4 ms·
It's not that you couldn't have used fixed SIMD, it's so it can rescale the SIMD automatically.
by Dylan16807 5y ago
It's not that you couldn't have used fixed SIMD, it's so it can rescale the SIMD automatically.
- owlbite 5y agoRight, but portable code != portable performance. See also: OpenCL. There's also the observation "keep simd vectors small but many" (e.g. Apple's arm chips) over "super long vectors" (intel avx512) is superior as it is much more flexible whilst delivering similar performance for tasks that are amenable to larger vectors. Having an architecture pushing towards the latter seems a retrograde step to me.