Search found 243 matches

by jorose
Sat Sep 14, 2019 4:06 pm
Forum: Computer Chess Club: General Topics
Topic: My failed attempt to change TCEC NN clone rules
Replies: 37
Views: 2126

Re: My failed attempt to change TCEC NN clone rules

I like that you differentiated more than the original rules, I think they could be improved. What do you mean by "weights format"? If you are referring to the actual format of the external weights file then I don't feel this should be included as I feel it should be encouraged to reuse that. Having ...
by jorose
Sat Sep 07, 2019 8:15 pm
Forum: Computer Chess Club: General Topics
Topic: The Price of Contempt and Armageddon
Replies: 2
Views: 306

The Price of Contempt and Armageddon

I recently implemented contempt in Winter, which is probably the last major feature I will be adding before releasing Winter 0.7. Unfortunately I have unexpectedly measured a regression in Winter with my implementation at 0 contempt. The good news is I think I understand what might be going on, so I...
by jorose
Thu Sep 05, 2019 3:44 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: New Tool for Tuning with Skopt
Replies: 60
Views: 4666

Re: New Tool for Tuning with Skopt

thomasahle, Very nice tool! Thx a lot for sharing it. jorose, Just wanted to mention I was successfully able to retune Winter's futility and static null move pruning margins with this tool. I measured a self play rating gain of +9 +- 3 Elo roughly. How many games did you have to run during tunning ...
by jorose
Mon Sep 02, 2019 3:07 pm
Forum: Computer Chess Club: General Topics
Topic: SIMEX 2.1
Replies: 26
Views: 2151

Re: SIMEX 2.1

Does Winter already support such a tool? What do I need to add to an engine in order to support a tool like SIMEX? I'm kind of curious what engines Winter (especially Winter dev) is related to :D EDIT: I'm trying things out in WINE now. I get the following error at the moment: Can't recognize 'sort ...
by jorose
Fri Aug 30, 2019 5:08 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: New Tool for Tuning with Skopt
Replies: 60
Views: 4666

Re: New Tool for Tuning with Skopt

Just wanted to mention I was successfully able to retune Winter's futility and static null move pruning margins with this tool. I measured a self play rating gain of +9 +- 3 Elo roughly.
by jorose
Tue Aug 27, 2019 8:54 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: New Tool for Tuning with Skopt
Replies: 60
Views: 4666

Re: New Tool for Tuning with Skopt

The game is limited to 160 plies. for ply in range(int(board.turn == chess.BLACK), 160): Good catch! I completely forgot I added that. I will change this to an optional user defined value, and otherwise just play till the fifty-moves rule decides it's time. Aside from draws getting recorded correct...
by jorose
Tue Aug 27, 2019 6:18 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: New Tool for Tuning with Skopt
Replies: 60
Views: 4666

Re: New Tool for Tuning with Skopt

So I am trying out your skript, but unfortunately it seems to not be counting threefold repetitions as draws. Is there an easy way to fix this issue?

EDIT: Ok I just checked the games, the issue is different than I thought. Games are adjudicating as undecided after just 85 moves?
by jorose
Sun Aug 25, 2019 2:28 pm
Forum: Computer Chess Club: General Topics
Topic: 29 Alpha-Beta's against Leela...
Replies: 62
Views: 8289

Re: 29 Alpha-Beta's against Leela...

Yes but they run around without any plan, They do enormous amounts of lines of play but have no clue what to do, That’s the reason the draw rate is increasing and increasing. Also most programs are clones of other programs, they make so many draws. Where is it good for ?? What is the sense of all t...
by jorose
Sun Aug 25, 2019 1:27 pm
Forum: Computer Chess Club: General Topics
Topic: 29 Alpha-Beta's against Leela...
Replies: 62
Views: 8289

Re: 29 Alpha-Beta's against Leela...

These are three different main lines, and they lead to three different games. I think 24... c3 leads the fastest to the goal ( = mobilisation of the rooks as quickly as possible) and is therefore clearly the best move in this position. In addition, it is the most interesting move here - and typical...
by jorose
Sun Aug 25, 2019 1:10 am
Forum: Computer Chess Club: General Topics
Topic: 29 Alpha-Beta's against Leela...
Replies: 62
Views: 8289

Re: 29 Alpha-Beta's against Leela...

According to Stockfish-dev, c3 and Rec8 are equivalent by transposition: Don't think so. I have received the following output in 3-variant mode (AMD Ryzen7-2700X - 16'384 MB): Analysis by Stockfish 280719 64 POPCNT 16CPU 1. -+ (-2.92): 24...c3 25.Sa3 Tec8 26.Sc2 Sb2 27.a3 h5 28.f3 hxg4 29.hxg4 Sc4 ...