News AMD updates Zen 5 Ryzen 9000 benchmark comparisons to Intel chips — details Admin mode boosts, chipset driver fix

[Admin]

AMD penned a blog post today to clarify Zen 5's gaming performance after withering criticism in the wake of its turbulent Ryzen 9000 launch.

AMD updates Zen 5 Ryzen 9000 benchmark comparisons to Intel chips — details Admin mode boosts, chipset driver fix : Read more

 

[Jagar123]

Meh, to me Zen 5 is still a bust. I'll be upgrading soon, just waiting for Arrow Lake and 9000x3d to see where I can get the most gaming performance at a good price. If it all stinks I'll just get a 7800x3d and be happy.

 

[redgarl]

What is getting out of it is that Windows is a damn unoptimized mess.

 

[redgarl]

Meh, to me Zen 5 is still a bust. I'll be upgrading soon, just waiting for Arrow Lake and 9000x3d to see where I can get the most gaming performance at a good price. If it all stinks I'll just get a 7800x3d and be happy.

Sure...

 

[spongiemaster]

What a trainwreck. AMD can't even agree with themselves in their press release. They state 2 different gaming performance improvements in this release.

For our Ryzen 9000 launch, AMD internal labs generated data that showed a 9% average generational uplift in 1080P gaming versus Ryzen 7000 Series,

Little further down

On a generational basis, Ryzen 9000 Series delivers a ~10% improvement in productivity and creative workloads, ~25% improvement in AI workloads, and 5-8% improvement in gaming over the Ryzen 7000 Series.

These are cut and pasted from the press release on AMD's website. Last time I checked, 9 doesn't fall between 5 and 8. No where in the release does it say 9% were their "original" results and now things have changed.

https://community.amd.com/t5/gaming...mmunity-update-gaming-performance/ba-p/704054

 

[Guardians Bane]

Meh, to me Zen 5 is still a bust. I'll be upgrading soon, just waiting for Arrow Lake and 9000x3d to see where I can get the most gaming performance at a good price. If it all stinks I'll just get a 7800x3d and be happy.

I'm doing the same thing. Waiting on both to see reviews of X3D vs ARL. Like you, I might just go with a 7800x3d also. Price/performance just isn't there with 9000x currently.

 

[Guardians Bane]

Sure...

Phoronix is testing on Linux. The article even said 9000 is better on Linux. What's the point you're making?

 

[sjkpublic]

Microsoft VBS is a kluge for 3rd party vendors and security. And it kills off any competing vendors to HyperV. I agree that both AMD and Intel have to deal with this kluge. My take is at the end of the day the overall performance between Intel and AMD is about the same. But AMD uses less power. Go green.

 

[Jagar123]

Phoronix is testing on Linux. The article even said 9000 is better on Linux. What's the point you're making?

I don't think they care. I mentioned gaming and he showed me Linux productivity results. Not really worth engaging with them.

 

[DougMcC]

Phoronix is testing on Linux. The article even said 9000 is better on Linux. What's the point you're making?

I assume the point is that the results suggest maybe the 9000 is actually pretty decent but windows is such a mess that it's unreasonably handicapped in those results.

 

[Guardians Bane]

I assume the point is that the results suggest maybe the 9000 is actually pretty decent but windows is such a mess that it's unreasonably handicapped in those results.

While true, it still isn't indicative of what most consumers get out of the box. And till MS and AMD come up with something to fix the bugs, those Linux results won't show most people what they can expect currently.

 

[baboma]

>I assume the point is that the results suggest maybe the 9000 is actually pretty decent but windows is such a mess that it's unreasonably handicapped in those results.

The "Windows sux" narrative is popular with the hoi polloi, and spin doctors always reach for it. It's low-browed scapegoating.

It's the HW vendor's responsibility to provide (and optimize) drivers for an OS, be it Windows or Linux. It's not the OS' responsibility.

The reality is that AMD launched Ryzen 9K with subpar driver support, and will need to be improved. AMD implicitly stated this in its blog piece.

Dual-CCD thread scheduling is an AMD issue, just as hybrid-core scheduling is/was an Intel issue. Heterogeneous thread scheduling AFAIK isn't a simple thing.

 

[Mattzun]

This still doesn't cover all of the Zen 5 weirdness on Windows.
It really does seem like they just weren't ready to launch and that they still don't have good drivers.

The "Admin" mode seems to affect both Zen4 and Zen5, so 24H2 changes won't improve relative performance.

