I may have missed the part where you explained why qualia can't fit into a state machine-model of the universe.

I think Mitchell's point is that we don't know whether state-machines have qualia, and the costs of making assumptions could be large.