Also, please study up on the design decisions made by the various authors. By understanding the problems the authors were trying to solve, you will find test situations that should favor each particular tool. Running one round of tests against a flat unfiltered network does not represent what professional testers see today.
I meant for the testing I've done at http://loquens-caesu.blogspot.com/
to encourage others to begin testing for themselves. End users are the ultimate judge of what works for their environment. Don't take my word for it. Set up a similar environment. Use the options I did. You will find similar results.
If you're going to go through the effort to run a test, at least try to understand the tools you are evaluating. Otherwise you come off favoring a car with automatic transmission because you don't understand the efficiency gains of stick shift.