Re:How do compilers optimize divisions?
by adwn on Tuesday February 21, 2017 @02:05AM
(#13693849)
That is not true; compare kr7's comment [1]:
> On Skylake [...] 32/64-bit floating point multiplication has a 4 cycle latency with a throughput of 2 instructions / cycle.
Of course, there are some operations that are very expensive (trigonometric functions, for example), but they're not necessary here, and they're also very expensive on the GPU.
> On Skylake [...] 32/64-bit floating point multiplication has a 4 cycle latency with a throughput of 2 instructions / cycle.
Of course, there are some operations that are very expensive (trigonometric functions, for example), but they're not necessary here, and they're also very expensive on the GPU.
[1] https://news.ycombinator.com/item?id=13693749