Jump to content

PC Community Thread


stepee

Recommended Posts

45 minutes ago, AbsolutSurgen said:

All of the chips are "bad" in that they have (if Intel's statement tells the story) microcode that requests too much voltage some of the time.  All transistors age or "degrade" over time due to voltage.  Higher voltage causes transistors to degrade faster.  All you will achieve right now by knowingly running extra voltage through your CPU is shorten its lifespan -- with nothing to gain if all you're doing is running a stress test.  Will it meaningfully reduce the life of your CPU? Probably not. But, why do it?

 

You find out now whether you need to RMA it or not, instead of waiting a month for the official "fix".  

Link to comment
Share on other sites

12 minutes ago, cusideabelincoln said:

 

You find out now whether you need to RMA it or not, instead of waiting a month for the official "fix".  

Even if it passes today, the transistors could have "aged" to a point where it will fail 6 months from now.

 

That is, if these failures are really due to voltage-induced transistor aging.

Link to comment
Share on other sites

30 minutes ago, AbsolutSurgen said:

Even if it passes today, the transistors could have "aged" to a point where it will fail 6 months from now.

 

That is, if these failures are really due to voltage-induced transistor aging.

 

If 2 hours of stress testing would cause this, then a person better not use their computer at all until the fix rolls out. CPUs will boost to their max frequency/voltage when loading a web page and playing games.

Link to comment
Share on other sites

16 minutes ago, AbsolutSurgen said:

:shrug:

Intel reportedly suggests doing this test.  I would much rather do this than run a 2-hour stress test.

 

 

That's basically a stress test with a GUI. Reinstalling the drivers up to 10x in a row? Run Y-cruncher for 15 minutes to get the same affect.

 

I swear you are contrarian just to be contrarian.

Link to comment
Share on other sites

I'm not completely buying this freakout. My 13700k just went under some intense workloads for the past 3 hours, it never went above 67 degrees and the usage was never above 80%. I have an MSI Z790-Pro and NZXT Kraken z63 either the defaults on it are fine or it's managing itself / not all MBs are tweaked too high. I've never heard the fans or GPU on this machine (4080) since I built it last spring. It's also on a NZXT H5 case w/ Noctua fans so the air is colder coming out than my office!

Link to comment
Share on other sites

12 minutes ago, SuperSpreader said:

I'm not completely buying this freakout. My 13700k just went under some intense workloads for the past 3 hours, it never went above 67 degrees and the usage was never above 80%. I have an MSI Z790-Pro and NZXT Kraken z63 either the defaults on it are fine or it's managing itself / not all MBs are tweaked too high. I've never heard the fans or GPU on this machine (4080) since I built it last spring. It's also on a NZXT H5 case w/ Noctua fans so the air is colder coming out than my office!


FWIW, the degradation isn’t instant. All CPUs degrade over time, the issue is that it’s “sped up” on 13th + 14th gen, so it’s happening over the course of 1-2 years rather than 8-10+

  • Halal 1
Link to comment
Share on other sites

Only positive is if you're a Linux user:

 

image.php?id=amd-ryzen-5-9600x-ryzen-7-9

 

Which might mean Zen 5 will claim its expected 15% boost over Zen 4 in future applications more optimized for it.

 

WWW.PHORONIX.COM

In total I ran nearly 400 benchmarks in full across these new Ryzen 5 9600X and Ryzen 7 9700X desktop processors along with all of the re-testing over the summer of the Intel Core 13th/14th Gen and Ryzen 5000/7000 series.

 

Link to comment
Share on other sites

Just now, cusideabelincoln said:

Only positive is if you're a Linux user:

 

image.php?id=amd-ryzen-5-9600x-ryzen-7-9

 

Which might mean Zen 5 will claim it's expected 15% boost over Zen 4 in future applications more optimized for it.

 

WWW.PHORONIX.COM

In total I ran nearly 400 benchmarks in full across these new Ryzen 5 9600X and Ryzen 7 9700X desktop processors along with all of the re-testing over the summer of the Intel Core 13th/14th Gen and Ryzen 5000/7000 series.

 

 

so…good news at least for a future zen5 deck?

Link to comment
Share on other sites

3 minutes ago, cusideabelincoln said:

 

Well for mobile Zen 5 isn't as bad as desktop because even though there's little performance gain it is using less power while doing so.

 

 So does it seem likely then that 9800x3D will be a bit of a disappointment?

Link to comment
Share on other sites

27 minutes ago, cusideabelincoln said:

 

Seems likely, unless there's some magic BIOS update but I have not heard anything about current BIOS's being a problem.

 

ok :(

 

So I guess we are stuck hoping devs actually optimize their games ughhhh - cmonnnnn dlss4 3for1 frame gen

Link to comment
Share on other sites

14 minutes ago, stepee said:

 

ok :(

 

So I guess we are stuck hoping devs actually optimize their games ughhhh - cmonnnnn dlss4 3for1 frame gen

 

Zen 5 seems to shore up some weaknesses (relative to Intel) that Zen 4 had. It performs much better in workstation applications, especially AVX512, and not a substantial upgrade for gaming. 

 

Maybe if you get lottery lucky with a Zen 5 that can do high memory speeds would gaming performance see a noticeable improvement.

