Because they play Go instead of chess.

Moderator: Ras
AgreeGraham Banks wrote: ↑Sun Jun 02, 2019 11:56 pm Why are super top professional Go players so stupid?
Because they play Go instead of chess.![]()
I would honestly hope that chess players are the most intelligent humans (apart of course from top talmudic scholars).
I'll take that back the day you beat him in an over the board match without computer access. Shouldn't be a problem for you to cut through the Go world where the top professional players are so stupid, right?
Might be an idea for you to take that non-self-evident 'truth' and shove it up your south facing Jewish orifice ..
Of course a bigger net is going to outperform a smaller net at equal node count.Laskos wrote: ↑Sun Jun 02, 2019 9:42 pmDepends on hardware balance. They don't seem to use CUDA (cuDNN) anywhere, only OpenCL, my speeds are not impressive on my RTX 2070. Might be due to batch sizes used not helping much with cuDNN, I don't know. But Leela's 40x256 latest net beats conclusively FaceBook latest 20x256 net at fixed nodes. Not sure at fixed time, especially LTC (I expect 40b to scale better than 20b).
No, it's not quantity vs quality. It's quantity of one kind versus quantity of a different kind.Raphexon wrote: ↑Mon Jun 03, 2019 7:24 am Of course a bigger net is going to outperform a smaller net at equal node count.
Smaller vs bigger is an issue of quantity (nps) vs quality. And if you remove the variable of quantity then of course the net that's optimized for quality is going to reign supreme.
I don't think "stupid" is the right word to use in this context. Imagine that you and me play a game of chess, some blitz game, and things get very intense, and you are running out of time, I check you with my queen and you move your king and I move her to a square where I can promote a pawn and I beat you.