Ah yes, the typical everyone else is to blame attitude. If it worked with pervious gen, there really shouldn't be any reason why those are suddenly broken. If things suddenly start working after driver update, whose the incompetent part (rhetorical question)?Typical clulessness used as argument.to badmouth AMD.
If 26 games out of 30 work flawlessly it means that the 4 remaining ones have half broken code, i wont abound on that matter but FI google "Resident Evil crash", and do the same with F1 24, it s just so much easy to blame AMD for the incompetence of some games developpers.
while every company can do better, AMD seems to doa better job than Intel as of late. 4 games on one version of the CPU gives this bad review ? I wish I could say more.Ah yes, the typical everyone else is to blame attitude. If it worked with pervious gen, there really shouldn't be any reason why those are suddenly broken. If things suddenly start working after driver update, whose the incompetent part (rhetorical question)?
They need to improve their own software/firmware and involvement with devs across the board. Things sure could be better.
It's not the CPU problem anyway but GPU drivers. Plus I'm confident AMD has more games that work with their drivers than other SoC makers like Qualcomm, Intel or Apple. At least this is my experience with my Intel 1250U, Apple M1 Pro and AMD 7840U.while every company can do better, AMD seems to doa better job than Intel as of late. 4 games on one version of the CPU gives this bad review ? I wish I could say more.
AVX-512 support + huge frequency increase is not free.Thing I just noticed re-reading the Zen 5 Info thread:
6.5 / 4.15 = 56% increase
Zen 5 Zen 4 Zen 3 Zen 2 Transistor Count 8.315B 6.5B 4.15B 3.9B
8.315 / 6.5 = 27.9% increase
How did Zen 4, which is supposedly more or less a tweak on Zen 3, have this much more transistor count? Obviously the node leap was massive, but...that's still huge. 56% more.
Zen 5 is a massive rework with a completely new decode, and they still only added 28% in comparison.
Clear Linux really does give me better benchmark scores than other operating systems and especially Windows. So they'll be playing catch up for sometime.So are we going to hype windows updates as much as new generations of CPUs as they apperently deliver more performance uplift. I predict 20% with 25H2
How does this compare to Bergamo numbers in reviews? Bergamo is running at mere 1.8GHz here and Turin at 4GHz. Then that Turin is a regular 128 core part.Probably BIOS support.
Other than that, some Turin numbers are now surfacing
It's a useful perf/watt comparison. Here we have Bergamo, which was the Zen4 most-optimized for perf/area and perf/watt, and standard Zen5 (Turin) is dropping a bomb on it in perf/watt. Bergamo may only be running @ 1.8 GHz, but it's still burning 360W to achieve that feat.Can't find definitive comparisons yet but those numbers don't sound exceptional.
That's TDP not actual power consumption in that workload. Those results should also be verifiable, so let's get to it?Bergamo may only be running @ 1.8 GHz, but it's still burning 360W to achieve that feat.
Clear Linux really does give me better benchmark scores than other operating systems and especially Windows. So they'll be playing catch up for sometime.
From my understanding it works pretty well on both Intel and AMD.clear linux is made by intel optimized for intel right?
can you imagine AMD linux?? 🚀🚀 🍷🍷🍷🍷
Unfortunately it seems to be a single switch in Windows that is leading to most of the gains seen in Windows 24H2... so I doubt we will see such gains againSo are we going to hype windows updates as much as new generations of CPUs as they apperently deliver more performance uplift. I predict 20% with 25H2
Interesting how 24H2 shows some regressions from the latest patch to 23H2, granted it is a “preview” build.Unfortunately it seems to be a single switch in Windows that is leading to most of the gains seen in Windows 24H2... so I doubt we will see such gains again
Yes, and it's not a pretty picture seeing current AMD software efforts Since AMD basically reached feature parity with Intel, they stand to benefit from most of what Intel is doing. I would not expect that tuning OS for specific microarchitecture could get you much further anyway. But you can always try to check out CachyOS with Zen4/Zen5 optimized repositories. https://www.phoronix.com/review/linux-os-amd-ryzen9-9950xclear linux is made by intel optimized for intel right?
can you imagine AMD linux?? 🚀🚀 🍷🍷🍷🍷
HWUB got their 10-11% gains having turned off said switch in both 23H2 and 24H2.Unfortunately it seems to be a single switch in Windows that is leading to most of the gains seen in Windows 24H2... so I doubt we will see such gains again
How can something that has yet to happen been disproven many times?
I don't think its bias, just reviewers getting to grips with new items on their checklists.lol so now the update went from affecting one game out of 40 for Intel to all of a sudden it gives Intel more gains than AMD in all games? All this FUD is getting rather old. Bias clouds everything.
That's TDP not actual power consumption in that workload. Those results should also be verifiable, so let's get to it?
Also I think top is Compression and bottom is Decompression. They are nearly exactly the same for both chips. You'll see in other comparisons there is quite a bit of a gap. So why isn't it showing it here?
I almost want to get cheeky and ask them to start adding a "Level of Expected Degradation" on 14900/14700Ks".HWUB got their 10-11% gains having turned off said switch in both 23H2 and 24H2.
Might be that VBS off benefits more from the update, might be game selection, might be something completetly different... Still a lot of open questions.
Sorry if I'm stating the obvious and if I missed the context, but in general compression is more sensitive to branch prediction, while decompression is more sensitive to memory bandwidth (for not too complex compression algorithms). That's a generic statement that might apply or not to 7-zip My point is that changes to uarch don't apply the same to compression and decompression.Hard to say. Could have something to do with memory bandwidth and how desktop platforms never has as much available. Here's a 7950X3D that doesn't show the same patterns:
I need benchmark to 7-zip in Ryzen 7950X3D, who has it?
www.overclock.net
And this reviewer also showed compression being slower than decompression, even on a 9950X: