This method applies external torques of various kinds to 108 to 512 molecules in a supercomputer simulation. It is fully described in the Omnia Opera papers on www.aias.us. Instead of using the complicated Euler angles, the torque is applied to unit vectors. I will look up the method and I think that it can be applied to the gyroscope. In the Evans Pelkie animation on youtube the torque is applied to 108 chiral molecules, fluorochlorobromomethane, both enantiomers and racemic mixture. I pioneered the method in the late seventies at the EDCL, and produced the first computer simulation of the Langevin function. At Univ. Zurich and ETH I produced the first computer simulation of the related Kielich functions. These simulations were carried on on the CDC 7600 and IBM 3084 and 3090 supercomputers at Aberystwyth, Bangor, Swansea, IBM Kingston, Cornell Theory Center, Univ. Zurich and ETH Zurich. All the code is on www.aias.us and is all archived on www.archive.org and www.webarchive.org.uk .Desktops are now approaching the power of those supercomputers. The chiral molecule is a kind of freely rotating gyroscope. In one simulation and early animation I could see the motion of molecules in a gas, with rare collisions. I am trying to find a way of putting that animation on www.aias.us or youtube. It is one of the first animations ever made (at IBM Kingston, Clementi environment where I was an IBM professor). I once ran an IBM 3084 for about a week solid, with interesting results.