Thread 107094695

Why don't phone manufacturers switch from phone CPUs to computer CPUs so we have to stop worrying about needing hardware accelerated decoding for modern image formats like jpeg xl?

Like what would be the tradeoff, 100 gram heavier battery and 0.2 inches thicker chassis? I can live with that desu.

>>107094695 (OP)
>Like what would be the tradeoff
10% as much battery life. Also nothing to do with the problem you want solved

>>107094716
I assume there's some kind of performance reason my PC can run GTA V at 100+ FPS on ultra settings and my phone can't...

>>107094716
Also why 10%? Like isn't the intel N100 supposed to be really power efficient. Obviously I'm not talking about putting an intel i9 on a phone LOL.

>>107094695 (OP)
For the same reason we don't use truck engines on mopeds.

>>107094742
if you compile it for arm64 it can run 100+ fps
because you're dynamically recompiling the x86 binary to arm64 you're incurring a performance penalty

>>107094695 (OP)
Why don't these devices get rid of DRM that is driving power and energy usage?
I thought they were trying to be environmentally friendly or something?

Reminder that you can not have safety AND environmental sustainability. These two cannot work with each other.

>>107095574
LOLNO. The base M4 macbook air, which is like lightyears ahead of most phone CPUs can do like 60 FPS on high settings.

But the whole thing sips on like 30 watts of power while doing that. Phones would explode if they tried that...

>>107095705
try reading it again until you understand
>if you compile it for arm64 it can run 100+ fps
>because you're dynamically recompiling the x86 binary to arm64 you're incurring a performance penalty

>>107095751
That's what the game porting shit does. Picrel is what LITERALLY JIT x86 emulation looks like.

Anyway you're ignoring my other point in that a phone using 30 watts under load = bomb.

>>107095817
you can just compile the game for arm architecture and it will not require such a hefty cpu.

>>107095871
Theoretically but it would be missing 1 gorrillion vector optimizations. ARMv9 was rumored to address this with 256-bit SVE but I think it never materialized in consumer ARM CPUs.

Still base M4 having the 95W performance of a 2500K at just 30W is still pretty cool.

>>107095949
Adding to this proofs:

>"The Apple M4 is an ARMv9.2a based design."
>"However, it lacks SVE (and SVE2) support."

>"There were rumors that the Apple M4 supported Scalable Vector Extensions but now again by this Apple code comment and the associated ISA being exposed by the LLVM compiler, SVE/SVE2 is not present for the Apple M4."

What a massive letdown desu senpai.

https://www.phoronix.com/news/Apple-M4-Added-To-LLVM-Clang