Wow, what a great set of tips. This is also a great guide for benchmarking any software, so long as its execution is highly deterministic like compression algor

Wow, what a great set of tips. This is also a great guide for benchmarking any software, so long as its execution is highly deterministic like compression algorithms are. This will probably be my goto reference when someone shows me a benchmark with bad methodology. If nothing else it should give him/her an appreciation of the multitude of things that can affect a benchmark result.