Link to comment
Share on other sites

PSA to anyone who eyeballs the 9950X and 9900X: They need the special drivers and Windows setup that the 7950X3D and 7900X3D do if you want to maximize performance, especially in games.

 

Why?

 

For some reason - current theory is aggressive core parking - there is a huge latency penalty when threads bounce from one CCD to the other - a much, much larger one than on any previous Zen processor. 

 

 

Quote

For reference, on the Ryzen 9 7950X, going to another CCD is around 76ns. But in Ryzen 9 9950X, we're seeing an average latency of 180ns, over twice the cost of the previous generation of Ryzen. Making this all the more confusing, Granite Ridge (desktop Ryzen 9000) reuses the same IOD and Infinity Fabric configuration as Raphael (Ryzen 7000) – all AMD has done is swap out the Zen 4 CCDs for Zen 5 CCDs. So by all expectations, we should not be seeing significantly higher inter-CCD latency here.

Our current working theory is that this is a side-effect of AMD's core parking changes for Ryzen 9000. That cores are being aggressively put to sleep, and that as a result, it's taking an extra 100ns to wake them up. If that is correct, then our core-to-core latency test is just about the worst case scenario for that strategy, as it's sending data between cores in short bursts, rather than running a sustained workload that keeps the cores alive over the long-haul.

 

Latency also seems to be worse even within a single CCD compared to Zen 4. This could explain why Zen 5 performs worse than Zen 4 in a select few workloads. Zen 5's stellar Linux performance also reinforces this latency; since Linux is better at scheduling threads than Windows it can keep the cores from parking as often to better utilize the actual IPC gains the architecture has.

Link to comment
Share on other sites

  • 2 weeks later...

Windows 11 might have been doing gamers, AMD (and sometimes Intel) dirty

 

GV56fvKawAAB_hM?format=jpg&name=large

 

GV56fvPbgAAeyHZ?format=jpg&name=large

 

 

Gears 5 showing a massive improvement is probably related to a Windows issue, not an AMD issue, since Intel also sees huge gains. Without more testing, it's hard to say which games' improvements are from general Windows optimizations or AMD-specific ones.

 

Supposedly these improvements should also apply to Zen 3 processors.

Link to comment
Share on other sites

11 minutes ago, cusideabelincoln said:

Windows 11 might have been doing gamers, AMD (and sometimes Intel) dirty

 

GV56fvKawAAB_hM?format=jpg&name=large

 

GV56fvPbgAAeyHZ?format=jpg&name=large

 

 

Gears 5 showing a massive improvement is probably related to a Windows issue, not an AMD issue, since Intel also sees huge gains. Without more testing, it's hard to say which games' improvements are from general Windows optimizations or AMD-specific ones.

 

Supposedly these improvements should also apply to Zen 3 processors.


I’m interested to see 5900x benchmarks 

Link to comment
Share on other sites

1 minute ago, Spork3245 said:


I’m interested to see 5900x benchmarks 

 

Yeah we need to see tests for dual-CCD CPUs for each generation as well, to see if Windows is also improving its thread scheduling and not needlessly bouncing them from one core to another.

 

I will also speculate that all Ryzen CPUs might see a performance boost because the Ryzen architecture uses its L3 cache differently than Intel (victim vs inclusive cache).

Link to comment
Share on other sites

On 8/26/2024 at 3:28 PM, cusideabelincoln said:

 

Yeah we need to see tests for dual-CCD CPUs for each generation as well, to see if Windows is also improving its thread scheduling and not needlessly bouncing them from one core to another.

 

I will also speculate that all Ryzen CPUs might see a performance boost because the Ryzen architecture uses its L3 cache differently than Intel (victim vs inclusive cache).


I went ahead and got the preview build. I hope my PC doesn’t explode  :shrug:

  • Shocked 1
Link to comment
Share on other sites

Well, that was quick. That boost has to be more than just optimizations. Probably a fresh install of Windows clears out any bugs and bloat of updates and patches.

 

Other testers are getting results, which seem to be all over the place. Big improvements, no improvements, and sometimes regressions.

 

WWW.HARDWARELUXX.DE

Optimierte Sprungvorhersage: Zen 5 mit Windows 11 24H2 getestet.

 

Link to comment
Share on other sites

2 minutes ago, cusideabelincoln said:

Well, that was quick. That boost has to be more than just optimizations. Probably a fresh install of Windows clears out any bugs and bloat of updates and patches.

 

Other testers are getting results, which seem to be all over the place. Big improvements, no improvements, and sometimes regressions.

 

WWW.HARDWARELUXX.DE

Optimierte Sprungvorhersage: Zen 5 mit Windows 11 24H2 getestet.

 


I didn’t do a fresh install, just updated after opting into the insider builds. That’s with a 5900x. The prior benchmarks were saved from when I first got my 4090. I have other games that I’ll try tomorrow.

Link to comment
Share on other sites

Okay, yea, my old bench is in DX11 mode with 4xMSAA and the one I did just now is DX12 with 2xMSAA. However, DX12 mode is unstable AF despite the better performance. I just ran it again in DX11 4xMSAA: avg fps was 56 this time which keeps inline with the ~10% boost in performance most are getting. 

  • Halal 1
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...