Testing depend largely on good Openings

Discussion of anything and everything relating to chess playing software and machines.

Moderator: Ras

Chessqueen
Posts: 5685
Joined: Wed Sep 05, 2018 2:16 am
Location: Moving
Full name: Jorge Picado

Re: Testing depend largely on good Openings

Post by Chessqueen »

lkaufman wrote: Sat Jul 09, 2022 5:39 pm
Ovyron wrote: Sat Jul 09, 2022 8:43 am
lkaufman wrote: Fri Jul 08, 2022 5:49 am the path to a draw is too clear now.
No, the path to draw isn't clear at all. Try making a "balanced" book, one where no matter what the opponent plays, you'll reach a balanced opening that you will be able to draw.

Well, I know people that would be able to bust one of those lines, and prove the "draw that was clear" wasn't so at all. It's just that testers of engines never try to find a hole on those "balanced" openings, so if there's a move that gives a side some big advantage, it'll not be found, because nobody is looking for it.

This is specially true on the age of NNUE, where you can build a big book on an off-beat line and win just because your side has a lot more time on the clock than the opponent missing your lines. Computer chess on the competitive side is still live and kicking, if anybody tried to use those balanced openings on playchess they'd be utterly destroyed by better moves by people that dedicate their lives to finding holes in lines.

I've seen positions where Stockfish NNUE has the score backwards, both engines show some +0.20 but after the game continues both engines agree it's really -0.40, so a black player wants to get into this line, while engine testers will be happy to settle to play some +0.15 position that is actually 0.00 missing black's improvement.

I've seen completely lost positions that are scored at +0.30, but there's already a sequence of moves that force some +2.00 advantage, so white looks great aiming for this line in case the opponent allows it and loses. An engine tester already knows what moves black will play so has no reason to try this.

What engine testing is missing is competition, trying to do the best that you can for your side causes you to find and play better openings. Using balanced openings with reversed colors leads to draw death, because there's no incentive to find better lines for either side.
Have people tried entering latest Stockfish on good hardware in Rapid (not blitz) online tournaments with NO BOOK? If so, does it lose any games, and if so what percent? Or perhaps same with just a very short generic book from GM games, like 5 or 6 moves only?
My opinion is that Stockfish or Komodo performs better with short Opening book of no more than 6 moves than with a very long generic book from GM games of 16 moves, because some of those long opening books have bad lines like in the case that I have posted several times, and if you let Komodo or Stockfish analyze the position after the bad move was chosen by the Opening book it could make a difference of minus 0.42 or +0.41 which will determine weather White will draw or lose . :roll:

Position analyzed after the opening move 7. Nd2? which favor Black, and can win easy after several moves, if the Black side is the strongest engine like Stockfish 15 :shock:
[fen]rnbq1rk1/ppp1ppbp/5np1/8/2pP4/5NP1/PP1NPPBP/R1BQ1RK1 b - - 1 1 [/fen]
FEN: rnbq1rk1/ppp1ppbp/5np1/8/2pP4/5NP1/PP1NPPBP/R1BQ1RK1 b - - 1 1

