News Intel's CPU instability and crashing issues also impact mainstream 65W and higher 'non-K' models — damage is irreversible, no planned recall

[stuff and nonesense]

An aside, as consumers we need 2 (ideally more but we currently have 2) different suppliers making CPUs of similar performance. Bulldozer was a huge misstep, hot and slow. After Sandy bridge Intel stagnated… 4 cores just a little more performance each year, no need to give more, people will buy it, it isn’t bulldozer…

Ryzen appears, it isn’t as fast as Intel but is is closer, it is comparable, it is new, it can be developed. Suddenly Intel gives 6 cores and increases its count over time. Competition between the companies means we as consumers get more performance at reasonable prices.

1700x was £500 new, a 7700x can be bought today for £280. I know it’s heading for replacement but the value is there. Similar increases in value are there for intel 13700k is showing as £330.

Personally, I don’t want Intel to fail, (it won’t), I want Intel to push AMD and AMD to push Intel. I want to see value increase, prices to be stable or fall and chips to be stable and reliable. I want to see chips and motherboards properly tested and validated. I want engineers to drive sales and marketing, not the other way around. Where problems are found I want to see them recognised/acknowledged in a timely manner, investigated and resolved so that the disruption and “worry” that a customer may experience is minimised.

Too much to ask?

 

[YSCCC]

An aside, as consumers we need 2 (ideally more but we currently have 2) different suppliers making CPUs of similar performance. Bulldozer was a huge misstep, hot and slow. After Sandy bridge Intel stagnated… 4 cores just a little more performance each year, no need to give more, people will buy it, it isn’t bulldozer…

Ryzen appears, it isn’t as fast as Intel but is is closer, it is comparable, it is new, it can be developed. Suddenly Intel gives 6 cores and increases its count over time. Competition between the companies means we as consumers get more performance at reasonable prices.

1700x was £500 new, a 7700x can be bought today for £280. I know it’s heading for replacement but the value is there. Similar increases in value are there for intel 13700k is showing as £330.

Personally, I don’t want Intel to fail, (it won’t), I want Intel to push AMD and AMD to push Intel. I want to see value increase, prices to be stable or fall and chips to be stable and reliable. I want to see chips and motherboards properly tested and validated. I want engineers to drive sales and marketing, not the other way around. Where problems are found I want to see them recognised/acknowledged in a timely manner, investigated and resolved so that the disruption and “worry” that a customer may experience is minimised.

Too much to ask?

Same hope for me, best is they push the performance and lower price so we can have our choice and not forced to use one or other.

But beside performance I do hope all are based on stability, human makes mistakes, especially under stress, but owning to one's mistake early and act accordingly is the key, if they don't, they should not hope others will just pass by and let it go for free

 

[TheHerald]

Lol you getting mad for no reason, just picking the single models fitting your arguement don't make it valid, if you go to TOTL, intel price advantage goes away you know?

And how they handle any issue IS all that important, they deny any responsibility until now, where it gone so bad not only DIY geeks, but service providers come out and complain, that alone is all that matters. AMD can go bad, but they act quick and replace it, so it is solved, RPL? ppl complained and RMAed since 13900k, I even read in intel community where they have cases ppl eventually gave up on the back and forth RMA and buy a 14900k, tought it have ironed out the issue and bad luck strike again. THAT is not ok and a full recall or other compensation is needed to teach them a hard lesson, not "ar we will now honor all the RMA".

I am pretty confident MY own 14900K is fine since I undervolt it since day 1 with all those power limit set manually and used contact frame, but that is me being lucky and tweaked it from day 1, not any other man that didn't coincidentally did it.

I'm not getting mad at all.

I'm focusing on a particular aspect and I'm not getting an answer. How would you know if your x3d hasn't degraded a bit and as a result will die just outside warranty? You don't, do you?

 

[stuff and nonesense]

I'm not getting mad at all.

I'm focusing on a particular aspect and I'm not getting an answer. How would you know if your x3d hasn't degraded a bit and as a result will die just outside warranty? You don't, do you?

“Any subsequent failures I’d guess were covered by RMA and no, we don’t know how many of the early chips are still in circulation and we don’t know how much residual damage may have been done.
That we don’t hear of subsequent problems tells us that it was handled well.”

It was a problem for a short period early in the lifetime of the platform. It was rectified. It was the motherboard makers who had exceeded limits (not recommendations). The limits were enforced and the problem was solved.

 

[thestryker]

you know they just pulled all their Zen 5 back out for some QC issues?

FWIW that's just an incorrect etching on the IHS (I still think it's good they reacted immediately), and there's a big difference between that and a technical problem within the design somewhere.

and X3D burning is a very quick, direct melting issue, it either burnt out, or don't really degrade

We don't really know what residual damage may have been done to the chips that survived. With how quickly the issue was able to be identified I think it certainly minimized the potential long term issues, but there's no guarantee.

If they announce they found the issue and rectify in first half year? Fine no recall needed, but only admit at end of life cycle while denied a ton of RMA?

I think there's one of two things at play here: There's a hardware bug they're covering up by trying to find a proper way of mitigating it (I don't think we'll ever know if this is the case) or they had a hard time finding the source of the problem beyond knowing it was excess voltage. As soon as Intel started talking about default profiles it was pretty apparent it was some sort of voltage issue. I do wish at that point they'd provided worst case scenario advice as well (things like capping at 5.3Ghz/not letting the CPU exceed 1.4V etc) just so people might have taken it more seriously.

To me Intel's response has been the worst part of the whole thing as they've basically just done what one would expect to find in a publicly traded company handbook for minimizing problems. I still don't think a recall is a reasonable expectation here, but they definitely need to give concrete answers as to what they're doing for their customers. Personally speaking I'd be fine with them adding 2 years to the boxed CPU warranty, working something similar out with OEMs and providing software which could identify CPUs likely to have been running damaging levels of voltage (this should be possible based on what they've said thusfar about the problem).

 

[stuff and nonesense]

Personally speaking I'd be fine with them adding 2 years to the boxed CPU warranty, working something similar out with OEMs and providing software which could identify CPUs likely to have been running damaging levels of voltage (this should be possible based on what they've said thusfar about the problem).

In other words, doing the bare minimum. In the mean time they screwed up and sold components that were faulty, faulty over 2 (1 1/2) generations). They took customers money, when the problem became apparent they waffled, obfuscated, deflected and only when the pressure was directly applied to them did they offer the promise of a band aid.

It really is indefensible.

 

[TheHerald]

“Any subsequent failures I’d guess were covered by RMA and no, we don’t know how many of the early chips are still in circulation and we don’t know how much residual damage may have been done.
That we don’t hear of subsequent problems tells us that it was handled well.”

It was a problem for a short period early in the lifetime of the platform. It was rectified. It was the motherboard makers who had exceeded limits (not recommendations). The limits were enforced and the problem was solved.

There is not a single 7800x 3d that is out of warranty yet. His point was how do we know that until the intel fix hits on August, the cpus have already been semi degraded and will die due to that just after the warranty expires. You can make that argument for the 7800x 3d.

 

[YSCCC]

“Any subsequent failures I’d guess were covered by RMA and no, we don’t know how many of the early chips are still in circulation and we don’t know how much residual damage may have been done.
That we don’t hear of subsequent problems tells us that it was handled well.”

It was a problem for a short period early in the lifetime of the platform. It was rectified. It was the motherboard makers who had exceeded limits (not recommendations). The limits were enforced and the problem was solved.

Exactly, rectified early so the warranty still keep it guaranteed for 2+ years vs warranty nearly expired and the "fix" comes along makes all the difference.

FWIW that's just an incorrect etching on the IHS (I still think it's good they reacted immediately), and there's a big difference between that and a technical problem within the design somewhere.

We don't really know what residual damage may have been done to the chips that survived. With how quickly the issue was able to be identified I think it certainly minimized the potential long term issues, but there's no guarantee.

It apparently being the etching or it can be other issues such as contaminated pads or so, we don't know, but that is a good sign that they will not just let it go, for residual damage yes, we won't know for sure, but since the issue is addressed almost right away with the product life cycle and after that no more issues are reported vs the 2 years constant issue reporting of the RPL makes all the difference that matters.

To me Intel's response has been the worst part of the whole thing as they've basically just done what one would expect to find in a publicly traded company handbook for minimizing problems. I still don't think a recall is a reasonable expectation here, but they definitely need to give concrete answers as to what they're doing for their customers. Personally speaking I'd be fine with them adding 2 years to the boxed CPU warranty, working something similar out with OEMs and providing software which could identify CPUs likely to have been running damaging levels of voltage (this should be possible based on what they've said thusfar about the problem).

To me it's the same, maybe the problem is really difficult to identify, but it's how they kick the ball around for practically all the RPL's product life and no visible "we are sorry, here's the extra compensation we would want to comfort you", IMO, a full recall would be a great reputation salvation for them, literally spending billions to gain back their image, or as you suggest, maybe extend the warranty for 2 more years would do, and in the interim, don't deny those RMAs as frequent as they will get reported

 

[stuff and nonesense]

There is not a single 7800x 3d that is out of warranty yet. His point was how do we know that until the intel fix hits on August, the cpus have already been semi degraded and will die due to that just after the warranty expires. You can make that argument for the 7800x 3d.

“we don’t know how many of the early chips are still in circulation and we don’t know how much residual damage may have been done.”

Third time I have said that

 

[YSCCC]

“we don’t know how many of the early chips are still in circulation and we don’t know how much residual damage may have been done.”

Third time I have said that

And I do belive also IF this patch was offered 6 months after the first 13900k massive complain trend started, nobody will ask if intel will issue full recall, now it's almost 2 years and he just justify that normal RMA is adequate...