Become a fan of Slashdot on Facebook

 


Forgot your password?
Close
typodupeerror
×
Windows IT

Windows 11's First Update Makes AMD CPU Performance Even Worse (theverge.com) 50

Posted by msmash from the closer-look dept.
AMD warned last week that its chips are experiencing performance issues in Windows 11, and now Microsoft's first update to its new OS has reportedly made the problems worse. From a report: TechPowerUp reports that it's seeing much higher latency, which means worse performance, after the Windows 11 update went live yesterday. AMD and Microsoft found two issues with Windows 11 on Ryzen processors. Windows 11 can cause L3 cache latency to triple, slowing performance by up to 15 percent in certain games. The second issue affects AMD's preferred core technology, that shifts threads over to the fastest core on a processor. AMD says this second bug could impact performance on CPU-reliant tasks. TechPowerUp measured the L3 cache latency on its Ryzen 7 2700X at around 10ns, and Windows 11 increased this to 17ns. "This was made much worse with the October 12 'Patch Tuesday' update, driving up the latency to 31.9ns," says TechPowerUp. That's a huge jump, and the exact type of issue AMD warned about.
This discussion has been archived. No new comments can be posted.

Windows 11's First Update Makes AMD CPU Performance Even Worse More

Windows 11's First Update Makes AMD CPU Performance Even Worse

Comments Filter:

  • My Windows update rule:

    For major version updates: wait one year
    For minor version updates ("Creators" update) wait six months

  • no way this was an accident. you have to stop letting this shit slide.

    • Re:antitrust. (Score:4, Insightful)

      by omnichad ( 1198475 ) on Wednesday October 13, 2021 @11:19AM (#61887803) Homepage

      It's not a Wintel world anymore. Microsoft actually wants Windows 11 to run well on AMD. Since this affects new CPUs as well as a few years old, these bugs aren't about driving new PC sales either.

      • Microsoft actually wants Windows 11 to run well on AMD.

        Yea, then why can't they prove it? And how did I see this coming years ahead of time?

        • MS only had years to develop on AMD chips. Just years. Whereas AMD can instantly change the hardware they manufactured and sold years ago. How can you blame MS for this? :P

      • Re: (Score:2)

        by gTsiros ( 205624 )

        "wintel" refers to windows OS and intel *architecture*

        Technically, amd64 is an AMD arch, not an intel one, but also technically, it is an extension of intel's "x86", so it's still "intel" architecture. Yes, AMD made it.

        A pc with windows and via CPU is still a "wintel"

        The better (ok, ok, not "better", just weirder) question is, is a pc with windows and intel *itanium* a "wintel" pc?

    • Is there "Intel Inside" the Microsoft Windows team?

      We can't make our processors better so we'll just knee-cap AMDs :-/

  • How much did Intel have to pay MS for this? (Score:5, Interesting)

    by wildstoo ( 835450 ) on Wednesday October 13, 2021 @11:22AM (#61887813)
    n/t

  • Microsoft Releases in a nutshell. (Score:5, Insightful)

    by nightflameauto ( 6607976 ) on Wednesday October 13, 2021 @11:33AM (#61887847)

    Stop using your end users as alpha / beta testers Microsoft. A company rolling in your kind of bank can afford to pay a large testing pool of internal users and quality control folks. Maybe it's time you try that?

  • Performance issues. (Score:3)

    by Ostracus ( 1354233 ) on Wednesday October 13, 2021 @11:34AM (#61887849) Journal

    AMD warned last week that its chips are experiencing performance issues in Windows 11, and now Microsoft's first update to its new OS has reportedly made the problems worse.

    Must be a guy CPU, and a she OS.

  • Aren't most hardware manufacturers working with Microsoft for windows releases, why isn't AMD?
    • Why is your assumption that AMD is not working with Microsoft? When MS first announced Windows 11 compatibility requirements, consumers quickly noticed the Zen 1 CPUs were not on the list. When asked about that, MS was less than forthcoming not committing to they were or not compatible. MS only said they "could be" compatible. These were 4 year old CPUs so MS did not have the excuse that they were too new. Finally MS confirmed after months that 4 year old CPUs like most of Intel 7th Gen and AMD Zen 1 were n

      • The requirements, and probably the performance problems, are both about security and "security." One to keep out malware, and the other to make Hollywood and Steam happy.

        • The requirements, and probably the performance problems, are both about security and "security." One to keep out malware, and the other to make Hollywood and Steam happy.

          Why is someone else's blame other than MS? They still control the software; the hardware was known years in advance.

          • Who's saying this isn't Microsoft's fault? But it's possible that the security features they were relying on in the CPU don't run as well as expected or that they're using them wrong.

            • Who's saying this isn't Microsoft's fault?

              Well AMD has already pointed out it's 2 separate issues: The first is in Windows that is causing the L3 cache latency to be higher than it should be and the other is a scheduling issue in AMD's driver. Both issues have been fixed and will be released in the next week or so according to AMD.

            • Who's saying this isn't Microsoft's fault?

              You said: "One to keep out malware, and the other to make Hollywood and Steam happy." No. MS has control of their software.

              But it's possible that the security features they were relying on in the CPU don't run as well as expected or that they're using them wrong.

              As noted by the summary, the problem is confirmed on Ryzen 7 2700X (launch September 6, 2018) and other AMD CPUs. It seems this patch not tested before the patch went out.

              • Every single response to my comment seems to think I said "Who's saying this IS Microsoft's fault?"

                I did not say that. I'm agreeing that it is definitely them but explaining a few ways it could have happened on their end.

        • I am confused as to why you think that Steam actually cares about the efficacy of their DRM.

  • Comment removed (Score:5, Informative)

    by account_deleted ( 4530225 ) on Wednesday October 13, 2021 @12:14PM (#61887991)
    Comment removed based on user account deletion

    • The problems are already fixed in the dev channel Windows Insider OS builds. Build 22471.1000 shows normal L3 read, write, copy bandwidth and latency. The CPPC2 bug seems non-existent as well (getting normal ST scores for CBR20 and SuperPi 1.5 Mod XS). I'm running it on my 3900X without issue.

      Much ado about nothing. Switch to Win11 in 2025 if you don't like what you see right now. Who can blame you for waiting anyway?

  • It is the dress. The dress makes you look fat. It cannot possibly be you that makes you look fat.

  • If I'm going to adopt a new Windows OS, I always give it 12-18 months of patch cycle before I use it.

  • I'm a computational physicist. The code I write (in C) makes Linux computers crunch numbers, often more than one number at a time using OpenMP. But I am not an expert in operating systems or architectures; I am a numericist.

    But I've never had to do anything to address L3 cache. It just ... caches stuff, right? Am I right in understanding that the CPU figures out what data it might want again and keeps it in cache, and works this out on its own without any sort of intervention from the programmer or the OS?

    W
    • > What exactly does Windows 11 do that breaks the CPU's automatic logic

      "AMD’s Ryzen processors based on the Zen, Zen+, and Zen 2 cores all share a common L3, but the structure of AMD’s CCX modules left the CPU functioning more like it had 2x8MB L3 caches, one for each CCX cluster, as opposed to one large, unified L3 cache like a standard Intel CPU." link [extremetech.com]
  • anyone know how some software update can make a certain cpu slower? and then another update make it fast again? it's not like there's drivers for cpus or is there? as far as I understand applications are compiled to cpu instructions, which are then run on the cpu, is this a compiler bug then? and the fix is installing recompiled apps?

Slashdot Top Deals

This place just isn't big enough for all of us. We've got to find a way off this planet.

Close