Hacker News new | past | comments | ask | show | jobs | submit login

Interesting thanks for the answer. From what you said it feels like it could be good for AVX-like CPU accelerated instructions with all those latencies it would be an optimization like loop unrolling ; but for GPU, really ?



Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: