This really doesn't tell you much at all, because: - Only one program is tested, exercising a very small portion of each language (and its built-in APIs). - T

This really doesn't tell you much at all, because: - Only one program is tested, exercising a very small portion of each language (and its built-in APIs). - The runtime is too small; less than a second in both cases. I'd really need to see examples where the benchmark has been run in a loop multiple times, preferably with a warmup run beforehand, and an execution time measured in tens of seconds. At least the author made the effort to figure out the reasons behind the differences (based on the hypothesis that Rust would be faster), so credit is due there.