This article highlights how small modifications to either the source code of a benchmark program or the compilation options may impact its behavior on a specific machine. It argues that for evaluating machines, benchmark providers and users be careful to ensure reproducibility of results based on the machine code actually running on the hardware and not just source code. The article uses color to grayscale conversion of digital images as a running example.
@article{arxiv.1309.0534,
title = {Machines are benchmarked by code, not algorithms},
author = {Raphael 'kena' Poss},
journal= {arXiv preprint arXiv:1309.0534},
year = {2013}
}