4 ms·
Weirdly enough, I've studying this kind of thing lately. So it seems from the various papers Z3 is oriented around infinite precision real arithmetic. It's tha
by joe_the_user 13y ago
Weirdly enough, I've studying this kind of thing lately.
So it seems from the various papers Z3 is oriented around infinite precision real arithmetic. It's that kind of overkill for most practical problems?
How does it compare with iSat or similar approaches in speed? One of the problems with SMT, Satisfiability Modulo Theories, is that it involves a somewhat loose connection between a SAT problem and an underlying theory prover, which can slow down the whole process if the underlying theory prover is not very fast.
http://en.wikipedia.org/wiki/Satisfiability_Modulo_Theories http://en.wikipedia.org/wiki/Satisfiability_Modulo_Theories
- colanderman 13y agoZ3 supports many more theories than just real arithmetic. I've primarily dabbled with difference logic and bit vectors, both of which I believe benefit strongly from SMT's SAT underpinnings.