I was surprised to find that the same program runs faster on the Apple Silicon M1 Max chip than on the M2 Ultra. This shouldn’t happen, as the latter is supposed to have more cores and higher computational performance. I’d appreciate it if somebody can give me an explanation about this.