Currently there is no such thing as "cloning" in yices. This means that---whenever we fork the execution---one of the two branches will use a fresh solver, that does not benefit from incremental solving. If such branch is the only one that's NOT pruned/reverted, then the choice of giving it the fresh solver was sub-optimal, as we throw away incremental solving for no reason.
NOTE: keep in mind that this is only meaningful as long as yices devs don't implement a cloning mechanism.
Currently there is no such thing as "cloning" in yices. This means that---whenever we fork the execution---one of the two branches will use a fresh solver, that does not benefit from incremental solving. If such branch is the only one that's NOT pruned/reverted, then the choice of giving it the fresh solver was sub-optimal, as we throw away incremental solving for no reason.
NOTE: keep in mind that this is only meaningful as long as yices devs don't implement a cloning mechanism.