benchmark results for hello world, pls ensure that terminal itself is not the bottleneck, by redirecting output to a non-busy file

main.py

#!/bin/sh

# Results
# 
# Macbook M1 Pro 16GB RAM
# Chip: M1 Pro
# 
# best of 5: 154 nsec per loop (+/- 10)
#
# lines printed: `wc -l`
# :: 13888889 (yep not so slow are we now haha!)
python3 -m timeit "print('Hello, World!')"

main.rs

#![feature(test)]
extern crate test;
use test::Bencher;

#[bench]
/// [[benchmark-results]]
/// 
/// Macbook M1 Pro 16GB RAM
/// Chip: M1 Pro
/// 
/// 1,342 ns/iter (+/- 150)
/// 
/// lines printed: `wc -l`
/// :: 288007 (+/- 50000)
fn hello_world(b: &mut Bencher) {
    b.iter(move || {
        // With println! bench and test run without output capture
        // this inflates the result of benchmarking
        // ensure the use of, `--nocapture`
        // ```sh
        // cargo +nightly bench -Z unstable-options -- --nocapture
        // ```
        println!("Hello, World!");
    });
}

#[bench]
/// [[benchmark-results]]
/// 
/// Macbook M1 Pro 16GB RAM
/// Chip: M1 Pro
/// 
/// 13 ns/iter (+/- 0)
/// 
/// lines printed: `wc -l`
/// :: 460836007 (does it even matter? it's a 3+ GB file)
fn better_hello_world(b: &mut Bencher) { 
    use std::io::Write;
    let stdout = std::io::stdout();
    let lock = stdout.lock();
    let mut out = std::io::BufWriter::new(lock);
    b.iter(|| {
        test::black_box(write!(out, "Hello, World!\n"));
    });
}