That makes sense. However if the classical computer were strategically directed to the corners of the [search] space that are interesting, maybe interesting res

That makes sense. However if the classical computer were strategically directed to the corners of the [search] space that are interesting, maybe interesting results could arise. It would be hard to have any sort of confidence in those I suppose if it must ignore large swaths of the space, however. It may serve as a discovery process informing more refined experiments though. Also, as I mentioned to another replier, I didn't say which type of computer must be used-- it should be possible with a quantum computer running software rewriting the NP-hard problem in terms most natural [i.e., native] to it. :-)