>>poorly selected operation to do in a benchmark. Considering the history of benchmarks, it is quite likely the opposite: The library was selected very careful

>>poorly selected operation to do in a benchmark. Considering the history of benchmarks, it is quite likely the opposite: The library was selected very carefully, to argue a "point". :-(