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

This seems like a use-case where the Mill CPU architecture would work quite well.

With the Mill's Loop pipelining and Mill's Vector ops this could be pretty well-optimized.




Pretty sure you could do a single-instruction vectorized version of the C code on Gold with 2c latency, yeah.




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

Search: