kayabaNerve/full-chain-membership-proofs

Abstract the constraint system over some underlying proof system

Opened this issue · 0 comments

The current AC code generates WL WR WO WV c values, yet is tightly bound to BP+. Ideally, the high-level AC folder becomes its own crate and solely calls out to BP+. In the future, this would let us move to BP++ for a 3.6% reduction in proof size, or to whatever other proofs come along.

I'm not too concerned about this since we can always just hard-code if it uses BP or BP+ or BP++, purging the old code. It wouldn't be the end of the world. Solely not optimal design commentary.