Jump to content

My latest invention.. sorta


Monk

Recommended Posts

I've been messing with CUDA support by offloading some functions for games into the GPU of a video card. So far, I've been able to manipulate some instructions by creating a virtual machine to directly call CUDA instructions for certian floating point operations, ie

 

foobar() -> cpu_foobar ? : 0 : 1 -> gpu_foobar()

 

I have a test CSS server that I've been screwing with, but when you exit the server it crashes. Another cool trick is using a video cards onboard memory for maps and temporary files, which is a factor of 30 faster than normal system memory :)

 

I'll take some screenshots when it isn't so messy. If I can move the core sleep functions into the GPU, it will be more accurate than any hardware timers due to the sheer speed and no latency..

Link to comment
Share on other sites

  • 1 month later...
  • 3 months later...

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. Terms of Use