This build is giving me no problems at all. I’d say the Julia native ARM version is around 30% faster than the Intel version running on Rosetta (Apple’s Intel emulation layer) based on very informal timings. I time sections of a simulation model and print them out after each run, which I eyeball comparing across many runs.
Ars Technica articles have claimed that even Apple emulation of the Intel instruction set runs faster than native Intel–as shipped in Apple Intel MacBooks, which don’t use the latest/greatest Intel chip.
Nice work. Thank you.