I can't find anything on how the X3D chips are improving with this update. I'm really interested to see how much this helps the 5800X3D and 7800X3D chips. I would imagine that the improvement should scale the same as other 5000 and 7000 series chips (perhaps even a bit more), but what if it doesn't? What if the gains were primarily seen by non X3D parts, that would actually be quite devastating for AMD considering how popular X3D is. Again, I think that's extremely unlikely.

That's a huge Micro-software bottleneck to target multi generational Zen cpus only.

Just proves how much a mess is Windows.

I remember when Intel and MS teamed for Alder Lake optimization before the launch. Now, you see what kind of grip Intel has for being able to pull this up.

I can't wait for the day I can drop Windows for good.

redgarl said:

Just proves how much a mess is Windows.

I remember when Intel and MS teamed for Alder Lake optimization before the launch. Now, you see what kind of grip Intel has for being able to pull this up.

I can't wait for the day I can drop Windows for good.

Click to expand...

The better question is, how much of Windows is optimized for Intel in a way that causes issues for AMD cpus?

If you have "Get the latest updates as soon as they're available" turned on, it seems to just auto-install this straight away. Does require a reboot though.

I think re-testing the r75800x3D is probably a major priority right now. A 10% increase would make this CPU the best investment ever.

I just updated, we'll see how it goes. I can't help but think that some of the stability issues we see are because AMD doesn't test with actual user environments. They didn't see this issue because they were running all their benchmarks as an Administrator, but customers don't do that. Maybe this update will get us closer to the environments that AMD is actually testing with and we'll see more stability as a result?

So, there is no issue with Ryzen, just windows crap again, but still that was not mentioned.
Also, add here the new bios with higher tdp, for example msi, where additional 2 digit % are added in multi thread.
To be honest, I feel bias here towards puting amd in a bad spot. Disapointed. We need to read other sites to get the real picture.

mistyrain said:

So, there is no issue with Ryzen, just windows crap again, but still that was not mentioned.
Also, add here the new bios with higher tdp, for example msi, where additional 2 digit % are added in multi thread.
To be honest, I feel bias here towards puting amd in a bad spot. Disapointed. We need to read other sites to get the real picture.

Click to expand...

Wait, I thought Techspot was a shill for AMD, isn't that what others have complained about in the past? haha. In all seriousness, couldn't these things just be considered an error on the part of the author and editor? Why does it have to portray some kind of bias?

mistyrain said:

So, there is no issue with Ryzen, just windows crap again, but still that was not mentioned.
Also, add here the new bios with higher tdp, for example msi, where additional 2 digit % are added in multi thread.
To be honest, I feel bias here towards puting amd in a bad spot. Disapointed. We need to read other sites to get the real picture.

Click to expand...

Yes they are totally bias against AMD in showing Zen 5 has similar performance to Zen 4.

Wait Zen 4 is also AMD... -gasp-

guillejp said:

There's a patch. Fine.
Does the patch work as expected?

Click to expand...

Yes it works but not as well as 24H2 performance upgrade. Or it could be that I have Zen 3 and I did not test in purely CPU limited scenarios.

EdmondRC said:

I can't find anything on how the X3D chips are improving with this update. I'm really interested to see how much this helps the 5800X3D and 7800X3D chips.

Click to expand...

I posted this on TPU's forum earlier. It's my quick and dirty test on 5800X3D on 23H2 without and with the update. I was not testing CPU limited 1080p low. I was testing 1440p and mostly GPU limited but I did see some gains in 0.1% lows after the update. Averages mostly did not change and in two of the tests there was a slight regression but those run at so high framerates that it was unnoticeable in testing anyway. Most of my games are older games with only Ghost of Tsushima being this years release. Of course I have many more games but I tested games that I already had installed. I did not bother installing extra ones. I also tested some games but decided to omit them from results - for example Batman Arkham Knight that has a 90fps framecap and the 2013 (if I remember correct) Tomb Raider reboot.
https://www.techpowerup.com/forums/attachments/clipboard_08-28-2024_01-jpg.361016/
Here's the post itself if you cant see the image here: https://www.techpowerup.com/forums/...to-windows-11-23h2.326010/page-2#post-5319139

Contrary to what some people around here might say but this issue with performance on Ryzen chips can all be blamed on Spectre and Meltdown mitigations. That is, according to Dave Plummer who used to work for Microsoft back in the day. He mentioned something about how the kernel placed kernel memory in RAM that was incompatible with Ryzen's new branch predictor. Again, this is all thanks to the mitigations that had to be put into place because of Spectre and Meltdown.

Always knew that Spectre and Meltdown would rear its ugly head again, just not as I expected it would.

Contrary to what some people might say in that there's some kind of nefarious thing or that Windows is a mess or some kind of joke... NOPE!!! Not at all! Once again people are casting blame in the wrong direction just because they can't pass up a chance to stick it to Microsoft. Basically it's the TDS of the Windows world.

trparky said:

Contrary to what some people around here might say but this issue with performance on Ryzen chips can all be blamed on Spectre and Meltdown mitigations. That is, according to Dave Plummer who used to work for Microsoft back in the day. He mentioned something about how the kernel placed kernel memory in RAM that was incompatible with Ryzen's new branch predictor. Again, this is all thanks to the mitigations that had to be put into place because of Spectre and Meltdown.

Always knew that Spectre and Meltdown would rear its ugly head again, just not as I expected it would.

Contrary to what some people might say in that there's some kind of nefarious thing or that Windows is a mess or some kind of joke... NOPE!!! Not at all! Once again people are casting blame in the wrong direction just because they can't pass up a chance to stick it to Microsoft. Basically it's the TDS of the Windows world.

Click to expand...

Haha, well said at the end. Seems like malicious intent is often assumed, when anyone that has worked in software could share that these things happen all the time. Not every Engineer is going to be able to predict how their fix will affect other systems. Fairly often they're going to make a change that successfully resolves the issue reported issue and later need to come up with a different solution when their fix is found to have created a new problem.

Notmare said:

Fairly often they're going to make a change that successfully resolves the issue reported issue and later need to come up with a different solution when their fix is found to have created a new problem.

Click to expand...

That's what happens when code is as old as some of the code in Windows is. The older the code, the more difficult it can be to figure out what's going on and to fix bugs.

And yes, I'm sure that someone will come along and say that Microsoft could rewrite a lot of that old code but then those would be the same people who would complain that the new code is buggy, so Microsoft is truly in a damned if they do, damned if they don't kind of a situation.

See Windows 11 as a case study on how the community reacts to Microsoft rewriting stuff.

trparky said:

That's what happens when code is as old as some of the code in Windows is. The older the code, the more difficult it can be to figure out what's going on and to fix bugs.

And yes, I'm sure that someone will come along and say that Microsoft could rewrite a lot of that old code but then those would be the same people who would complain that the new code is buggy, so Microsoft is truly in a damned if they do, damned if they don't kind of a situation.

See Windows 11 as a case study on how the community reacts to Microsoft rewriting stuff.

Click to expand...

As a reminder to those folks who would say that, you can always switch to Linux for "free." But I suppose that could put those folks in their own damned it they do, damned if they don't situation. haha

Notmare said:

As a reminder to those folks who would say that, you can always switch to Linux for "free." But I suppose that could put those folks in their own damned it they do, damned if they don't situation. haha

Click to expand...

Exactly. Talk about out of the frying pan and into the fire.