Search Result

[mczak]

That may be true, but it would only contribute a tiny bit to it being a perf/w leader. The truth is all these cpus are operating so far above their ideal...

[mczak]

I'm wondering how much single channel limits here. This is only Vega 3 after all (and for the cpu part memory bandwidth is usually no big deal neither)....

[mczak]

I think that's pretty much as expected. In the past the APUs always featured the newest display and video blocks (sometimes newer than in any released...

[mczak]

These chips aren't cpus. AFAIR nvidia doesn't do scoreboarding neither with current gpus (they did in some earlier chips).
Obviously, the idea is...

[mczak]

Too bad the hw can't afford decent sized line buffers. Or alternatively just use L2 for scanout when it intends to switch mem clocks...

[mczak]

Nope. A later version of the patch changed it to the universally used (albeit a bit more obscure unless you look up the cpu features docs) clearcpuid...

[mczak]

I think UMIP could be disabled via clearcpuid=514 on the boot command line, so sgdt from userspace should no longer crash?
I have no idea if the...

[mczak]

Actually no. It is fairly likely Icelake Celerons / Pentiums will support AVX2 imho. This is because the client icelake cpus support AVX 512. Hence, I'd...

[mczak]

The idea to require avx2 is completely insane (try again in another 10 years or so maybe...). Requiring fma is just as terrible (it's a separate feature,...

[mczak]

Ah yes of course, I was thinking 1 or 2 pre-rasterization....

[mczak]

The hardware VS does not correspond to API VS (which is also why the second part of the sentence explicitly says "API shaders".
The hardware...

[mczak]

Yes but that's what I'm saying - it is intentionally limited to 16 threads because scaling is bad. If you lift the limit it can use more threads but they...

[mczak]

Yes, scaling is bad, and for that reason it does not even try to use more than 16 threads (+ the main thread).


It's a result of disabling...

[mczak]

Nice! Albeit something seems to be very wrong with the Celeron N3160. This 4-core Celeron (Airmont core, Braswell) isn't exactly a speed demon, but it...

[mczak]

Apple has nothing do to with llvmpipe, and (to my knowledge) they've never used it neither.
Funnily enough though you could argue llvmpipe has roots...