Some apps and other code doesn’t appear to run faster on M1 chips, and some even runs more slowly. Could this be a result of it not using the best acceleration for vectors and matrices?
floating point
Many processors like the ARM64 have instructions to perform fused multiply-add operations. Do they deliver reduced error and better performance?
A new tailor-made log view shows what’s happened in interactions with the App Store, and can give clues as to the cause of delays and failures.
Close to 0.0, there are huge numbers of different floating-point Doubles, but their value gets greater, they are far fewer.
All you need to understand about numbers on your Mac, both integers and floating point. Common errors and problems, from endianness to NaNs.
M1 Macs may be as fast as greased lightning, but how good is their numerical accuracy? Can it match that of Intel processors?
The Eclectic Light Company 