Open metal-pony opened 2 months ago
The reduction matrix could be built into the regular data structures used to track the sieve items. No need to build it from scratch each time it's needed. I.e., tracking items by cell number instead of m-value.
Similarly look at the internal puzzle tracking.
The new caching could also allow faster testing for derivative puzzles if categorized by cell index.
The reduction matrix could be built into the regular data structures used to track the sieve items. No need to build it from scratch each time it's needed. I.e., tracking items by cell number instead of m-value.
Similarly look at the internal puzzle tracking.