ButtPlugForPM@alien.topB to Hardware@hardware.watchEnglish · 1 year agoCan AMD ever catch up to the lead NVIDIA has in A.i/ray tracing?message-squaremessage-square25fedilinkarrow-up11arrow-down10file-text
arrow-up11arrow-down1message-squareCan AMD ever catch up to the lead NVIDIA has in A.i/ray tracing?ButtPlugForPM@alien.topB to Hardware@hardware.watchEnglish · 1 year agomessage-square25fedilinkfile-text
Just wondering,what AMD would need to do…to at least MATCH nvidias offering in A.I/dlss/Ray tracing tech
minus-squareCANT_BEAT_PINWHEEL@alien.topBlinkfedilinkEnglisharrow-up0·1 year agoCan they make dedicated tensor to units or is that patented?
minus-squareGomaEspumaRegional@alien.topBlinkfedilinkEnglisharrow-up0·1 year agoThey don’t even need to make dedicated tensor units, since programmable shaders already have the necessary ALU functionality. The main issue for AMD is their software, not their hardware per se.
minus-squareResponsibleJudge3172@alien.topBlinkfedilinkEnglisharrow-up0·1 year agoNah, throughput of tensor cores is far to high to compete against
minus-squareGomaEspumaRegional@alien.topBlinkfedilinkEnglisharrow-up1·1 year agoWell, sure the application specific IP is always going to be more performant. But on a pinch, shader ALUs can do tensor processing just fine. But without a proper software stack, the presence of tensor cores is irrelevant ;-)
Can they make dedicated tensor to units or is that patented?
They don’t even need to make dedicated tensor units, since programmable shaders already have the necessary ALU functionality.
The main issue for AMD is their software, not their hardware per se.
Nah, throughput of tensor cores is far to high to compete against
Well, sure the application specific IP is always going to be more performant. But on a pinch, shader ALUs can do tensor processing just fine. But without a proper software stack, the presence of tensor cores is irrelevant ;-)