AMD Ryzen Branch Prediction Optimizations Now Available to Windows 11 23H2

[RJARRRPCGP]

i'd still try other sata cables, they can be kooky sometimes .

My cable is like brand new. I checked with other HDDs before I hooked the oldest HDD I can confirm having. I find USB to be more kooky than SATA! I also actually had more issues with PATA than with SATA!
With USB, you can't tell that it's got a bad connection sometimes, unless you have at least one of the following symptoms:

Windows logs errors in the event log about the USB drive.

You have Windows on a USB stick and it loads very slowly and likely a BSOD occurs. Especially if it loads very slowly and then it reboots itself.

General strange slowness.

Explorer.exe crashes.

 

[AleXXX666]

Why I should use some "preview" updates? I am not free beta-tester for ya buggy updates. I will wait for the "release" update version and use it.

Windows ME was a peice of garbage. It's only saving grace was the addition of System Restore... which was a necessity due to its instability.

in wrong user's hands even super-stable server edition of some Linux distro could be "piece of garbage", lol.

 

[OkieDan]

in wrong user's hands even super-stable server edition of some Linux distro could be "piece of garbage", lol.

It wasn't the OS being in the wrong hands it was the OS itself. It was much less stable than Win98 SE that preceeded it and XP which succeeded it, to say otherwise is revisionist history.

 

[AleXXX666]

It wasn't the OS being in the wrong hands it was the OS itself. It was much less stable than Win98 SE that preceeded it and XP which succeeded it, to say otherwise is revisionist history.

well, 2000 was more stable than xp, then xp wasn't piece of cake either.

 

[OkieDan]

well, 2000 was more stable than xp, then xp wasn't piece of cake either.

I agree, XP was no "piece of cake ".

 

[chrcoluk]

Intel has it's share of problems with Windows due to Big / Little cores concept, impacting performance in some cases even more than just a few percent. Reviewers of course monitored that and made sure this didn't happen during benchmarking, but it's not something you can constantly tweak as a user.

The biggest issue I (and others doing the same testing) have with windows scheduler on Intel, is Windows really wants to prefer using HT over e-cores when there is more than 8 threads on a process, so this unoptimal thing is not an AMD exclusive thing like some have made out. Intel taking the brute force way of removing HT entirely to prevent Windows from doing dumb things.

Also looks like HUB are not doing the promised Intel testing on the new build, or its delayed, as the video never came.

 

[HD64G]

 

[Luckz]

I bet the performance uplift we see with 24H2 update is what Windows 10 performance have been looking all along?

This is probably why Windows 11 gaming performance has been mediocre (compared to 10) for years.

Yeah, https://www.purepc.pl/test-procesor...4h2-preview-bedzie-wzrost-wydajnosci?page=0,4
Looks like all it does it catch up to W10.
Cherrypicked Spoderman there, other games are less pronounced https://www.purepc.pl/test-procesor...4h2-preview-bedzie-wzrost-wydajnosci?page=0,4

 

[Dyno]

Is this BRANCH PREDICTION PATCH implemented into Windows 11 24H2?

 

[Dyno]

So what will AMD and Microsoft do if you recently bought a AMD Zen5 and are running Windows 11 24H2? They should push this patch out again. There are alot of people with AMD cpu's that don't have this patch installed.

 

[Event Horizon]

Is this BRANCH PREDICTION PATCH implemented into Windows 11 24H2?

So what will AMD and Microsoft do if you recently bought a AMD Zen5 and are running Windows 11 24H2? They should push this patch out again. There are alot of people with AMD cpu's that don't have this patch installed.

The patch is included with 24H2 as far as I know.