you wrote

“Of course, one can also wonder whether the physical world might provide computational resources even beyond quantum computing”.

On that note, is there some formal theory about the computational power of “analog computers”, where the inputs and states wouldn’t be represented and manipulated as bits but directly mapped to physical values, like position?

The difficulty being that error handling has to be intrinsic to the theory (a bit like in QC).

Or is it mostly a matter of saying that all the values are encoded in “unary notation” (e.g. making subset sum look polynomial)?

]]>