Hacker News .hnnew | past | comments | ask | show | jobs | submitlogin

>quite a few RISC-style instructions require multiple (sometimes many) clock cycles to complete, such as mul/div, floating point math

Which seems like stuff you want support for, but this is seemingly arguing against?



It seems contradictory because the "one clock per instruction" is mostly a misconception, at least with respect to anything even remotely modern.

https://retrocomputing.stackexchange.com/a/14509




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

Search: