these big jumps it's best to wait and let the guinea pigs test. If you don't then expect to get burned.
AMD is delaying their ryzen release a bit so likely they are making sure their i's are dotted so they don't pull an Intel.
Yes, I doubt that the Ryzen 9000 delay is coincidental. Plus they now know they have plenty of time to validate and needn't be pressure into a premature release.
I'm still running a Ryzen 3900x on a B350 board, so I won't be affected either. But it seems like at this point, that thousands if not millions of users have been, in one way or another.
I don't see this ending happily for anyone, really.
I'm interested to see what the root cause ends up being. Apparently Intel has been shipping properly working chips for a few months now, however, they have the same microcode revision as the faulty chips which makes one wonder if it's really possible to fix the faulty chips with a microcode update like Intel has been suggesting.
Socket 3: UMC UM8498 | DX2-66 SX955 WB | 32MB FPM | GD5426 VLB | Win3.1/95
Super Socket 7: MVP3 | 600MHz K6-III+ | 256MB SDRAM | MX440 AGP | 98SE/NT4
Slot 1: 440BX | 1300MHz PIII-S SL5XL | 384MB ECC Reg | Quadro FX500 AGP | XP SP3
I mostly noticed the recent updates that came out for my ASUS Z690 board. With the new setting to use Intel's recommended power specifications instead of the juiced unlimited n0t-validated settings the motherboard manufacturer uses. Though it has not been a problem for the relatively cool 12600K that I use.
Cosmic wrote on 2024-07-25, 16:22:I'm interested to see what the root cause ends up being. Apparently Intel has been shipping properly working chips for a few months now, however, they have the same microcode revision as the faulty chips which makes one wonder if it's really possible to fix the faulty chips with a microcode update like Intel has been suggesting.
Yeah, it's two pronged. On the one hand they claim to have fixed the manufacturing defect that (supposedly) only affected 13th gen chips. But on the other hand the question is how many were sold before this was discovered and rectified? More so, they only very recently even admitted to having this problem, how many RMA's were wrongfully denied in that time frame? How could they possibly go back and rectify that mistake. They don't even really manufacture those things anymore, I'm sure that they have SOME stock lying around but hundreds? Thousands?
Then there is the question of the voltage issues with the 14th gen stuff. They claim that this is unrelated to manufacturing defect that caused the oxidation issues on the previous generation of chips.....but then again this is Intel, they also claimed that there was no problem to begin with. (Also that 14th gen was actually any different than 13th gen, but that's just me being cheeky)
Furthermore, if the voltage problem ISN'T related to manufacturing defect, and simply a bad batch of microcode...this would mean that it COULD be fixed by this miracle microcode update.
That would be great if true, personally I'm skeptical.
But even if they do do fix all the issues with this patch, it still leaves likely millions of chips that have been being repeatedly overvolted for what, a year and a half in some cases?
At what point does electromigration come into play? How can it even be reliably tested?
Don't those customers deserve some consideration? It wasn't their fault the damn things that they paid top dollar for, are frying themselves.
Extended warranty? Maybe, but again Intel can't possibly replace all of these time bombs.
That isn't to say ANYTHING about the microcode possibly throttling performance. It's fairly likely that it will.
It's a big nasty mess.
1994 Users: Your chips are broke. Intel: No way no how, everything perfectly fine. Narrator: Intel's chips were indeed broke
2000 Users: Your chips are broke. Intel: No way no how, everything perfectly fine. Narrator: Intel's chips were indeed broke
2016..18 Users: Your chips are broke. Intel: No way no how, everything perfectly fine. Narrator: Intel's chips were indeed broke
2024 Users: Your chips are broke. Intel: No way no how, everything perfectly fine. Narrator: Intel's chips were indeed broke,
Fdiv, ghz bug, spectre/meltdown and this.... lot of other bugs too f00f etc.
Unicorn herding operations are proceeding, but all the totes of hens teeth and barrels of rocking horse poop give them plenty of hiding spots.
Intel is most likely pushing their chips beyond silicon degradation, to the point that it even affects conservative workstation build.
I must be some kind of standard: the anonymous gangbanger of the 21st century.
The Serpent Rider wrote on 2024-07-26, 02:20:Intel is most likely pushing their chips beyond silicon degradation, to the point that it even affects conservative workstation build.
We actually do model this. And that's really all I'm gonna say.
No doubt about it, but the quality of silicon in mass production could vary beyond projection. So buying a high-end CPU becomes a lottery. 13600/14600 chips and lower chips are mostly unaffected it seems.
I must be some kind of standard: the anonymous gangbanger of the 21st century.
DosFreak wrote on 2024-07-25, 15:24:Doesn't affect me since I don't jump on the latest and greatest (at least not for new tech). Waited 3yrs to switch from Intel to AMD (5950x) back in 2020 since stability both at the CPU and chipset level is always the foremost concern for me. For these big jumps it's best to wait and let the guinea pigs test. If you don't then expect to get burned.
AMD is delaying their ryzen release a bit so likely they are making sure their i's are dotted so they don't pull an Intel.
More likely they dont want a repeat of their last problem with their fire starter motherboards.
It could be that the motherboard manufacturers are not following the tight voltage controls the chips now require.
On the other hand it being an American company it could be a competence problem.
Boeing 777 anyone?
Pssst... the other vendor's CPUs have bugs too. They all do. Some just get more publicity...
jmarsh wrote on 2024-07-26, 04:19:Pssst... the other vendor's CPUs have bugs too. They all do. Some just get more publicity...
Cool, but none of these result in the CPU rapidly degrading. The exploding 7800X3D processors would be a much better example.
Gigabyte GA-8I915P Duo Pro | P4 520 | GF6600 | 2GiB DDR | 256GB SSD | DR-DOS/WinXP/Arch32/OBSD
MSI MS-5169 | K6-2/350 | TNT2M64 | 384MiB | 120GB HDD | 95/NT/2k/XP/OBSD
Dell Precision M6400 | C2D T9600 | FX2700M | 16GiB | 128GB SSD | 2k/Vista/Arch/OBSD
This is Nothing New.
It been going on Since the 1990’s after the 486 CPU.
Back in 1990’s In the Server rooms we use to remove all the top covers and cool the machines with direct Air conditioning ducts.
If you were to walk into an Internet portal server room filled with hundreds of servers you could feel the heat from computer exhaust fans and the hair on your body would stand up from all the static electricity.
Its Case Swap time.
Get a Case with Good Ventilation
Get Lots of Fans
Lots of Cooling
Liquid CPU Coolers
Remove the Side Cover and place a BIG Fan in front.
Replace your Thermal paste and pads.
Keep your PC clean of Dust and Debris
Good cable management
Don’t put your PC to close to the Wall or other Obstacles. Give it room to breath.
These OEM computers cases don’t have good ventilation
These USB powered fans are an Easy fix.
All in all, something like, they haven't got their ducks in a row because their chickens aren't spherical enough to fit the physics and now the cows have come home to roost.... as predicted by the prophet Larson.
Unicorn herding operations are proceeding, but all the totes of hens teeth and barrels of rocking horse poop give them plenty of hiding spots.
jmarsh wrote on 2024-07-26, 04:19:Pssst... the other vendor's CPUs have bugs too. They all do. Some just get more publicity...
https://www.youtube.com/watch?v=OVdmK1UGzGs
You wish it was only a bug...
If you wanna check a blue ball playing retro PC games
MIDI Devices: RA-50 (modded to MT-32) SC-55
iraito wrote on 2024-07-26, 15:00:You wish it was only a bug...
I don't really have a horse in this race.
My main concern is having assembly code that I write do what it's meant to do... and after waiting what seemed like an eternity for AMD to implement AVX-512 (seriously, it took so long that Intel had stopped putting it in their CPUs by the time AMD produced one with it), they botched it so that simple register loads randomly fail.
I'm in the same boat as some of the others here, I'm unaffected since I'm always at least 2 or 3 generations behind. All of the systems in my home are AMD anyway, and all are still on B450 boards running Ryzen 2600 or 5600 processors. I've never been one to want the latest or greatest, this is true for both hardware and operating systems, I live to give them time to "mature". Simply put, to get the bugs worked out of it and any defects or vulnerabilities to be exposed. Lastly, I only need what works for me, and that rarely ever means I need to get the newest or fastest hardware out there.
I normally keep my opinions to myself on this stuff since so many people are so entrenched in their "Intel/AMD/Nvidia are GODS" mindsets that presenting an alternate view sparks rage filled debate. That being said, this is Vogons, so I'll let it slide this once. 😀
I've never been a "fanboy" of any specific manufacturer. I used what I could get my hands on, that started with Intel obviously, then expanded out to Cyrix and AMD. After the Pentium 4 era, I found myself buying more AMD products than Intel for the reasons I said above, as well as cost. My last non-AMD retail purchase of a new processor was a Pentium 4, I've been running AMD since then, so it's been a while. While I would happily buy an Intel processor if it fit my needs and budget, that just hasn't happened in a very long time, and I'm kinda glad.
I'm not going to get into any specifics, but Intel have soured my taste for them as a company with some of their ... practices ... over the years that I found to be very shady and anti-consumer. While AMD and just about every over big name in the industry are guilty of it to some degree as well, I mean, they are a business, they are not anyone's friend, they exist to make money. It seemed that Intel was doing it more often, or at least, getting caught.
So, while I looked at building non-AMD based machines, the combination of not needing what they had to offer and not liking some of the things they did over the years, conspired to keep me on AMD. These latest issues are just one more example. Will it hurt Intel? I don't think so, not in the long run, they are too large. If AMD had done something like this, it probably would have had more of an impact on them simply because they have such a smaller share of the market.
Anyway, that's my long-winded way of saying, I'm not surprised, I'm unaffected, and I'm more worried about my 90's hardware than I am Intel's problems. 😜
Retro Blog: https://theclassicgeek.blogspot.com/
Archive: https://archive.org/details/@theclassicgeek/
3D Things: https://www.thingiverse.com/classicgeek/collections
StriderTR wrote on 2024-07-26, 16:43:I'm in the same boat as some of the others here, I'm unaffected since I'm always at least 2 or 3 generations behind. All of the […]
I'm in the same boat as some of the others here, I'm unaffected since I'm always at least 2 or 3 generations behind. All of the systems in my home are AMD anyway, and all are still on B450 boards running Ryzen 2600 or 5600 processors. I've never been one to want the latest or greatest, this is true for both hardware and operating systems, I live to give them time to "mature". Simply put, to get the bugs worked out of it and any defects or vulnerabilities to be exposed. Lastly, I only need what works for me, and that rarely ever means I need to get the newest or fastest hardware out there.
I normally keep my opinions to myself on this stuff since so many people are so entrenched in their "Intel/AMD/Nvidia are GODS" mindsets that presenting an alternate view sparks rage filled debate. That being said, this is Vogons, so I'll let it slide this once. 😀
I've never been a "fanboy" of any specific manufacturer. I used what I could get my hands on, that started with Intel obviously, then expanded out to Cyrix and AMD. After the Pentium 4 era, I found myself buying more AMD products than Intel for the reasons I said above, as well as cost. My last non-AMD retail purchase of a new processor was a Pentium 4, I've been running AMD since then, so it's been a while. While I would happily buy an Intel processor if it fit my needs and budget, that just hasn't happened in a very long time, and I'm kinda glad.
I'm not going to get into any specifics, but Intel have soured my taste for them as a company with some of their ... practices ... over the years that I found to be very shady and anti-consumer. While AMD and just about every over big name in the industry are guilty of it to some degree as well, I mean, they are a business, they are not anyone's friend, they exist to make money. It seemed that Intel was doing it more often, or at least, getting caught.
So, while I looked at building non-AMD based machines, the combination of not needing what they had to offer and not liking some of the things they did over the years, conspired to keep me on AMD. These latest issues are just one more example. Will it hurt Intel? I don't think so, not in the long run, they are too large. If AMD had done something like this, it probably would have had more of an impact on them simply because they have such a smaller share of the market.
Anyway, that's my long-winded way of saying, I'm not surprised, I'm unaffected, and I'm more worried about my 90's hardware than I am Intel's problems. 😜
I can get behind this. I'm pretty loyal to Nvidia as a customer, just because I don't think any Nvidia GPU or driver release has been explicitly bad in all the time I've owned one since the Riva 128. By contrast, AMD has some high points and some low points, especially in their drivers.
Processors on the other hand, Intel and AMD seem to have traded places fairly regularly in terms of producing duds. Although this silicon degradation issue certainly seems like a new one to me. Really takes the cake. But I've switched back and forth between AMD and Intel. Stuck with Pentiums though my childhood, then with AMD for Athlon XP and Athlon 64. Went back to Intel for Core 2 Duo and later one nondescript Intel i5 or another. Then back to AMD again for a 3700X and a 5800X3D. On the one hand, if a platform is stable and the CPU performs well, it's unlikely anything will come along later and fuck it up the way GPU drivers can. On the other hand, if the platform is immature or the CPU is a dud, it can be a long hard road to redeption. Which seems to be the boat Intel is in now.
I wish them luck. As much as I love Nvidia, they need more competition. I'd hate to see AMD get all cocky and evil without Intel being all cocky and evil themselves, keeping AMD honest.
Win95/DOS 7.1 - P233 MMX (@2.5 x 100 FSB), Diamond Viper V330 AGP, SB16 CT2800
Win98 - K6-2+ 500, GF2 MX, SB AWE 64 CT4500, SBLive CT4780
Win98 - Pentium III 1000, GF2 GTS, SBLive CT4760
WinXP - Athlon 64 3200+, GF 7800 GS, Audigy 2 ZS
Bring Back the Slot-1 CPU Design.
Laying everything Down on the Motherboard ( CPU, RAM, SSD ) is taking up Valuable Real estate
And Causing Motherboards to Over Heat.
Its the Same old Problems that Plagued the computers of the 1990’s
Over heating Capacitors, Ruined PCB Motherboards and cards.
The Computer PCB Motherboard Needs a Complete Redesign.