VBS seems to have an unusually large effect on Zen 5 vs Zen4 in some games

Its weird that you need core parking on all dual CCD Zen 5s, when it was only needed for X3D versions of Zen 4
Being able to uninstall the core parking driver doesn't explain why its needed

I didn't see any explanation why memory compatibility is better in Zen 4 than Zen 5

 

[codex5600x]

They say this generation it's not for gaming, a CPU ist not use only for gaming, they have a CPU dedicated for gaming.
The 9000 series are focus for web, ai, workstations, etc were need a good performance with less heat and power consumption

 

[phxrider]

I have been saying, and still say, people should hold off on final judgement for a few months while the kinks are worked out with Zen 5. For the moment it looks like a flop, but I'd give them (AMD and MS, and potentially game devs) time to optimize things for it.

 

[TheHerald]

I assume the point is that the results suggest maybe the 9000 is actually pretty decent but windows is such a mess that it's unreasonably handicapped in those results.

No. You don't think amd tested and developed this thing first and foremost for Windows? How can it be windows fault? Doesn't make any sense.

 

[ottonis]

Here is a slightly different perspective and take on these gaming benchmarks:
Back in the day, having a Pentium 60 or a Pentium 120 could make a difference between "completely unplayable" and "sufficiently playable" gaming.

Nowadays, however, even several years old CPUs and current-gen, low-tier CPUs offer more than plenty of fps in the vast majority of games - they are arguably all very well playable. Frankly, I don't care if CPU A is 5% faster than CPU B as long as we are talking about framerates that provide a smooth gaming experience that isn't choppy or doesn't get in the way of fast action.
YMMV, and I am sure that professional game players may be bothered even by a few fps difference, but for the vast majority of standard users, all these benchmarks only provide theoretical differences, most of which are not relevant to everyday gaming experience.

Personally, I am much more interested in a) performance per Watt and b) AI and transcoding speed and efficiency, for example for upscaling videos and other content creation stuff. This is where I want my CPU to save me as much time as reasonably possible, without breaking the bank and without generating half a kW of heat dissipation.

 

[-Fran-]

When will AMD stop clowning themselves?

They could publish their internal testing settings for everyone to check and scrutinise and some of this would have been prevented.

The Windows stuff has no excuse though. Complete miss.

Let's see how the patches improve the situation for Zen, as it seems like it'll be an across-the-board uplift?

Regards.

 

[stuff and nonesense]

No. You don't think amd tested and developed this thing first and foremost for Windows? How can it be windows fault? Doesn't make any sense.

Check out Wendell, Zen 5 Gaming: Where’s my 5%. Windows vs Patch Windows.
He gives a non-partisan, non-tribal, non-judgemental assessment of windows 11 current against windows 11 insider agains linux. It’s a 40 minute video so those with short attention spans may not watch it through.

He also states that he doesn’t know why… watch it then say windows can’t be better. (Hint, win 11 insider build shows some improvements)

To be fair to Microsoft, they write an operating system for a huge number of motherboards, processors, GPUs, memory, keyboard, mice, sound devices, miscellaneous display programmes, RGB, fan control, monitor control etc etc. There isn’t a standard PC. With that I can understand (not necessarily agree with) Microsoft culling support for older CPUs, using TPM gave them the premise to do so. Windows code base is built on the ancient(?) original NT stuff, iterated, some parts culled, some parts extended, some parts new. Multicore wasn’t a thing at initial design, Win 2000 advanced server was only 4 core and for the main part only exotic workstations had dual core. Athlon x2, pentium 4D really weren’t used well, things got better as you can see improvements in throughput (to a limit) as cores increase but it is limited.

I can’t speak for Linux.. watch Wendell’s video. It looks better in some games, worse in others BUT it doesn’t require the cludges Windows demands to make the scheduling system work.

 

[mitch074]

These are cut and pasted from the press release on AMD's website. Last time I checked, 9 doesn't fall between 5 and 8. No where in the release does it say 9% were their "original" results and now things have changed.

One is 1080p, the other is all resolutions combined. Considering GPU bottlenecking at 1440p and 4K, not unreasonable.

 

[-Fran-]

Check out Wendell, Zen 5 Gaming: Where’s my 5%. Windows vs Patch Windows.
He gives a non-partisan, non-tribal, non-judgemental assessment of windows 11 current against windows 11 insider agains linux. It’s a 40 minute video so those with short attention spans may not watch it through.

He also states that he doesn’t know why… watch it then say windows can’t be better. (Hint, win 11 insider build shows some improvements)

To be fair to Microsoft, they write an operating system for a huge number of motherboards, processors, GPUs, memory, keyboard, mice, sound devices, miscellaneous display programmes, RGB, fan control, monitor control etc etc. There isn’t a standard PC. With that I can understand (not necessarily agree with) Microsoft culling support for older CPUs, using TPM gave them the premise to do so. Windows code base is built on the ancient(?) original NT stuff, iterated, some parts culled, some parts extended, some parts new. Multicore wasn’t a thing at initial design, Win 2000 advanced server was only 4 core and for the main part only exotic workstations had dual core. Athlon x2, pentium 4D really weren’t used well, things got better as you can see improvements in throughput (to a limit) as cores increase but it is limited.