Dragon-2.6.1-64bit-avx2:
1 00:00 21,467k 21,466,862k -1.07 Nf6-d5
1 00:00 21,468k 21,467,507k -0.67 Nb8-c6 Nd2xc4 a7-a5
1 00:00 117k 9,784k -0.59 Qd8-d5
1 00:00 118k 9,750k -0.22 Bc8-e6 Nf3-g5
1 00:00 118k 9,731k +0.44 b7-b5
2 00:00 119k 9,697k +0.44 b7-b5 b2-b3
3 00:00 120k 9,674k +0.44 b7-b5 b2-b3 c4-c3
4 00:00 122k 9,609k +0.60 b7-b5 Qd1-c2 Nf6-d5 a2-a4 c7-c6
5- 00:00 123k 9,574k +0.48 b7-b5 b2-b3
5 00:00 124k 9,550k +0.47 b7-b5 b2-b3 c4-c3 Nd2-b1 Nf6-d5
6 00:00 125k 9,517k +0.47 b7-b5 b2-b3 c4-c3 Nd2-b1 Nf6-d5 e2-e4
7 00:00 127k 9,483k +0.46 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5
8 00:00 129k 9,441k +0.46 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1
9 00:00 134k 9,366k +0.44 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5
10 00:00 143k 9,072k +0.44 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Qd8-d5+ Qd1-f3 Qd5xf3+ Ne1xf3 Nf6-d5 Bc1-e3 c3-c2
11 00:00 155k 8,769k +0.44 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Qd8-d5+ Qd1-f3 Qd5xf3+ Ne1xf3 Nf6-d5 Bc1-e3 c3-c2 Nb1-a3
12 00:00 172k 8,661k +0.44 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Qd8-d5+ Qd1-f3 Qd5xf3+ Ne1xf3 Nf6-d5 Bc1-e3 Rf8-e8 Rf1-c1 c3-c2
13 00:00 201k 8,197k +0.44 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Nf6-d5 Ne1-c2 a7-a5 Ra1xa5 Ra8xa5 b4xa5 Nb8-c6 Qd1-d3 Nc6xa5 Nb1xc3 Na5xb3
14 00:00 333k 8,184k +0.43 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Qd8-d5+ Qd1-f3 Qd5xf3+ Ne1xf3 Nf6-d5 Bc1-e3 Rf8-e8 Rf1-c1 Nb8-c6 b4-b5 Nc6-b4 Nb1xc3 Nd5xe3+ f2xe3 Re8xe3
15 00:00 672k 8,505k +0.43 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 Qd8-d5+ Qd1-f3 e5xd4 e3xd4 Qd5xf3+ Ne1xf3 Nf6-d5 Bc1-e3 Rf8-e8 Rf1-c1 Nb8-c6 b4-b5 Nc6-b4 Nb1xc3 Nd5xe3+ f2xe3 Re8xe3 Nc3-d1 Re3-e7
16 00:00 868k 8,471k +0.43 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 Qd8-d5+ Qd1-f3 e5xd4 e3xd4 Qd5xf3+ Ne1xf3 Nf6-d5 Bc1-e3 Rf8-e8 Rf1-c1 Nb8-c6 b4-b5 Nc6-b4 Nb1xc3 Nd5xe3+ f2xe3 Re8xe3 Nc3-d1 Re3-e7 Ra1-a4
17 00:00 1,309k 7,380k +0.43 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 Qd8-d5+ Qd1-f3 e5xd4 e3xd4 Qd5xf3+ Ne1xf3 Nf6-d5 Bc1-e3 Rf8-e8 Rf1-c1 Nb8-c6 b4-b5 Nc6-b4 Nb1xc3 Nd5xe3+ f2xe3 Re8xe3 Nc3-d1 Re3-e7 Ra1-a4 Bg7-h6 Rc1-c4
18 00:00 1,828k 7,303k +0.45 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Qd8-d5+ Qd1-f3 Qd5xf3+ Ne1xf3 Nf6-d5 Bc1-e3 Rf8-e8 Rf1-c1 Nd5xe3+ f2xe3 Re8xe3 b4-b5 c7-c5 Rc1xc3 Re3xc3 Nb1xc3 c5xd4 Nc3-d5 d4-d3 Ra1-d1 Nb8-d7 Rd1xd3 Nd7-c5 Nd5-e7+ Kg8-f8
19- 00:00 2,625k 6,943k +0.38 b7-b5 b2-b3
19+ 00:00 2,932k 6,789k +0.40 b7-b5 b2-b3 c4-c3
19 00:00 3,074k 6,720k +0.43 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Nb8-c6 Nb1xc3 Qd8xd4 Qd1xd4 Nc6xd4 Ra1-a6 Rf8-d8 h2-h3 Nd4xb3 Ne1-f3 Rd8-d3 Bc1-b2 Nf6-d5 Nc3xd5
20- 00:00 5,445k 6,171k +0.36 b7-b5 b2-b3
20+ 00:00 5,584k 6,142k +0.38 b7-b5 b2-b3 c4-c3
20 00:00 5,756k 6,119k +0.38 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Nb8-c6 Nb1xc3 Qd8xd4 Qd1xd4 Nc6xd4 Ra1-a6 Rf8-d8 h2-h3 Nf6-d5 Nc3xd5 Rd8xd5 Bc1-e3 Bg7-f8 Be3xd4 Rd5xd4 Ne1-c2 Rd4-d3
21 00:01 7,506k 6,029k +0.34 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Qd8-d5+ Qd1-f3 Qd5xf3+ Ne1xf3 Nf6-d5 Bc1-e3 Rf8-e8 Rf1-c1 Nd5xe3+ f2xe3 Re8xe3 b4-b5 c7-c5 Rc1xc3 Re3xc3 Nb1xc3 c5xd4 Nc3-d5
22+ 00:01 7,830k 6,013k +0.41 b7-b5 b2-b3 c4-c3
22 00:01 8,003k 5,996k +0.42 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Nb8-c6 Nb1xc3 Qd8xd4 Qd1xd4 Nc6xd4 Ra1-a6 Rf8-d8 h2-h3 Nf6-e8 Nc3-a4 Nd4xb3 Bc1-f4 Ne8-d6 Ne1-f3
23 00:01 8,274k 5,951k +0.48 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Nb8-c6 Nb1xc3 Qd8xd4 Qd1xd4 Nc6xd4 Ra1-a6 Rf8-d8 h2-h3 Nf6-e8 Nc3-a4 Nd4xb3 Bc1-f4 Bg7-f8 Ne1-c2 Nb3-d4 Nc2xd4
24- 00:01 9,587k 5,774k +0.41 b7-b5 b2-b3
24 00:02 11,672k 5,532k +0.42 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Nb8-c6 Nb1xc3 Qd8xd4 Qd1xd4 Nc6xd4 Ra1-a6 Rf8-d8 h2-h3 Ra8-b8 Ra6xa7 Nf6-e8 Nc3-e4 Rb8xb4 Ra7-a4 Rb4xb3 Bc1-g5 f7-f6 Bg5-e3 Nd4-f5
25- 00:03 16,550k 5,156k +0.35 b7-b5 b2-b3
25 00:04 20,853k 4,934k +0.34 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Nb8-c6 Nb1xc3 Qd8xd4 Qd1xd4 Nc6xd4 Ra1-a6 Rf8-d8 h2-h3 Ra8-b8 Bc1-e3 Nf6-e8 Nc3-a2 Nd4xb3 Ne1-f3 Nb3-d4 Ra6xa7 Nd4xf3 Kg2xf3 Bg7-d4 Be3xd4 Rd8xd4 Rf1-c1 Rd4-d2
26 00:05 26,369k 4,826k +0.30 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Nb8-c6 Nb1xc3 Qd8xd4 Qd1xd4 Nc6xd4 Ra1-a6 Rf8-d8 h2-h3 Ra8-b8 Bc1-e3 Nf6-e8 Nc3-a2 Nd4xb3 Ne1-f3 Nb3-d4 Nf3xd4 Bg7xd4 Be3xd4 Rd8xd4 Ra6xa7 Rd4-d3 Rf1-e1 Rd3-d2
27+ 00:05 28,391k 4,803k +0.37 b7-b5 b2-b3 c4-c3
27 00:06 32,021k 4,761k +0.31 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Nb8-c6 Nb1xc3 Qd8xd4 Qd1xd4 Nc6xd4 Ra1-a6 Rf8-d8 h2-h3 Ra8-b8 Bc1-e3 Nf6-e8 Nc3-a2 Ne8-d6 Ra6xa7 Nd6-b5 Ra7-a6 Nd4xb3 Ne1-f3 Nb3-d4 Nf3xd4 Bg7xd4 Be3xd4 Nb5xd4 Rf1-d1 Nd4-c2 Rd1xd8+ Rb8xd8
28+ 00:07 33,357k 4,748k +0.38 b7-b5 b2-b3 c4-c3
28 00:07 33,915k 4,739k +0.39 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Nb8-c6 Nb1xc3 Qd8xd4 Qd1xd4 Nc6xd4 Ra1-a6 Rf8-d8 h2-h3 Ra8-b8 Bc1-e3 Nf6-e8 Nc3-a2 Nd4xb3 Ne1-f3 Nb3-d4 Nf3xd4 Bg7xd4 Be3xd4 Rd8xd4 Ra6xa7 h7-h5 Rf1-e1
29+ 00:08 40,595k 4,580k +0.46 b7-b5 b2-b3 c4-c3
29- 00:09 42,197k 4,550k +0.39 b7-b5 b2-b3
29+ 00:15 64,206k 4,253k +0.42 b7-b5 b2-b3 c4-c3
29- 00:15 65,859k 4,237k +0.34 b7-b5 b2-b3
29+ 00:17 72,860k 4,172k +0.39 b7-b5 b2-b3 c4-c3
29 00:20 82,861k 4,117k +0.63 b7-b5 a2-a4 c7-c6 Nf3-e5 Nf6-d5 Nd2-b1 Bc8-b7 Nb1-c3 a7-a6 Nc3-e4 f7-f6 Ne5-f3 Nb8-d7 Qd1-c2 Qd8-c8 Rf1-d1 f6-f5 Ne4-g5 Nd7-f6 Nf3-e5 c6-c5 d4xc5 Qc8xc5 Ng5-e6 Qc5-d6 Ne6xf8 Qd6xe5 a4xb5 Kg8xf8 Qc2xc4
30- 00:31 124,738k 3,944k +0.56 b7-b5 b2-b3
30- 00:33 131,197k 3,933k +0.47 b7-b5 b2-b3
30 00:36 143,434k 3,917k +0.45 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Nb8-c6 Nb1xc3 Qd8xd4 Qd1xd4 Nc6xd4 Ra1-a6 Rf8-d8 h2-h3 Nf6-e8 Nc3-e4 Ra8-b8 Ra6xa7 Rb8xb4 Ne4-c5 Rb4-b5 Ne1-d3 Nd4xb3 Bc1-g5 f7-f6 Bg5-e3
31+ 00:38 150,495k 3,908k +0.52 b7-b5 b2-b3 c4-c3
31+ 00:40 158,928k 3,898k +0.61 b7-b5 b2-b3 c4-c3
31 00:40 159,595k 3,898k +0.57 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 c7-c5 Nf3-e1 Nf6-d5 d4xc5 Nd5-f4 Qd1xd8 Nf4xe2+ Kg1-h1 Rf8xd8 Bc1-e3 Rd8-d1 a3xb4 c3-c2 Ne1xc2 Rd1xf1+ Bg2xf1 Bc8-b7+ Bf1-g2 Bb7xg2+ Kh1xg2 Bg7xa1 Nc2xa1 Nb8-c6 Na1-c2 e7-e5 b4-b5 Nc6-d4 Nc2xd4 e5xd4 Be3-d2 Ra8-b8 Nb1-a3 d4-d3 c5-c6 Ne2-d4 Bd2-f4 Nd4xb5 Bf4xb8
32- 00:42 166,299k 3,884k +0.50 b7-b5 b2-b3
32- 00:54 209,905k 3,852k +0.41 b7-b5 b2-b3
32 01:14 284,653k 3,800k +0.36 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Nb8-c6 Nb1xc3 Qd8xd4 Qd1xd4 Nc6xd4 Ra1-a6 Rf8-d8 h2-h3 Nf6-e8 Nc3-e4 Ra8-b8 Ra6xa7 Rb8xb4 Ne1-d3 Rb4xb3 Ne4-c5 Rb3-b6 Nc5-b7 Nd4-b5 Nb7xd8 Nb5xa7
33+ 01:20 306,207k 3,793k +0.43 b7-b5 b2-b3 c4-c3
33 01:20 306,901k 3,792k +0.43 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Nb8-c6 Nb1xc3 Qd8xd4 Qd1xd4 Nc6xd4 Ra1-a6 Rf8-d8 h2-h3 Nf6-e8 Nc3-e4 Nd4xb3 Bc1-e3 Nb3-d4 Ne4-c5 Ne8-d6 Be3xd4 Bg7xd4 Ne1-f3 Nd6-b5 Rf1-d1
34- 01:47 402,907k 3,751k +0.36 b7-b5 b2-b3
34+ 01:54 428,794k 3,748k +0.38 b7-b5 b2-b3 c4-c3
34 01:56 437,538k 3,747k +0.40 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Nb8-c6 Nb1xc3 Qd8xd4 Qd1xd4 Nc6xd4 Ra1-a6 Rf8-d8 Bc1-g5 Nd4-e6 Ne1-f3 Ne6xg5 Nf3xg5 Nf6-e8 Nc3-a4 Bg7-f8 Na4-c5 Rd8-d4 Ng5-e4 Rd4xb4 Rf1-a1 Rb4-b5 Nc5-d7 Ra8-d8 Nd7xf8 Kg8xf8
35- 02:09 484,512k 3,735k +0.33 b7-b5 b2-b3
35+ 02:13 497,383k 3,732k +0.35 b7-b5 b2-b3 c4-c3
35 02:15 506,588k 3,730k +0.37 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Nb8-c6 Nb1xc3 Qd8xd4 Qd1xd4 Nc6xd4 Ra1-a6 Rf8-d8 h2-h4 Nf6-g4 Bc1-g5 Rd8-d7 Nc3-e4 Nd4xb3 Ne1-f3 h7-h6 Bg5-f4 Rd7-e7 Ne4-c5 Nb3xc5 b4xc5 Re7-e2 c5-c6 Ra8-c8 Ra6xa7 Re2-c2 Ra7-a6 Bg7-f8 Ra6-a4 Rc2xc6
36+ 02:23 534,795k 3,725k +0.44 b7-b5 b2-b3 c4-c3
36- 02:24 539,815k 3,724k +0.37 b7-b5 b2-b3
36+ 02:36 582,911k 3,716k +0.40 b7-b5 b2-b3 c4-c3
36- 03:05 686,314k 3,706k +0.32 b7-b5 b2-b3
36 04:19 959,704k 3,699k +0.28 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Nb8-c6 Nb1xc3 Qd8xd4 Qd1xd4 Nc6xd4 Ra1-a2 Rf8-d8 Bc1-f4 Nd4-e6 Bf4-e5 Nf6-d5 Be5xg7 Kg8xg7 Nc3xd5 Rd8xd5 Ra2-a4 Ra8-b8 h2-h4 h7-h6 Ne1-c2 Rd5-d2 Nc2-a3 Rd2-d3 b4-b5 Rd3xb3 Ra4xa7 Kg7-f6 Rf1-e1 g6-g5 h4xg5+ h6xg5 Re1-e3 Rb3xe3 f2xe3 Kf6-e7
37+ 04:35 1,019,907k 3,697k +0.35 b7-b5 b2-b3 c4-c3
37 04:38 1,030,939k 3,697k +0.36 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Nb8-c6 Nb1xc3 Qd8xd4 Qd1xd4 Nc6xd4 Ra1-a2 a7-a5 b4xa5 Nd4xb3 Ne1-f3 Ra8xa5 Ra2xa5 Nb3xa5 Bc1-f4 Rf8-c8 Rf1-b1 c7-c6 Nf3-e5 Rc8-e8 Rb1-a1 Na5-b3 Ra1-a3 Nb3-d4 Ra3-a6 Nf6-d5 Nc3xd5 Bg7xe5 Bf4xe5 Re8xe5 Nd5-b4 Re5-c5 Nb4-d3 Rc5-c2 Nd3-e5
38+ 05:03 1,123,024k 3,696k +0.43 b7-b5 b2-b3 c4-c3
38- 05:04 1,125,384k 3,695k +0.36 b7-b5 b2-b3
38+ 05:21 1,188,599k 3,695k +0.39 b7-b5 b2-b3 c4-c3
38 05:25 1,204,201k 3,695k +0.41 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Nb8-c6 Nb1xc3 Qd8xd4 Qd1xd4 Nc6xd4 Ra1-a2 a7-a5 b4xa5 Nd4xb3 Ne1-f3 Ra8xa5 Ra2xa5 Nb3xa5 Bc1-f4 Rf8-c8 Rf1-b1 c7-c6 Bf4-e5 Na5-c4 Be5xf6 Bg7xf6 Nc3-e4 Bf6-b2 Ne4-d2 Nc4xd2 Nf3xd2 Bb2-e5 Nd2-c4 Be5-d4 Rb1-b7 c6-c5 Rb7-d7 Rc8-c6 f2-f4 Rc6-a6 Kg2-h3 Ra6-c6 Rd7-a7 Rc6-e6 Ra7-c7
39- 06:19 1,400,661k 3,690k +0.34 b7-b5 b2-b3
39+ 06:40 1,479,550k 3,691k +0.36 b7-b5 b2-b3 c4-c3
39 06:45 1,495,066k 3,691k +0.39 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Nb8-c6 Nb1xc3 Qd8xd4 Qd1xd4 Nc6xd4 Ra1-a2 a7-a5 b4xa5 Nd4xb3 Ne1-f3 Ra8xa5 Ra2xa5 Nb3xa5 Bc1-g5 c7-c6 Bg5xf6 Bg7xf6 Nc3-e4 Kg8-g7 Rf1-b1 Rf8-e8 Rb1-b4 Bf6-e7 Rb4-a4 Be7-d8 g3-g4 Na5-b7 Ra4-c4 Kg7-f8 Ne4-c5 Nb7xc5 Rc4xc5 Re8-e4 Rc5xc6 Re4xg4+ Kg2-f1
40 07:05 1,568,281k 3,686k +0.40 b7-b5 b2-b3 c4-c3 Nd2-b1 b5-b4 a2-a3 Bc8-f5 Nf3-e1 Bf5-e4 a3xb4 Be4xg2 Kg1xg2 e7-e5 e2-e3 e5xd4 e3xd4 Nb8-c6 Nb1xc3 Qd8xd4 Qd1xd4 Nc6xd4 Ra1-a2 a7-a5 b4xa5 Nd4xb3 Ne1-f3 Ra8xa5 Ra2xa5 Nb3xa5 Bc1-g5 Rf8-c8 Rf1-b1 c7-c5 Nf3-e5 Rc8-a8 f2-f4 Nf6-g4 Rb1-b5 Ng4xe5 f4xe5 Na5-c6 Rb5xc5 Nc6xe5 Rc5-d5 Ra8-c8 Nc3-e2 f7-f6 Bg5-e3 Rc8-c2 Rd5-d8+ Kg8-f7
Last edited by Chessqueen on Sat Jul 09, 2022 6:16 pm, edited 10 times in total.
Chessqueen
Posts: 5685
Joined: Wed Sep 05, 2018 2:16 am
Location: Moving
Full name: Jorge Picado

Re: Testing depend largely on good Openings

Post by Chessqueen »

lkaufman wrote: Sat Jul 09, 2022 5:39 pm
Ovyron wrote: Sat Jul 09, 2022 8:43 am
lkaufman wrote: Fri Jul 08, 2022 5:49 am the path to a draw is too clear now.
No, the path to draw isn't clear at all. Try making a "balanced" book, one where no matter what the opponent plays, you'll reach a balanced opening that you will be able to draw.

Well, I know people that would be able to bust one of those lines, and prove the "draw that was clear" wasn't so at all. It's just that testers of engines never try to find a hole on those "balanced" openings, so if there's a move that gives a side some big advantage, it'll not be found, because nobody is looking for it.

This is specially true on the age of NNUE, where you can build a big book on an off-beat line and win just because your side has a lot more time on the clock than the opponent missing your lines. Computer chess on the competitive side is still live and kicking, if anybody tried to use those balanced openings on playchess they'd be utterly destroyed by better moves by people that dedicate their lives to finding holes in lines.

I've seen positions where Stockfish NNUE has the score backwards, both engines show some +0.20 but after the game continues both engines agree it's really -0.40, so a black player wants to get into this line, while engine testers will be happy to settle to play some +0.15 position that is actually 0.00 missing black's improvement.

I've seen completely lost positions that are scored at +0.30, but there's already a sequence of moves that force some +2.00 advantage, so white looks great aiming for this line in case the opponent allows it and loses. An engine tester already knows what moves black will play so has no reason to try this.

What engine testing is missing is competition, trying to do the best that you can for your side causes you to find and play better openings. Using balanced openings with reversed colors leads to draw death, because there's no incentive to find better lines for either side.
Have people tried entering latest Stockfish on good hardware in Rapid (not blitz) online tournaments with NO BOOK? If so, does it lose any games, and if so what percent? Or perhaps same with just a very short generic book from GM games, like 5 or 6 moves only?
My opinion is that Stockfish or Komodo performs better with short Opening book of no more than 6 moves than with a very long generic book from GM games of 16 moves, because some of those long opening books have bad lines like in the case that I have posted several times, and if you let Komodo or Stockfish analyze the position after the bad move was chosen by the Opening book it could make a difference of minus 0.42 or +0.41 which will determine weather White will draw or lose . :roll:


Position analyzed after Komodo Dragon chose the opening move 7.Na3! which does NOT favor Black, and is able to Draw against Stockfish 15
[fen]rnbq1rk1/ppp1ppbp/5np1/8/2pP4/N4NP1/PP2PPBP/R1BQ1RK1 b - - 1 1 [/fen]
FEN: rnbq1rk1/ppp1ppbp/5np1/8/2pP4/N4NP1/PP2PPBP/R1BQ1RK1 b - - 1 1

