Interpretation of torque and rotational impulse vectors

I am having trouble understanding how to interpret the single vector argument supplied to llSetTorque() and llApplyRotationalImpulse() - my understanding (such as it is) of 'torque' is that it is the effect of a force applied at some distance from a fulcrum, equal to the cross product of the force vector and the vector from the axis of rotation to the point where the force is acting. Is it this cross-product value that the functions are expecting? And what, actually, is the axis of rotation of a prim? Apologies if this is a silly question, but I am having difficulty visualising this, for some reason.

OK, I see that this thread has the lowest view count of any in the forum (apart from one labelled 'Sorry, Wrong Forum'). I knew I should have given it a snappier title

Jigsaw,

I can't explain it to you, but I know a man who can.

Try the LSL Wiki, specifically: torque and physics

hth
Leon

PS. Also have a look at the Vehicle page.

Thanks Tom - I did look at the entries for torque and physics generally, but they were a bit vague as to precisely how one should calculate the value of a rotational impulse (or indeed a torque) in order to achieve some desired effect. Vehicles appear only to have an 'angular motor' which accelerates to a given rotational speed (in radians/sec) about the object's natural axis over some specified time period, which isn't really equivalent to a torque. I guess I will have to adopt Newton's original approach, and conduct some physical experiments in-world

Jigsaw,

From what I gather on the forums, the accuracy is a little off as well ... especially propagation errors? Think it will be a little trail and error in any case?

Leon

PS. I haven't as yet played with any of this, so ymmv.

A prim always rotates around its centre (although prim torture can make that appear to be elsewhere SL still treats it as if the whole prim is there, but chunks of it are ghosted.

The vector that you give is the torque around each axis - I assume in Nm which is the SI unit of torque, but in SL who can tell! - so a torque of <0,0,1> will make it spin around the Z-axis in a counterclockwise direction <1,0,0> would do the same but around the X axis, <50,50,50> is 50 times faster around each axis - it would tumble like a mad thing.

There's also a local value - is the torque applied to the local axis of the prim or not but I assume that's OK?

Does this help?

Thanks Eloise, yes I think it does help - so I should stop trying to equate it with some sort of real-world mechanical torque, and just treat it as a magical 'spin' force or impulse, with magnitude proportional to the vector magnitude, and axis of rotation as defined by the vector 'pointing' angle. Hmmm.

*throws away copy of Elementary Mechanics Volume I and strides off into strange new world ...

Hey, this was the first thread I clicked on when browsing this forum! Very eye-catching. Sorry I can't help, though.