Efficient float32 arithmetic in JavaScript

Re:Efficient float32 arithmetic in JavaScript by andhow on Thursday November 07, 2013 @02:51PM (#6691952)

Referring to http://agner.org/optimize, both Nehalem (Intel Core i7) and Jaguar (AMD Kabini) instruction tables: you are right that addss, subss show the same latency as addsd and subsd, resp. However, mulss and divss show better latencies than mulsd and divsd.

[ Parent | on HN ]

Re:Efficient float32 arithmetic in JavaScript by Dylan16807 on Thursday November 07, 2013 @04:35PM (#6692574)

But if your code vectorizes, can't you get twice the throughput on sections?

[ Parent | on HN ]

Re:Efficient float32 arithmetic in JavaScript by andhow on Thursday November 07, 2013 @06:33PM (#6693297)

Agreed, I was just replying to the OP that, even on desktop using scalar ops, there are advantages to single vs. double precision ops.

[ Parent | on HN ]