I don't recall Wendell ever identifying his sources of information. Feel free to find out if you want to support your stance.
In this video, Level 1 Techs analyzed failure data from "thousands of systems already deployed in the data center around the W680 chipsets".
He says these are from game servers belonging to two different games companies, but he doesn't name them. So, I guess we don't know whether either is Alderon Games. However, he does go into details about how they're configured.
- [12:51] "they use SuperMicro; one of them uses Asus and the Crash rate is pretty similar between these two "
- [13:28] "fully 50% of the systems deployed for both companies with either one of these processors to within one percentage Point are experiencing the same stability issues"
- [13:37] "even disabling ecores has not fully resolved the issue"
- [13:41] "for one of these companies the error rate also seems to be going up over time on the server side"
- [15:04] He quotes one of the datacenter operators as saying: "we had really good luck with the 12900 K based systems that we had and we always had good luck with the Xeon. Something just isn't right with the 13900 K and 14900 K. We already replaced a lot of customer 13 900k systems with 14 900k systems and the issues don't seem to be fully resolved. We've been steering customers toward 7950 X systems."
- [15:29] "We talking one of the game developers about this they said I think I'm going to lose about $100,000 in Lost players from their multiplayer server crashes."
- [17:41] "the two populations of systems were a little different the one provider uses dual dim configurations and that seemed to suffer a lot. The single dim configurations seem to work a little better. The 2x 48 gig dims versus the 4x 32 gig dims opt for 2x 48 gig dims every time.
- [18:01] "the most stable configuration for testing YC cruncher 24 hours at a time on the Linux side was definitely configuring a Max multiplier of 53 and configuring the ddr5 speed to 4200. for the 4 dim configuration 5200 was fine for single dim."
- [18:24] (discusses testing setup & procedures; crash behavior)
- [18:58] "one of the first things that I did in setting up both machines for both providers was to fully update the BIOS to whatever was current as of June 25th 2024"
- [19:13] "in the end ddr5 4200 and disabling ecores were the most drastic things that positively impacted stability but mostly disabling ecores didn't have as much impact as making the memory Run Dog slow"
So, that's what I associated with Alderon Games. It actually
could be, but L1Techs didn't say which two companies provided him with access to the "thousands of game servers" with "about half of these CPUs having some kind of issue".
IMO, it'd be
more damning if neither of these were Alderon Games. In fact, if one of them were, you'd think he would have highlighted the 100% failure rate. Maybe the reason they weren't included is that they had already switched to AMD by that point?
I'm suggesting he's an angry developer who saw an opportunity and ran with it. I have no doubt they saw some failures and that it created a negative experience for them. I also have no doubt some of them likely were due to Intel's CPU problem.
First of all, the CEO of the company is the one publicizing this issue. CEOs don't usually take up causes for no good reason, because they have a friggin' company to run.
Secondly, it's not like they're anti-Intel, because otherwise they wouldn't have used Intel CPUs in their game servers to begin with. It sounds to me like, after everything they went through trying to troubleshoot the issue with Intel, he finally decided to get his ducks in a row and go public. It clearly wasn't a rash decision and I don't see what he had to gain, personally or professionally, by doing it. He was already getting a free supply of replacement CPUs from Intel, but had switched to AMD by the time he went public.
When their claimed experience doesn't line up with any of the other reporting on the situation that makes them the outlier who isn't to be taken at face value.
So, we're talking about the gaming laptops, here? What other data do you have on Raptor Lake HX laptops, other than what Intel said?
If they ran them outside of specifications in some fashion that ended up killing them that isn't really on Intel now is it?
Ah, okay. So, if a gaming laptop crashes because it was designed to max TDP, now we're back to victim blaming?
IMO, it's plausible Intel is trying to weasel their way out of the laptop CPU failures, because they're well aware of how much more expensive this is. I can believe they're trying to avoid having to underwrite machines running with too much power, if that's what ultimately makes the difference.
I still think you're throwing a red herring, by trying to use the comments about laptops to discredit the reports of server failures.
I find the data quite compelling that what really makes the difference is the workload and
not manufacturing variations, but I've also made it clear that even manufacturing variations wouldn't let Intel off the hook.
You're free to believe or not believe whatever you like, even if it's akin to a flat earth version of reality. I think you're trying to set an impractical standard of evidence, simply because you do not want to believe the truth. Believe whatever you like.