Douman
I re-wrote your benchmark to make sure your measurements do not get affected by optimizations in a best possible way. Please refer to [black_box](https://doc.rust-lang.org/std/hint/fn.black_box.html) for more details. Also be careful...
Did you wrap every computation into black box as I shown previously? I'm honestly not sure how it is possible you have such big difference between two versions of the...
On side note while playing with your benchmark I noticed that compiler didn't inline some internal functions properly so I re-worked it a bit to improve performance quite a bit