Comment by lifeinthevoid

> Well in fact we know (if P! = NP) that for any randomized ALG there's a good deterministic one

Oh, that's cool, do you have a reference for that?