Dragon-2.6.1-64bit-avx2:
1 00:00 35,501k 35,501,320k -2.54 Nb8-a6
1 00:00 35,501k 35,501,340k -1.07 Nf6-d5
1 00:00 35,501k 35,501,375k -0.67 Nb8-c6 Na3xc4 a7-a5
1 00:00 6,258k 135,178k -0.40 a7-a5
2 00:00 6,259k 134,704k -0.40 a7-a5 Na3xc4
3 00:00 6,260k 134,211k -0.63 a7-a5 Na3xc4 a5-a4
3 00:00 6,260k 133,947k -0.56 Nb8-c6 Na3xc4 Bc8-e6 b2-b3 Nc6xd4 Nf3xd4
4 00:00 6,261k 133,148k -0.56 Nb8-c6 Na3xc4 Bc8-e6 b2-b3 Nc6xd4
4 00:00 6,262k 132,741k -0.43 a7-a5 Na3xc4 a5-a4 Bc1-f4
5 00:00 6,263k 132,046k -0.43 a7-a5 Na3xc4 a5-a4 Bc1-f4 Nb8-c6
6 00:00 6,264k 131,217k -0.43 a7-a5 Na3xc4 a5-a4 Bc1-f4 Nb8-c6 Ra1-c1
7 00:00 6,265k 130,602k -0.43 a7-a5 Na3xc4 a5-a4 Bc1-f4 Nb8-c6 Ra1-c1 Nf6-d5
8 00:00 6,267k 129,629k -0.43 a7-a5 Na3xc4 a5-a4 Bc1-f4 Nb8-c6 Ra1-c1 Nf6-d5 Bf4-d2
9 00:00 6,269k 128,655k -0.43 a7-a5 Na3xc4 a5-a4 Bc1-f4 Nb8-c6 Ra1-c1 Nf6-d5 Bf4-d2 a4-a3 Nc4xa3
10 00:00 6,271k 127,081k -0.43 a7-a5 Na3xc4 a5-a4 Bc1-f4 Nb8-c6 Ra1-c1 Nf6-d5 Bf4-d2 a4-a3 Nc4xa3 Nc6xd4
11 00:00 6,276k 124,198k -0.43 a7-a5 Na3xc4 a5-a4 Bc1-f4 Nb8-c6 Ra1-c1 Nf6-d5 Bf4-d2 a4-a3 Nc4xa3 Nc6xd4 Nf3xd4
12 00:00 6,282k 119,969k -0.43 a7-a5 Na3xc4 a5-a4 Bc1-f4 Nb8-c6 Ra1-c1 Nf6-d5 Bf4-d2 a4-a3 Nc4xa3 Nc6xd4 Nf3xd4 Bg7xd4
13 00:00 6,315k 97,456k -0.48 a7-a5 Na3xc4 a5-a4 Bc1-f4 Nb8-c6 Ra1-c1 Nf6-d5 Bf4-d2 a4-a3 Nc4xa3 Nc6xd4 Nf3xd4 Bg7xd4 Qd1-b3 Bc8-e6 Bd2-h6
14- 00:00 229k 2,574k -0.55 a7-a5 Na3xc4
14 00:00 404k 2,245k -0.53 Nb8-a6 Na3xc4 c7-c5 b2-b3 Bc8-f5 Bc1-b2 Bf5-e4 Nc4-d2 Be4-d5 Rf1-e1 c5xd4 Bb2xd4 Ra8-c8 Ra1-c1 Rc8xc1 Qd1xc1 Na6-b4 a2-a3
15+ 00:00 410k 2,255k -0.46 Nb8-a6 Na3xc4 c7-c5
15 00:00 413k 2,263k -0.46 Nb8-a6 Na3xc4 c7-c5 b2-b3 Bc8-f5 Bc1-b2 Bf5-e4 Qd1-d2 b7-b6 Ra1-c1 Qd8-d5 d4xc5 Na6xc5 Qd2xd5 Nf6xd5 Nf3-e5
16 00:00 440k 2,285k -0.51 Nb8-a6 Na3xc4 c7-c5 b2-b3 Bc8-f5 Bc1-b2 Bf5-e4 Qd1-d2 b7-b6 Ra1-c1 Qd8-d5 d4xc5 Na6xc5 Qd2xd5 Nf6xd5 Nf3-e5 Be4xg2 Kg1xg2 Bg7xe5 Nc4xe5 Rf8-d8
17+ 00:00 726k 2,062k -0.44 Nb8-a6 Na3xc4 c7-c5
17 00:00 777k 2,075k -0.38 Nb8-a6 Na3xc4 c7-c5 b2-b3 Bc8-f5 Bc1-b2 Bf5-e4 Qd1-d2 b7-b6 Qd2-e3 Qd8-c7 Rf1-d1 Ra8-d8 Ra1-c1 Qc7-b7 Nc4-e5 Na6-b4 d4xc5
18- 00:00 893k 2,103k -0.45 Nb8-a6 Na3xc4
18 00:00 1,522k 2,123k -0.44 Nb8-a6 Na3xc4 c7-c5 b2-b3 Bc8-f5 Bc1-b2 Bf5-e4 Qd1-d2 b7-b6 Rf1-d1 Qd8-d5 d4xc5 Na6xc5 Qd2xd5 Nf6xd5 Nf3-e5 Be4xg2 Kg1xg2 Rf8-d8 Ne5-c6 Bg7xb2 Nc4xb2
19 00:00 1,551k 2,131k -0.44 Nb8-a6 Na3xc4 c7-c5 b2-b3 Bc8-f5 Bc1-b2 Bf5-e4 Qd1-d2 b7-b6 Rf1-d1 Qd8-d5 d4xc5 Na6xc5 Qd2xd5 Nf6xd5 Nf3-e5 Be4xg2 Kg1xg2 Rf8-d8 Ne5-c6 Bg7xb2 Nc4xb2
20+ 00:01 2,498k 2,098k -0.37 Nb8-a6 Na3xc4 c7-c5
20 00:02 5,310k 2,108k -0.39 Nb8-a6 Nf3-e5 c7-c5 d4xc5 Qd8-c7 c5-c6 c4-c3 b2xc3 Qc7xe5 c6xb7 Bc8xb7 Bg2xb7 Ra8-d8 Qd1-c2 Na6-c5 Bb7-f3 Nc5-e4 Bc1-f4 Qe5xc3 Bf3xe4 Nf6xe4 Qc2xe4 Qc3xa3
21+ 00:03 6,797k 2,099k -0.32 Nb8-a6 Na3xc4 c7-c5
21- 00:03 7,428k 2,101k -0.39 Nb8-a6 Na3xc4
21+ 00:04 9,182k 2,072k -0.36 Nb8-a6 Na3xc4 c7-c5
21- 00:04 9,188k 2,073k -0.43 Nb8-a6 Na3xc4
21 00:05 10,669k 2,069k -0.32 c7-c5 d4xc5 c4-c3 Na3-b5 Nb8-a6 Nb5xc3 Na6xc5 Bc1-e3 Nf6-e4 Nc3xe4 Nc5xe4 Be3-d4 Bg7xd4 Qd1xd4 Qd8xd4 Nf3xd4 Rf8-d8 Rf1-d1 Ne4-d6 Ra1-c1 Bc8-d7 f2-f4 Ra8-c8 Rc1xc8 Rd8xc8
22- 00:05 11,891k 2,061k -0.39 c7-c5 d4xc5
22+ 00:06 13,193k 2,046k -0.37 c7-c5 d4xc5 c4-c3
22 00:07 16,010k 2,033k -0.26 c7-c5 d4xc5 c4-c3 Na3-b5 Nb8-a6 Nb5xc3 Na6xc5 Bc1-e3 Nf6-e4 Nc3xe4 Nc5xe4 Be3-d4 Bg7xd4 Nf3xd4 Ne4-f6 Nd4-b3 Ra8-b8 Qd1-c1 Qd8-b6 Qc1-c5 Qb6xc5 Nb3xc5 Rf8-d8
23- 00:10 20,270k 2,016k -0.33 c7-c5 Na3xc4
23+ 00:10 22,028k 2,015k -0.31 c7-c5 Na3xc4 Bc8-e6
23 00:12 25,053k 2,021k -0.25 c7-c5 Na3xc4 Bc8-e6 b2-b3 Nb8-c6 Bc1-b2 c5xd4 Nf3xd4 Nc6xd4 Bb2xd4 Be6-d5 Bg2xd5 Nf6xd5 Bd4xg7 Kg8xg7 Qd1-d4+ Nd5-f6 Qd4-b2 Qd8-c7 Ra1-c1 Rf8-d8 e2-e4 b7-b5 Nc4-e5
24- 00:13 26,895k 2,026k -0.32 c7-c5 Na3xc4
24+ 00:14 28,757k 2,020k -0.30 c7-c5 Na3xc4 Bc8-e6
24- 00:14 29,885k 2,025k -0.35 c7-c5 Na3xc4
24+ 00:14 30,288k 2,025k -0.31 c7-c5 Na3xc4 Bc8-e6
24- 00:15 30,638k 2,024k -0.41 c7-c5 Na3xc4
24 00:15 32,317k 2,025k -0.37 c7-c5 Na3xc4 Bc8-e6 Nf3-e5 Qd8-c8 d4xc5 Nf6-d7 Ne5xd7 Nb8xd7 c5-c6 b7xc6 Nc4-a5 Be6-d5 Qd1-c2 Bd5xg2 Kg1xg2 c6-c5 Rf1-d1 Nd7-e5 Bc1-f4 Ne5-c6 Qc2xc5 Nc6xa5 Qc5xa5 Qc8-b7+ Qa5-d5 Qb7xb2 Qd5-e4 a7-a5 Qe4xe7 Ra8-e8
25+ 00:18 38,394k 2,025k -0.30 c7-c5 Na3xc4 Bc8-e6
