Ruby versus Matz's Ruby fastest programs Always look at the source code. These are only the fastest programs. Do some of them use manually vectorized SIMD? Look at the other programs. They may seem more-like a fair comparison to you. fannkuch-redux source secs mem gz busy cpu load Ruby 181.72 26,276 1455 716.75 99% 100% 99% 96% Matz's Ruby 3,668.48 17,512 1455 14,149.20 95% 96% 98% 98% binary-trees source secs mem gz busy cpu load Ruby 23.80 566,560 1008 67.07 58% 96% 63% 64% Matz's Ruby 316.78 1,413,456 1008 997.03 73% 72% 75% 95% fasta source secs mem gz busy cpu load Ruby 25.63 120,804 1069 28.41 0% 10% 100% 1% Matz's Ruby 266.99 101,684 1020 295.69 33% 3% 3% 71% n-body source secs mem gz busy cpu load Ruby 232.08 22,968 1137 236.00 0% 1% 100% 0% Matz's Ruby 2,280.53 3,816 1137 2,298.53 0% 0% 0% 100% mandelbrot source secs mem gz busy cpu load Ruby 262.79 69,328 880 1,049.78 100% 100% 100% 100% Matz's Ruby 1,961.35 53,240 970 7,844.90 100% 100% 100% 100% spectral-norm source secs mem gz busy cpu load Ruby 118.16 24,764 326 120.08 0% 1% 100% 0% Matz's Ruby 547.06 19,728 758 2,056.03 93% 95% 94% 94% k-nucleotide source secs mem gz busy cpu load Ruby 77.22 146,188 576 271.55 81% 84% 93% 93% Matz's Ruby 174.82 126,732 576 573.54 69% 83% 86% 90% reverse-complement source secs mem gz busy cpu load Ruby 11.45 2,009,880 482 21.50 34% 70% 57% 26% Matz's Ruby 21.02 492,816 264 21.29 1% 0% 100% 0% regex-redux source secs mem gz busy cpu load Ruby 9.66 277,900 724 21.43 44% 96% 38% 43% Matz's Ruby 15.83 223,648 724 36.10 94% 44% 44% 46% pidigits source secs mem gz busy cpu load Ruby 2.11 541,320 485 4.16 53% 6% 100% 39% Matz's Ruby Failed Ruby ruby 3.0.0p0 (2020-12-25 revision 95aff21468) [x86_64-linux] --jit Matz's Ruby ruby 1.8.7 (2008-08-11 patchlevel 72) [x86_64-linux] So old that I haven't been able to get rubygems to work no backport, no gmp all Ruby programs & measurements all Matz's Ruby programs & measurements