3 ms·
Doesn't it still depend on CUDA for hardware acceleration? For it to become truly open, it should be untied from CUDA first. So far it looks more like a way to
by shmerl 8y ago
Doesn't it still depend on CUDA for hardware acceleration? For it to become truly open, it should be untied from CUDA first.
So far it looks more like a way to advance CUDA usage even further, by giving a free higher level library that's locked into it.
- remind_me_again 8y agoDon't be naive.
- xvilka 8y agoSee also https://www.phoronix.com/scan.php?page=news_item&px=LPC2018-Video-Open-CUDA https://www.phoronix.com/scan.php?page=news_item&px=LPC2018-...
- shmerl 8y agoYep, that's nice. There should be more push to dislodge CUDA grip on the AI industry.
- maccard 8y ago> Doesn't it still depend on CUDA for hardware acceleration? Assuming by hardware you mean GPU acceleration, then yes. It's not really a push for Cuda usage, most games that use physx don't use the GPU acceleration (not everyone has PhysX cards, and those that do are usually busy using the GPU for rendering), so in practice for 99% of use cases, it is open.
- shmerl 8y agoModern cards commonly have rendering and compute pipelines available to be used in parallel. So in theory nothing stops games from using compute queues for physics.
- HelloNurse 8y agoThe issue is that CUDA, unlike OpenCL, is only available on Nvidia hardware.