25 00:20 40,581k 2,024k -0.28 c7-c5 Na3xc4 Bc8-e6 Nf3-e5 Qd8-c8 d4xc5 Nf6-d7 Ne5xd7 Nb8xd7 c5-c6 b7xc6 Nc4-a5 Be6-d5 Qd1-c2 Bd5xg2 Kg1xg2 c6-c5 Rf1-d1 Nd7-e5 Bc1-f4 Ne5-c6 Qc2xc5 Nc6xa5 Qc5xa5 Qc8-b7+ Qa5-d5 Qb7xb2 Ra1-c1 Qb2xe2 Rd1-e1 e7-e6 Qd5-a5
26- 00:21 42,551k 2,021k -0.35 c7-c5 Na3xc4
26+ 00:21 43,293k 2,018k -0.33 c7-c5 Na3xc4 Bc8-e6
26- 00:22 44,522k 2,016k -0.38 c7-c5 Na3xc4
26+ 00:22 45,014k 2,013k -0.34 c7-c5 Na3xc4 Bc8-e6
26 00:24 49,911k 2,007k -0.29 c7-c5 d4xc5 c4-c3 Na3-b5 Nb8-a6 Nb5xc3 Na6xc5 Bc1-e3 Nf6-e4 Nc3xe4 Nc5xe4 Qd1-b3 Ne4-d6 Ra1-d1 Qd8-a5 Nf3-g5 Bc8-d7 Rd1xd6 e7xd6 Bg2-d5 Bd7-a4 Bd5xf7+ Kg8-h8 Qb3-d5 Qa5xd5 Bf7xd5 Bg7xb2
27- 00:31 63,412k 1,992k -0.36 c7-c5 Na3xc4
27+ 00:32 64,288k 1,989k -0.34 c7-c5 Na3xc4 Bc8-e6
27 00:32 65,271k 1,988k -0.30 c7-c5 Na3xc4 Bc8-e6 Nf3-e5 Qd8-c8 d4xc5 Nf6-d7 Ne5xd7 Nb8xd7 c5-c6 b7xc6 Nc4-a5 Be6-d5 Qd1-c2 Bd5xg2 Kg1xg2 c6-c5 Rf1-d1 Nd7-b6 Bc1-e3 Qc8-e6 Ra1-c1 Rf8-c8 b2-b3 h7-h5 Be3xc5 Bg7-h6 e2-e3 h5-h4 Qc2-d3 Rc8-d8 Qd3-e2
28- 00:35 69,763k 1,978k -0.37 c7-c5 Na3xc4
28- 00:55 107,850k 1,938k -0.46 c7-c5 Na3xc4
28+ 01:01 117,965k 1,931k -0.43 c7-c5 Na3xc4 Bc8-e6
28 01:01 118,635k 1,931k -0.42 c7-c5 Na3xc4 Bc8-e6 Nf3-e5 Qd8-c8 d4xc5 Nf6-d7 Ne5xd7 Nb8xd7 c5-c6 b7xc6 Nc4-a5 Be6-d5 Bg2xd5 c6xd5 Qd1xd5 Nd7-f6 Qd5-f3 Qc8-e6 Rf1-d1 Nf6-e4 Ra1-b1 Ra8-d8 Bc1-f4 Rd8xd1+ Rb1xd1 Bg7xb2 Qf3-b3 Qe6xb3 Na5xb3 Bb2-f6 Bf4-e3 Ne4-c3 Rd1-d2 Rf8-c8 Be3xa7
29+ 01:03 123,129k 1,930k -0.35 c7-c5 Na3xc4 Bc8-e6
29 01:05 125,559k 1,931k -0.30 c7-c5 Na3xc4 Bc8-e6 Nf3-e5 Qd8-c8 d4xc5 Nf6-d7 Ne5xd7 Nb8xd7 c5-c6 b7xc6 Nc4-a5 Be6-d5 Bg2xd5 c6xd5 Qd1xd5 Nd7-f6 Qd5-f3 Qc8-e6 Na5-b3 Nf6-e4 Ra1-b1 a7-a5 Rf1-d1 Ra8-d8 Bc1-e3 a5-a4 Nb3-d4 Qe6-d5 h2-h4 f7-f5 a2-a3 e7-e5 Nd4-c2
30- 01:10 136,027k 1,923k -0.37 c7-c5 Na3xc4
30+ 01:18 150,152k 1,913k -0.35 c7-c5 Na3xc4 Bc8-e6
30- 01:23 159,306k 1,909k -0.40 c7-c5 Na3xc4
30 01:28 169,524k 1,905k -0.41 c7-c5 Na3xc4 Bc8-e6 Nf3-e5 Qd8-c8 d4xc5 Nf6-d7 c5-c6 b7xc6 Ne5xd7 Nb8xd7 Nc4-a5 Be6-d5 Bg2xd5 c6xd5 Qd1xd5 Nd7-f6 Qd5-f3 Qc8-e6 Na5-b3 Nf6-e4 Ra1-b1 Rf8-d8 Bc1-e3 a7-a5 Rf1-d1 a5-a4 Nb3-d4 Qe6-d5 b2-b3 a4xb3 Nd4xb3 Qd5xd1+ Rb1xd1 Rd8xd1+ Kg1-g2 Ra8-a4
31+ 01:35 182,044k 1,901k -0.34 c7-c5 Na3xc4 Bc8-e6
31- 01:36 182,557k 1,901k -0.41 c7-c5 Na3xc4
31 01:47 204,843k 1,898k -0.45 c7-c5 Na3xc4 Bc8-e6 Nf3-e5 Qd8-c8 d4xc5 Nf6-d7 Ne5xd7 Nb8xd7 c5-c6 b7xc6 Nc4-a5 Be6-d5 Bg2xd5 c6xd5 Qd1xd5 Nd7-f6 Qd5-f3 Qc8-e6 Na5-b3 Nf6-e4 Ra1-b1 a7-a5 Rf1-d1 a5-a4 Nb3-d4 Qe6-d5 Bc1-e3 Rf8-d8 Kg1-g2 f7-f5 Nd4-b3 Qd5-b5 Nb3-d2 Ne4-f6 Rd1-c1 Nf6-d5 b2-b3 Nd5xe3+ Qf3xe3 a4xb3 Rb1xb3
32+ 02:01 229,953k 1,892k -0.38 c7-c5 Na3xc4 Bc8-e6
32+ 02:10 247,192k 1,892k -0.29 c7-c5 Na3xc4 Bc8-e6
32 02:14 254,051k 1,891k -0.33 c7-c5 Na3xc4 Bc8-e6 Nf3-e5 Qd8-c8 d4xc5 Nf6-d7 Ne5xd7 Nb8xd7 c5-c6 b7xc6 Nc4-a5 Be6-d5 Qd1-c2 Bd5xg2 Kg1xg2 c6-c5 Rf1-d1 Nd7-e5 Bc1-f4 Ne5-c6 Qc2xc5 Nc6xa5 Qc5xa5 Qc8-b7+ Qa5-d5 Qb7xb2 e2-e3 e7-e6 Qd5-d7 h7-h6 a2-a4 Rf8-d8 Qd7xd8+ Ra8xd8 Rd1xd8+ Kg8-h7 Ra1-d1 Qb2-b7+ Kg2-g1
33- 02:23 270,971k 1,890k -0.40 c7-c5 Na3xc4
33+ 02:29 282,443k 1,888k -0.38 c7-c5 Na3xc4 Bc8-e6
33+ 03:07 352,551k 1,881k -0.26 c7-c5 Na3xc4 Bc8-e6
33- 03:08 354,235k 1,881k -0.34 c7-c5 Na3xc4
33+ 03:14 364,864k 1,879k -0.29 c7-c5 Na3xc4 Bc8-e6
33 03:17 371,314k 1,878k -0.32 c7-c5 Na3xc4 Bc8-e6 Nf3-e5 Qd8-c8 d4xc5 Nf6-d7 Ne5xd7 Nb8xd7 c5-c6 b7xc6 Nc4-a5 Be6-d5 Bg2xd5 c6xd5 Qd1xd5 Nd7-f6 Qd5-f3 Qc8-e6 Na5-b3 Nf6-e4 Ra1-b1 a7-a5 Rf1-d1 Rf8-d8 Bc1-e3 a5-a4 Nb3-d4 Bg7xd4 Rd1xd4 Rd8xd4 Be3xd4 Ra8-d8 Qf3-e3 Qe6xa2 Rb1-d1 Qa2-d5 Rd1-d3 Qd5-e6 f2-f3 Ne4-f6 Qe3xe6 f7xe6 Rd3-d2 Kg8-f7 Bd4-c3 Rd8-b8
34+ 03:46 424,374k 1,872k -0.25 c7-c5 Na3xc4 Bc8-e6
34- 03:47 426,644k 1,871k -0.32 c7-c5 Na3xc4
34+ 03:53 437,264k 1,870k -0.29 c7-c5 Na3xc4 Bc8-e6
34- 03:58 446,098k 1,870k -0.36 c7-c5 Na3xc4
34 04:32 508,229k 1,865k -0.42 c7-c5 Na3xc4 Bc8-e6 Nf3-e5 Qd8-c8 d4xc5 Nf6-d7 Ne5xd7 Nb8xd7 c5-c6 b7xc6 Nc4-a5 Be6-d5 Bg2xd5 c6xd5 Qd1xd5 Nd7-f6 Qd5-f3 Qc8-e6 Na5-b3 Nf6-e4 Ra1-b1 a7-a5 Rf1-d1 Rf8-d8 Bc1-e3 a5-a4 Nb3-d4 Bg7xd4 Be3xd4 Ne4-g5 Qf3-e3 Qe6xa2 Rb1-a1 Qa2-d5 h2-h4 Ng5-e6 Bd4-c3 Qd5-c4 Rd1xd8+ Ra8xd8 Qe3-a7 Qc4xe2 Qa7xe7 Rd8-d1+ Ra1xd1 Qe2xd1+ Kg1-h2
Lazy_Frank
Posts: 74
Joined: Mon Jul 23, 2018 10:56 pm
Location: Latvia
Full name: Raivis Baumanis

