TalkChess.com

You can enable "Verbose Move Stats" in engine options (or through command line, --verbose-move-stats).

It's at least intended that roundrobin works the best when all GPUs are the same, and for multiplexing it's not that strict.

I wrote a blog post about backends recently: http://blog.lczero.org/2019/04/backend- ... l?m=1#more

In the LC0 folder there's a file CUDNN.txt that contains the "NVIDIA SOFTWARE LICENSE AGREEMENT". If I put this agreement in with the Chess Fighter download is that sufficient? Yes, it should be enough. Lc0 had also the problem from the opposite side, GPL by default doesn't allow bundling closed so...

I think that you would prefer to call the policy network move an N node search, where N is the number of legal moves, since all of these moves have been given a probability. But the actual node reached by each move has not been visited, so technically I think they are correct. Policy head performs ...

I'd also add the following points which are often brought up (sorted from most surprising to least surprising): - No SLI bridge is needed (or useful at all) when using multiple GPUs. - For RTX cards, default Leela configuration is much slower because cudnn backend is default instead of cudnn-fp16. -...

1. T51 doesnt start from total random play ( they feed 100k game data from T50). Similarily, T40 start from seeded data from T35 and perform head to head vs 3200+ rated Laser within 2 weeks ( in tcec bonus tests) Test51 started from random play, no test50 games were used. Test51 did use the same ra...

It is using mcts search in tcec. The alphabeta version has not been released yet. And the engine playing in TCEC is officially “Allie+Stein” as it is using Mark’s excellent SL “Stein” network. Without that, Allie could not play chess. Just want to note, that it's not correct to say that "Stein" is ...

How can I register there? I was in a mood to contribute a few times, but couldn't find a way. Also there are some mistypes here and there to fix, e.g. "The diagonals are the north-east directed, distinguished from the anti-diagonals in north-east directions." in https://www.chessprogramming.org/Ches...

dkappe wrote: ↑

Wed Mar 06, 2019 6:43 pm

I don’t take a stand on whether it is right or wrong, but Allie doesn’t use lc0 code. Also, the network uses lots of non-leela training data.

Allie uses Lc0 NN backends code: https://github.com/manyoso/allie/tree/master/lib/neural

Was a set of input planes representing a position at time (t - T + 1) oriented for the side to move at time t or time (t - T + 1)? All the history input planes are oriented for the current side to move, i.e. at time t. Would it matter if they weren’t? (That’s a serious question btw) In LcZero they ...