I'm no expert on these things, but shouldn't you be testing these birds with 100% fuel and in the manual-specified takeoff configuration?
Besides, measuring the takeoff in "seconds before hitting the crossing" seems inherently flawed unless the takeoff speeds are identical.
What you could do would be to run the recorder, set everything up according to hoyle, and at the moment the vsi goes positive ROC, freeze frame, go outside top-down view and get a screenshot of the plane's shadow on the RW, then compare results.
[This message has been edited by Dinger (edited 03-16-2000).]