Re: Testing depend largely on good Openings

Post by Lazy_Frank »

Ovyron wrote: Sat Jul 09, 2022 3:40 pm
Lazy_Frank wrote: Sat Jul 09, 2022 9:32 am At the moment i am very interested to spot out where is chess draw limit (pick the bad line in range ~ 1.3-1.5 evaluation and sort out it is drawable or not), clearly opposite approach.
It's probably drawable. A more interesting limit is the 1.90 eval, I think the ones I've seen have some 50% chance of being drawable as the score falls or already winning if the score increases. At 2.00 it's probably winning.

These scores hold at about early middle game, before I've seen 2.00 positions that are drawable still near the opening, later, say, by move 50 a score of 1.40 can be decisive. Probably someone smart could map these out, but Stockfish NNUE overestimates its chances on the opening and underestimates them near the endgame.
That is exactly what i try to do - create map or collection of dubious opening lines. Currently i covered 2 ply and 3 ply openings.
Not so much in 2 ply: just 1. a4 f6 and 1. h3 h5.
At 3-ply covered and approved 86 lines.
4-ply in process (674 lines in consideration to be good, 4140 lines remains to check).
Tree expands numerously be each next ply.
Chessqueen
Posts: 5685
Joined: Wed Sep 05, 2018 2:16 am
Location: Moving
Full name: Jorge Picado

Re: Testing depend largely on good Openings

Post by Chessqueen »

Chessqueen wrote: Sat Jul 09, 2022 5:49 pm
lkaufman wrote: Sat Jul 09, 2022 5:39 pm
Ovyron wrote: Sat Jul 09, 2022 8:43 am
lkaufman wrote: Fri Jul 08, 2022 5:49 am the path to a draw is too clear now.
No, the path to draw isn't clear at all. Try making a "balanced" book, one where no matter what the opponent plays, you'll reach a balanced opening that you will be able to draw.

Well, I know people that would be able to bust one of those lines, and prove the "draw that was clear" wasn't so at all. It's just that testers of engines never try to find a hole on those "balanced" openings, so if there's a move that gives a side some big advantage, it'll not be found, because nobody is looking for it.

This is specially true on the age of NNUE, where you can build a big book on an off-beat line and win just because your side has a lot more time on the clock than the opponent missing your lines. Computer chess on the competitive side is still live and kicking, if anybody tried to use those balanced openings on playchess they'd be utterly destroyed by better moves by people that dedicate their lives to finding holes in lines.

I've seen positions where Stockfish NNUE has the score backwards, both engines show some +0.20 but after the game continues both engines agree it's really -0.40, so a black player wants to get into this line, while engine testers will be happy to settle to play some +0.15 position that is actually 0.00 missing black's improvement.

I've seen completely lost positions that are scored at +0.30, but there's already a sequence of moves that force some +2.00 advantage, so white looks great aiming for this line in case the opponent allows it and loses. An engine tester already knows what moves black will play so has no reason to try this.

What engine testing is missing is competition, trying to do the best that you can for your side causes you to find and play better openings. Using balanced openings with reversed colors leads to draw death, because there's no incentive to find better lines for either side.
Have people tried entering latest Stockfish on good hardware in Rapid (not blitz) online tournaments with NO BOOK? If so, does it lose any games, and if so what percent? Or perhaps same with just a very short generic book from GM games, like 5 or 6 moves only?
My opinion is that Stockfish or Komodo performs better with short Opening book of no more than 6 moves than with a very long generic book from GM games of 16 moves, because some of those long opening books have bad lines like in the case that I have posted several times, and if you let Komodo or Stockfish analyze the position after the bad move was chosen by the Opening book it could make a difference of minus 0.42 or +0.41 which will determine weather White will draw or lose . :roll:

Position analyzed after the opening move 7. Nd2? which favor Black, and can win easy after several moves, if the Black side is the strongest engine like Stockfish 15 :shock:
[fen]rnbq1rk1/ppp1ppbp/5np1/8/2pP4/5NP1/PP1NPPBP/R1BQ1RK1 b - - 1 1 [/fen]
FEN: rnbq1rk1/ppp1ppbp/5np1/8/2pP4/5NP1/PP1NPPBP/R1BQ1RK1 b - - 1 1

[pgn][Event "Computer chess game"]
[Site "DESKTOP-OFQ3C0P"]
[Date "2022.07.03"]
[Round "?"]
[White "Dragon-2.6.1-64bit-avx2"]
[Black "Stockfish_15_x64_avx2"]
[Result "0-1"]
[BlackElo "3555"]
[ECO "D77"]
[Opening "Neo-Grünfeld, 6.O-O dxc4"]
[Time "19:21:03"]
[WhiteElo "340"]
[TimeControl "900+10"]
[Termination "adjudication"]
[PlyCount "153"]
[WhiteType "program"]
[BlackType "program"]

1. d4 Nf6 2. c4 g6 3. g3 Bg7 4. Bg2 O-O 5. Nf3 d5 6. O-O dxc4 7. Nbd2 b5 8.
Ne5 Nd5 9. b3 Nc3 10. Qe1 Qxd4 11. Nxf7 Nd5 12. Ng5 h6 13. bxc4 bxc4 14. e3
Qe5 15. Nxc4 Qxg5 16. e4 Qh5 17. exd5 Bxa1 18. d6 exd6 19. Bxa8 Bh3 20. Bg2
Bxg2 21. Qe6+ Kh7 22. Kxg2 Bg7 23. f3 Qb5 24. Ne3 Re8 25. Qf7 Qd7 26. Qc4
Re5 27. Rd1 Qe7 28. Re1 Rc5 29. Qb3 Nc6 30. Bb2 Bxb2 31. Qxb2 h5 32. Qd2
Qf8 33. Re2 Ne5 34. Rf2 Qa8 35. Qd4 Kg7 36. h3 Kf7 37. g4 Ke6 38. Qb4 a5
39. Qa4 Qc6 40. Qb3+ Kd7 41. Qg8 Qb5 42. gxh5 gxh5 43. Qh7+ Kc8 44. h4 Nd3
45. Rd2 Nf4+ 46. Kg3 Ne2+ 47. Kh2 Nc3 48. Ng2 Qb1 49. Qh8+ Kb7 50. Qd4 Nb5
51. Qf4 Qb4 52. Qe3 a4 53. Qd3 d5 54. Kg3 d4 55. Nf4 Re5 56. Rd1 Qd6 57.
Rd2 Kb6 58. Kg2 Qf6 59. Kg3 c6 60. Ng2 a3 61. f4 Rc5 62. Kh2 Qf5 63. Qe2
Rc3 64. Kg1 Rc1+ 65. Kh2 Rb1 66. Qd3 Qxd3 67. Rxd3 Rf1 68. Rb3 Rf2 69. Kg3
Rxa2 70. Rb1 Rb2 71. Ra1 Rb3+ 72. Kf2 Nd6 73. f5 Nxf5 74. Nf4 c5 75. Nxh5
Nxh4 76. Ng7 c4 77. Ne6 {User Adjudication} *[/pgn]
Last edited by Chessqueen on Sat Jul 09, 2022 9:17 pm, edited 4 times in total.
Chessqueen
Posts: 5685
Joined: Wed Sep 05, 2018 2:16 am
Location: Moving
Full name: Jorge Picado

Re: Testing depend largely on good Openings

Post by Chessqueen »

Chessqueen wrote: Sat Jul 09, 2022 5:52 pm
lkaufman wrote: Sat Jul 09, 2022 5:39 pm
Ovyron wrote: Sat Jul 09, 2022 8:43 am
lkaufman wrote: Fri Jul 08, 2022 5:49 am the path to a draw is too clear now.
No, the path to draw isn't clear at all. Try making a "balanced" book, one where no matter what the opponent plays, you'll reach a balanced opening that you will be able to draw.

Well, I know people that would be able to bust one of those lines, and prove the "draw that was clear" wasn't so at all. It's just that testers of engines never try to find a hole on those "balanced" openings, so if there's a move that gives a side some big advantage, it'll not be found, because nobody is looking for it.

This is specially true on the age of NNUE, where you can build a big book on an off-beat line and win just because your side has a lot more time on the clock than the opponent missing your lines. Computer chess on the competitive side is still live and kicking, if anybody tried to use those balanced openings on playchess they'd be utterly destroyed by better moves by people that dedicate their lives to finding holes in lines.

I've seen positions where Stockfish NNUE has the score backwards, both engines show some +0.20 but after the game continues both engines agree it's really -0.40, so a black player wants to get into this line, while engine testers will be happy to settle to play some +0.15 position that is actually 0.00 missing black's improvement.

