I wouldn't trust that article, then, either. Here's some contradictory evidence: http://www.centralhobbies.com/Engine...s/hydetxt.html If you want to experiment yourself, set up a test. It's not hard to do. Move four or so servos around without a real load on them, then move them around after hanging weights on the servo arms. Time out the difference. A servo simply works harder when it has to use greater force to attain or maintain a position. Hence, the advice you gave about calculating run times by moving the servos around without any appreciable load on them really isn't very good advice.