like, we know what should be at 0 and what should be at PC, <some constant> and 99, respectively

and both are symbolic expressions and they both depend on only two unknowns

and we can reduce the expressions on each simulation step so the expressions are never too big (or is that a bad idea? would that block some laziness based optimizations?) so in theory the result should be neatly representable as a 2x2 matrix.......

Rain π±@grainloom@cybre.spaceAdvent of Code day 2 vague spoilers i guess?

ugh, that's more than enough for today.

don't do Haskell, kids.