Which is the more relevant for Octane? If I was gonna use Octane, do I buy a GeForce with more CUDA cores but less RAM, or a Quadro with more RAM but less cores?
Finally, is nVidia's "compute capability" they rank all their cards with on http://developer.nvidia.com/cuda-gpus a relevant metric, or just some arbitrary marketing gibberish (see http://developer.nvidia.com/cuda-faq)
RAM vs CUDA cores
Forum rules
For new users: this forum is moderated. Your first post will appear only after it has been reviewed by a moderator, so it will not show up immediately.
This is necessary to avoid this forum being flooded by spam.
For new users: this forum is moderated. Your first post will appear only after it has been reviewed by a moderator, so it will not show up immediately.
This is necessary to avoid this forum being flooded by spam.
Depends what you need and what type of scenes you will be rendering. Most people on the boards would rank memory higher than speed. FYI: There are a few GeForce cards with 2 GB.Richard7666 wrote:Which is the more relevant for Octane? If I was gonna use Octane, do I buy a GeForce with more CUDA cores but less RAM, or a Quadro with more RAM but less cores?
The compute capability isn't a performance ranking. It tells you the capabilities of the card. The higher the number, the more advanced (and recent) the card.Richard7666 wrote:Finally, is nVidia's "compute capability" they rank all their cards with on http://developer.nvidia.com/cuda-gpus a relevant metric, or just some arbitrary marketing gibberish (see http://developer.nvidia.com/cuda-faq)
See here for more details: http://en.wikipedia.org/wiki/CUDA#Versi ... ifications