bfcpp

> https://github.com/camel-cdr/bfcpp/blob/main/TUTORIAL.md#32-... > loops like [-], [>] and [<] could even be completely reduced to a single operation. This has got me wondering if it would make sense to pre-process the BF programs I've been using in genetic programming experiments. Loops are the #1 thing that slow down the search. But, I don't know that this optimization pass would be paid for in the typical case - at least not until the average complexity of the population reaches a certain point.

The explanation document is interesting: https://github.com/camel-cdr/bfcpp/blob/main/TUTORIAL.md

For anybody interested in how this works I've written a bit of a tutorial, where I try to explain the code base: https://github.com/camel-cdr/bfcpp/blob/main/TUTORIAL.md Also, if you want to play around with this online, here is the code in godbolt: https://godbolt.org/z/oYM8eW6sj