GPU Compiler 128x speed increase Dr Huiyang Zhou - associate professor ncsu.edu

[kurtjn]

I found some interesting info about speeding up GPU power by up to 128 times faster using a GPU compiler http://news.ncsu.edu/releases/wmszhoucompiler/
Dr Huiyang Zhou, an associate professor of electrical and computer engineering at North Carolina.

Read more: GPU compiler could "turn desktops into supercomputers" | News | PC Pro http://www.pcpro.co.uk/news/358027/gpu- ... z2eIEiXZuf

I don't know if this could be implimented with average/current hardware infrastructures, we might have to have a fused CPU-GPU unit. Dr Huiyang Zhou had released his code as open source http://people.engr.ncsu.edu/hzhou/ (An Open-Source GPGPU Compiler (CUDA-to-CUDA code optimizer) -bottom of the page. I am no programmer and I don't know if this information is helpful to otoy's dev or not but if it makes sense and could possibly be integrated here ya go.

[boris]

as far as I understand this is a optimization for general purpose computation on gpu's which octane probably would make slower because the gpu architecture itself is what's making octane so fast.
from your linked text:

But using a GPU for general computing functions isn’t easy. The actual architecture of the GPU itself is designed to process graphics, not other applications. Because GPUs focus on turning data into millions of pixels on a screen, the architecture is designed to have many operations taking place in isolation from each other. The operation telling one pixel what to do is separate from the operations telling other pixels what to do. This hardware design makes graphics processing more efficient, but presents a stumbling block for those who want to use GPUs for more complex computing processes.

[mib2berlin]

Also it is from 2010 but interesting.

Cheers, mib.