Alain Connes has written down a wonderful generalisation of geometry in terms of non-commutative algebras. In general these algebras are infinite dimensional, which makes them hard to work with numerically. To do so anyway we can consider finite dimensional matrix algebras. This allows us to put the system on the computer, in the form of complicated matrix models. Why are they complicated? Because we have to implement different constraints on the Dirac operator to obtain the correct spin structure on the space. Simulating this system leads to interesting hints of two dimensional physics.