Comment by TheRealKing

No, you can still trust compilers: 1) The hand-tuned BLAS routines are essentially a different algorithm with hard-coded information. 2) The default OpenBLAS uses OpenMP parallelism, so much speed likely originates from multithreading. Set OMP_NUM_THREADS environment variable to 1 before running your benchmarks. You will still see a significant performance difference due to a few factors, such as extra hard-coded information in OpenBLAS implementation.