I did quite a bit of benchmarking. I mostly compile code, which parallelizes well. For that workload, the higher core count on the 16" means it's (a bit) faster

I did quite a bit of benchmarking. I mostly compile code, which parallelizes well. For that workload, the higher core count on the 16" means it's (a bit) faster. For single-core work, the M1 wins.