TalkChess.com

Razoring was bad in Winter. That being said, I have not tried it in a very long time and Winter has undergone some huge changes since then, so perhaps I should try it again.

I enjoyed Winter's dynamic positional play in this game :) [Event "77th Amateur D3"] [Site "ChessGUI"] [Date "2019.12.03"] [Round "28.3"] [White "Winter 0.7 64-bit"] [Black "SmarThink 1.98 64-bit"] [Result "1-0"] [Time "8:54:20 a.m."] [ECO "D10"] [Opening "Exchange Variation, Slav"] [TimeControl "40...

By the way, do we know for sure that the Gusev game is not in the Fat Fritz training dataset? Its from a high level human game, so if CB uses human games in its training it is not impossible? Aside from that I like it as an example. It is a situation where I would expect CNNs to excell at, as they s...

Thanks for this. Will see if I can provide comments and something interesting. Thank you for your interest! I don't think I would have published this had not several people repeatedly asked me to. The interest in Winter has generally been very motivating and makes me very happy. Where can i downloa...

I have finally found the time and motivation to publish my NN training script (actually training repository) for Winter! Keep in mind that I mostly wrote this for my own personal use,so have mercy on my coding practices... That being said, I added some comments and it should be reasonably easy to un...

May I have an account for Winter?

Back when I was compiling Winter on Windows I used a mingw gcc compiler and it was fairly straightforward, but I failed when trying to compile with clang. I don't know much about C++ on Windows, so I am sure someone else might be of more use. Longterm I would like to support cmake, but I have never ...

I am not sure why you guys are trying to reach conclusions based on individual online games. There are so many things going on that will massively shift how such games go and I don't think I need to give a lecture on small sample sizes or anecdotal evidence to you guys. Also I am curious about the a...

I honestly also think NCM should consider updating the benchmark engine after SF11 release. Elo model is not particularly good for engines outside a certain rating range and at roughly +240 Elo, I think we might have reached that point already. For a 240+ Elo performance you need to score 80%.

No sense to buy fast GPU, when engine strength is +-20 to Stockfish. How much does a 10core CPU cost? Because if LC0 on that GPU plays +-20 to Stockfish, it might make sense to go for the GPU instead. I only checked Newegg, which I have heard isn't close to the cheapest nowadays. You can get an 8co...