You don't have to worry about randomness. Just searching a few extra nodes here and there will completely prevent duplicate games anyway, so that's not an issue. But randomly searching deeper makes no sense. You should do it for a reason, or else save the time for when you might need it more. If something convinces you that the tree in the current position is somehow unstable, and that a deeper search might get closer to the truth, that's where to search deeper.mainsworthy wrote:why Im thinking about this quarter ply , isnt because of one search, its because of average time per 20-40 moves , one quarter are power searches and the rest normal searches. so if you did all power or deeper searches, they wouldnt be enogh time on the clock, but there is time to do some deeper searches.
robert I think random 75/25, because your engine will be more unprodictable, and interesting without weakening it, infact it strengthens it(maybe).
thankyou Miguel I can see you know your stuff.
Doing something at a random point is almost always the worst algorithm one would think about, because such an algorithm is trivial to do. If you get a super-duper idea that doesn't beat random, then use random. But there must be better ways than that, which is my point.
In demand page replacement algorithms, we talk about "optimal replacement" which is an algorithm that requires future information impossible to obtain, and we talk about "random replacement" which is the worst algorithm we would even consider since it is so simple. And we use "random" to provide the worst-case scenario that other algorithms are compared against (and hopefully beat).
