3 ms·

Maybe its a dumb question, but isn't something like OpenCL meant to solve this problem?

• jvanderbot 3y ago
  From my understanding, using triangle / shaders to do HPC has given way to a specific, more general purpose GPU programming paradigm which is CUDA. Of course this knowledge is superficial and probably outdated, but if I'm not too far off base, it's probably more work to translate a general CUDA-like layer or CUDA libs to OpenCL.
  • slackito 3y ago
    The fact that you're comparing CUDA to using triangles and shaders makes me think you might be confusing OpenCL with OpenGL. OpenCL is meant for general computation (the C is for "computing") rather than graphics, like CUDA.
• VHRanger 3y ago
  In theory, yes. In practice, OpenCL became a giant mess. Some vendors put speed bumps by not supporting the transition from 2 to 3, or having shitty drivers for it. It also sat at the wrong level of abstraction for high performance compute, which is why CUDA ended up being used. Vulkan would have been reasonable to write compute shaders in, if there wasn't a ton of alternatives out there already now
  • surge 3y ago
    >In practice, OpenCL became a giant mess. Some vendors put speed bumps by not supporting the transition from 2 to 3, or having shitty drivers for it. Well, Nvidia's stock price in the age of AI indicates how bad of a screw up that was, they're locked out of the market growth until they play catch up. By then, Nvidia might have an insurmountable foothold.