User Profile

[piotrj3]

Also I would like to add, that common data set for desktop computers is not big but rather small, and when for arrays of size let's say ~~100 SIMD code...

[piotrj3]

?!?!?!?!?!?
There is so many things wrong in this that i cannot even comprehend.

How many assembly lines you wrote in your life? And...

[piotrj3]

Problem is not being "too wide". Problem is way it is implemented comparing to AVX1 or AVX2. Generally if you can split data into 256 bit chunks...

[piotrj3]

There is one more thing, because in developer guidelines, google says that if device is 64bit and has android 10, you can assume it has Vulkan 1.1. "Strongly...

[piotrj3]

Nvidia themselves added Vulkan RTX extensions before support arrived to DX12.... So idk why you bring Nvidia here, Nvidia supports raytracing via Vulkan,...

[piotrj3]

You have to remember that Nvidia's aproach is actually realistic physics wise raytracing via BRDF. Raytracing (especially physics wise realistic) is very...

[piotrj3]

Lies. Android 10 even requires GPU to support Vulkan 1.1 as minimum. So basicly Vulkan on mobile devices isn't even behind comparing to desktops. Also...

[piotrj3]

Well suprising part is that Sunnit with Nvidia voltas+Power9 is still more energy efficient then those ARM CPUs, especially considering Power9 is not...