"Fuck you Nvidia" - L. Torvalds.
We should take this mantra to the Nvidia offices and chant it until we go hoarse, may the proprietary blob crumble like the walls of Jericho.
We should take this mantra to the Nvidia offices and chant it until we go hoarse, may the proprietary blob crumble like the walls of Jericho.
On top of that I don't think there's a lot of overhead associated with using different gpus models simultaneously, allowing you to incrementally migrate your servers to a new gpu.