An experiment to see the differences when trying to solve a simple problem, adding one to all numbers given as arguments, with Java and Rust.
One obvious difference is the amount of code needed:
$ tokei -s code -e benches .
# curated output
--------------------------------------------------------------------------------------------------
Language Files Total Blanks Comments Code
--------------------------------------------------------------------------------------------------
Java 3 53 11 0 42
Rust 2 32 5 0 27
--------------------------------------------------------------------------------------------------Another one might be that Rust clearly is less wordy, denser, and possibly easier to grasp.
In any case, rust is doing the work far more efficiently.
Simple benchmarking is easy to do in Rust, yet I was unable to get it to work using JMH in IntelliJ after an hour. It certainly doesn't mean much, except for showing that such a task is not too intuitive and easily done with my java toolchain. Because of that, the Rust micro-benchmark stands without comparison.