I can’t speak for Linux.. watch Wendell’s video. It looks better in some games, worse in others BUT it doesn’t require the cludges Windows demands to make the scheduling system work.

AMD and nVidia don't have all features in their Linux drivers vs Windows, so some things (like RayTracing, HDR and such) are just not implemented the same way or at all in them.

That's the biggest contributing factor why in Linux some things perform so much better. If you do an image analysis, you will spot differences across Linux and Windows.

Regards.

 

[TheHerald]

Here is a slightly different perspective and take on these gaming benchmarks:
Back in the day, having a Pentium 60 or a Pentium 120 could make a difference between "completely unplayable" and "sufficiently playable" gaming.

Nowadays, however, even several years old CPUs and current-gen, low-tier CPUs offer more than plenty of fps in the vast majority of games - they are arguably all very well playable. Frankly, I don't care if CPU A is 5% faster than CPU B as long as we are talking about framerates that provide a smooth gaming experience that isn't choppy or doesn't get in the way of fast action.
YMMV, and I am sure that professional game players may be bothered even by a few fps difference, but for the vast majority of standard users, all these benchmarks only provide theoretical differences, most of which are not relevant to everyday gaming experience.

Personally, I am much more interested in a) performance per Watt and b) AI and transcoding speed and efficiency, for example for upscaling videos and other content creation stuff. This is where I want my CPU to save me as much time as reasonably possible, without breaking the bank and without generating half a kW of heat dissipation.

This is exactly why I buy Intel. More cores at every price point means I can have some heavy stuff on the background running while enjoying a flawless experience. Basically a 300 euro 13700k offers you 16cores of raw performance with high framerates for all your games.

And having so many cores means you can pull the power draw back while still being much much faster than it's competition for those transcoding workloads.

 

[stuff and nonesense]

AMD and nVidia don't have all features in their Linux drivers vs Windows, so some things (like RayTracing, HDR and such) are just not implemented the same way or at all in them.

That's the biggest contributing factor why in Linux some things perform so much better. If you do an image analysis, you will spot differences across Linux and Windows.

Regards.

That’s fine, it would be wrong to compare windows ray tracing against linux raster? rendering. Also I’d guess that windows is using dx12 against Vulcan, though this is a guess and it isn’t stated in the video.

Assume that in cyberpunk the 205 fps is used as a target, a reference. What was demonstrated was that there was an improvement with the as yet unreleased windows 11 build. Further improvements were achieved with bios adjustments. That the performance could be improved shows problems for windows .

Yes it’s running a virtualised security instance and there are other overheads, different overheads for drivers, different overheads for the OS etc etc and it’s hard to do apples to apples comparisons. Note at the start of the video, he wasn’t claiming a cure for the woes, he was pointing out inconsistencies between 2 ASUS motherboards, then how on the quicker of the 2 boards how VBS affected cyberpunk… and then tweaks… None of that was needed (admittedly for a very limited selection of games) to achieve very good results on Linux. He doesn’t say Linux is perfect, just that he likes it but he isn’t evangelising.
If Microsoft, AMD and Intel actually talk and iron out their needs/limitations perhaps we can all have a better Windows.

 

[jp7189]

What is getting out of it is that Windows is a damn unoptimized mess.

I recall reading that Intel was involved in the Win11 kernel development so that it would be optimized for thread director from the jump. It doesn't seem AMD gets the same consideration.

 

[jp7189]

>I assume the point is that the results suggest maybe the 9000 is actually pretty decent but windows is such a mess that it's unreasonably handicapped in those results.

The "Windows sux" narrative is popular with the hoi polloi, and spin doctors always reach for it. It's low-browed scapegoating.

It's the HW vendor's responsibility to provide (and optimize) drivers for an OS, be it Windows or Linux. It's not the OS' responsibility.

The reality is that AMD launched Ryzen 9K with subpar driver support, and will need to be improved. AMD implicitly stated this in its blog piece.

Dual-CCD thread scheduling is an AMD issue, just as hybrid-core scheduling is/was an Intel issue. Heterogeneous thread scheduling AFAIK isn't a simple thing.

I disagree with this. I would go so far as to say Intel is the reason we got Win11 when we did - because the Win10 kernel wasn't designed for heterogeneous cores (P and E on the same CPU)