Below are 2 theories of mine. #1 Absolute rotation changes : rotate solely with energy, not thrust XY, Z is the depth - toward / away from you. The centre of mass is in the picture's centre. The rotation we are talking about is the whole picture around Z. Now you start spinning the wheels with the same directions around Z. The inner rotation doesn't change the pictures's rotation much, but the outer does. Additionally you have the option to flip them 180° around the X-axis (from picture). One clockwise, one counter-clockwise to preserve the Z-axis vector.This should give a quick burst, turning around Z. (if you can't imagine that, leave the "Nerd out Zone" ) #2 Inertia Drive Again these 2 wheels! But now, you turn left one in opposite direction (mid=both-up, left/right=both-down). They don't move along Z (depth axis) while you start spinning them. According to the theory that the more you approach the speed of light, the higher your mass gets, these wheels should have increased mass now. At least while you start moving them along Z. While moving them along Z, you stop spinning, reducing their mass to normal. v_total = sqrt( v_aroundXY + v_alongZ ) When you now also stop moving them along Z, you should require less force than when you start their movement while they were spinning. I would be happy if somebody can explain me what I am missing, if I am missing something. It's a bit difficult to build an experiment IRL, because of the time you have to stop spinning and the air-resistances on earth. And games don't have accurate enough physics to allow such experiments while dedicated software often lacks visual feedback. Remembering a BUG in KSB (Kerbal Space Program): You could build 2 fuel tanks connected with a pipe, start spinning and transfer fuel between tanks to change the inertia vector of it. The transfer didn't affect the object which has these fuel tanks.