Nvidia blames Intel for GPU VRAM errors

[Kaz]

Quote

Nvidia's latest 552.12 driver patch notes reveal that the GPU manufacturer is aware of stability issues plaguing many Raptor Lake and Raptor Lake Refresh gamers. Nvidia specifies that users experiencing crashes with a 13th or 14th Gen CPU should consult Intel for their troubles.

Specifically, Nvidia's patch notes state that if 13th/14th Gen CPU owners are experiencing "...stability issues/out of video memory error messages/crash to desktop while the game is compiling shaders..." to consult two sites consisting of an Intel community page and a tutorial from Rad Game Tools on how to reduce CPU power limits to Intel's default specifications.

Tom's Hardware

 

Nvidia cards + Unreal engine compiling shaders on intel 13th/th14 gen processors with uncapped power limits is causing trouble.  That must have been really fun to troubleshoot.  I've heard of companies pointing fingers before, but man those motherboard partners really screwed up! (J/k).

 

If you're an nvidia gamer with a new intel processor you might want to enable intel's default power limit, or at least cap it.

Edited April 12 by Kaz
Link added

[Barefooter]

Wow that's not good!

[ENTERPRISE]

I recall this being covered somewhere before, I think JayzTwoCents , not related to these issues but more so the fact that mobo partners blatantly skirting Intel specifications to get ahead in benchmarks etc but introducing blatant temperature related issues. 

 

So yeah, this is not ideal. 

[Kaz]

3 hours ago, ENTERPRISE said:

I recall this being covered somewhere before, I think JayzTwoCents , not related to these issues but more so the fact that mobo partners blatantly skirting Intel specifications to get ahead in benchmarks etc but introducing blatant temperature related issues. 

 

So yeah, this is not ideal. 

Gamer's Nexus has mentioned this is a problem.  In the drive for highest/best performance, mobo partners go ham on processor settings.  That's why they stress that they run stuff to spec.  It's funny that it takes 4 companies to point it out.

[iamjanco]

Fairly recent news, perhaps relevant to the discussion; mostly applicable to higher end Intel Xeon processors:

 

LinkedIn: Intel Accused of Inflating Over 2,600 CPU Benchmark Results

 

PCWorld: Intel accused of inflating CPU benchmark results:

 

Quote

I’ll point out that both the Xeon processors and the SPEC 2017 test are some high-level hardware meant for “big iron” industrial and educational applications, and aren’t especially relevant for the consumer market we typically cover. But companies giving their chips a little extra oomph for the sake of attention-grabbing benchmarks isn’t exactly novel. Most recently, mobile chip suppliers across the industry (Qualcomm, Samsung, and MediaTek, supplying chips in almost every non-Apple phone) were accused of effectively faking Android performance results in 2020. Accusations of interference in companies’ own self-reported benchmarks, often without specific parameters and therefore unverifiable, are incredibly common.

 

Also see:

 

Source: SPEC CPU®2017 Run and Reporting Rules

                        1.4. A SPEC CPU 2017 Result is a Claim About Maturity of Performance Methods:

Quote

SPEC is aware of the importance of optimizations in producing the best performance. SPEC is also aware that it is sometimes hard to draw an exact line between legitimate optimizations that happen to benefit SPEC benchmarks, versus optimizations that exclusively target the SPEC benchmarks. However, with the list above, SPEC wants to increase awareness of implementers and end users to issues of unwanted benchmark-specific optimizations that would be incompatible with SPEC's goal of fair benchmarking.

 

The tester must describe the performance methods that are used in terms that a performance-aware user can follow, so that users can understand how the performance was obtained and can determine whether the methods may be applicable to their own applications. The tester must be able to make a credible public claim that a class of applications in the real world may benefit from these methods.

 

Phoronix: Targeted Intel oneAPI DPC++ Compiler Optimization Rules Out 2k+ SPEC CPU Submissions

 

[Kaz]

5 hours ago, iamjanco said:

Fairly recent news, perhaps relevant to the discussion; mostly applicable to higher end Intel Xeon processors:

That software engineer worked really hard.  I almost feel bad that it came out in the way that it did.  What he did was along the lines of a firmware update.  Using hardware in ways that are more applicable to the software's needs. 

 

Did they engineer their hardware to do this from the beginning?  Maybe...

[iamjanco]

39 minutes ago, Kaz said:

Did they engineer their hardware to do this from the beginning?  Maybe...

 

Maybe, but given the pursuit of corporate profits these days (read the stock market(s)), it wouldn't surprise me if that sort of thing has become the norm (fudging numbers, regardless of industry).

Edited April 14 by iamjanco

[Sir Beregond]

12th gen ok then? Wife's PC has a 3070 Ti and a 12700kf.