46
Modification Help / Re: [WIP] Autodetecting BLHack's aimbot.
« on: December 29, 2017, 07:21:07 PM »yes but the very nature of simd means that you can have a computer execute all of those in parallel...? and discussion of simd/mimd/misd/sisd is only ever in the context of parallelization???
That has nothing to do with threads, it's about using the same instructions on 4 (sometimes more) values simultaneously.
Instead of loading two floats from registers and adding them, you load 2 blocks of 4 floats from special simd registers and add those, resulting in a new block of 4 floats. Since that takes about the same time as a normal add, you can massively speed up some math code that needs to do the exact same instruction on different data, like adding 2 vectors or multiplying them.
Btw, this only works on 128-bit blocks, so compiling a normal vector math library that uses 3 floats (96 bits) with sse support actually changes mostly nothing.