I've seen completely lost positions that are scored at +0.30, but there's already a sequence of moves that force some +2.00 advantage, so white looks great aiming for this line in case the opponent allows it and loses. An engine tester already knows what moves black will play so has no reason to try this.

What engine testing is missing is competition, trying to do the best that you can for your side causes you to find and play better openings. Using balanced openings with reversed colors leads to draw death, because there's no incentive to find better lines for either side.
Have people tried entering latest Stockfish on good hardware in Rapid (not blitz) online tournaments with NO BOOK? If so, does it lose any games, and if so what percent? Or perhaps same with just a very short generic book from GM games, like 5 or 6 moves only?
My opinion is that Stockfish or Komodo performs better with short Opening book of no more than 6 moves than with a very long generic book from GM games of 16 moves, because some of those long opening books have bad lines like in the case that I have posted several times, and if you let Komodo or Stockfish analyze the position after the bad move was chosen by the Opening book it could make a difference of minus 0.42 or +0.41 which will determine weather White will draw or lose . :roll:


Position analyzed after Komodo Dragon chose the opening move 7.Ne5! which does NOT favor Black, and is able to Draw against Stockfish 15
[fen]rnbq1rk1/ppp1ppbp/5np1/8/2pP4/N4NP1/PP2PPBP/R1BQ1RK1 b - - 1 1 [/fen]
FEN: rnbq1rk1/ppp1ppbp/5np1/8/2pP4/N4NP1/PP2PPBP/R1BQ1RK1 b - - 1 1

Or even 7.Ne5! can hold the draw against Stockfish 15
[pgn][Event "Computer chess game"]
[Site "DESKTOP-OFQ3C0P"]
[Date "2022.07.03"]
[Round "?"]
[White "Dragon-2.6.1-64bit-avx2"]
[Black "Stockfish_15_x64_avx2"]
[Result "1/2-1/2"]
[BlackElo "3560"]
[ECO "D77"]
[Opening "Neo-Grünfeld, 6.O-O dxc4"]
[Time "19:21:03"]
[WhiteElo "3540"]
[TimeControl "900+10"]
[Termination "normal"]
[PlyCount "118"]
[WhiteType "program"]
[BlackType "program"]

1. d4 Nf6 2. c4 g6 3. g3 Bg7 4. Bg2 O-O 5. Nf3 d5 6. O-O dxc4 7. Ne5 Ne8 8.
Na3 c5 9. Be3 {(Bc1-e3 c5xd4 Be3xd4 Ne8-d6 Ne5xc4 Bg7xd4 Qd1xd4 Nd6xc4
Qd4xc4 Nb8-c6 Rf1-d1 Qd8-b6 Qc4-b5 Bc8-e6 Qb5xb6 a7xb6 Bg2-d5 Be6xd5 Rd1xd5
Rf8-d8 Rd5-b5 Ra8-a6 Kg1-f1 Rd8-c8 Ra1-d1 Kg8-f8 Rd1-d7 Rc8-b8 Rd7-d2
Kf8-e8 g3-g4 Rb8-d8 Kf1-e1 f7-f6 Rb5-b3 Rd8xd2 Ke1xd2 h7-h5 g4xh5 g6xh5
Kd2-c3 Nc6-e5 f2-f4 Ne5-c6 Na3-c4) +0.44/38 57} cxd4 {(c5xd4 Be3xd4 Ne8-d6
Ne5xc4 Bg7xd4 Qd1xd4 Nd6xc4 Qd4xc4 Nb8-c6 Rf1-d1 Qd8-b6 Qc4-b5 Bc8-e6
Qb5xb6 a7xb6 Bg2-d5 Be6xd5 Rd1xd5 Rf8-d8 Rd5-b5 Ra8-a6 Kg1-f1 Rd8-c8 Ra1-d1
Kg8-f8 Rd1-d7 Rc8-b8 g3-g4 Kf8-e8 Rd7-d2 f7-f6 Rb5-b3 Rb8-d8 Rd2xd8+ Nc6xd8
Kf1-e1 Nd8-e6 Ke1-d2 Ne6-c5 Rb3-h3 Nc5-e4+ Kd2-e1 Ne4-g5 Rh3-c3 Ng5-f7
f2-f3) -0.23/36 18} 10. Bxd4 {(Be3xd4 Ne8-d6 Ne5xc4 Bg7xd4 Qd1xd4 Nd6xc4
Qd4xc4 Nb8-c6 Rf1-d1 Qd8-b6 Qc4-b5 Bc8-e6 Qb5xb6 a7xb6 Bg2-d5 Be6xd5 Rd1xd5
Rf8-d8 Rd5-b5 Ra8-a6 Kg1-f1 Rd8-c8 Ra1-d1 Kg8-f8 Rd1-d7 Rc8-b8 Kf1-e1
Kf8-e8 Rd7-d2 Rb8-d8 Rd2xd8+ Nc6xd8 Ke1-d2 Nd8-e6 Rb5-b3 Ne6-d4 Rb3-d3
Nd4-c6 Rd3-c3 f7-f6 Rc3-b3) +0.48/36 29} Nd6 {(Ne8-d6 Ne5xc4 Bg7xd4 Qd1xd4
Nd6xc4 Qd4xc4 Nb8-c6 Rf1-d1 Qd8-b6 Qc4-b5 Bc8-e6 Qb5xb6 a7xb6 Bg2-d5 Be6xd5
Rd1xd5 Ra8-a5 Ra1-d1 Ra5xd5 Rd1xd5 Rf8-a8 Kg1-f1 Kg8-f8 Kf1-e1 Kf8-e8
Rd5-b5 Ra8-a6 g3-g4 f7-f6 Ke1-d2 Nc6-d4 Rb5-b4 Nd4-c6 Rb4-b3 Ra6-a4 Rb3xb6
Ra4xg4 Rb6xb7) -0.33/35 55} 11. Nexc4 {(Ne5xc4 Bg7xd4 Qd1xd4 Nd6xc4 Qd4xc4
Nb8-c6 Rf1-d1 Qd8-b6 Qc4-b5 Bc8-e6 Qb5xb6 a7xb6 Bg2-d5 Be6xd5 Rd1xd5 Rf8-d8
Ra1-d1 Rd8xd5 Rd1xd5 Ra8-a5 Rd5-b5 Ra5-a6 Kg1-f1 Kg8-f8 Kf1-e1 Kf8-e8
Ke1-d2 Nc6-d4 Rb5-b4 Nd4-c6 Rb4-b3 Nc6-d4 Rb3-d3 Nd4-f5 f2-f3 Nf5-d6 Rd3-b3
Ke8-d8 g3-g4 Kd8-e8) +0.50/36 24} Bxd4 {(Bg7xd4 Qd1xd4 Nd6xc4 Qd4xc4 Nb8-c6
Rf1-d1 Qd8-b6 Qc4-b5 Bc8-e6 Qb5xb6 a7xb6 Bg2-d5 Be6xd5 Rd1xd5 Rf8-d8 Ra1-d1
Rd8xd5 Rd1xd5 f7-f6 Kg1-f1 Kg8-f7 Kf1-e1 Kf7-e8 f2-f3 Nc6-e5 Rd5-b5 Ra8-a6
Ke1-d1 Ne5-f7 Rb5-b4 Nf7-d6 Kd1-c2 Ke8-d7 Kc2-b3 Kd7-c6 Na3-c2 Ra6-a8
Nc2-d4+ Kc6-c7 a2-a4 e7-e5 Nd4-b5+ Nd6xb5 Rb4xb5) -0.24/38 23} 12. Qxd4
{(Qd1xd4 Nd6xc4 Qd4xc4 Nb8-c6 Rf1-d1 Qd8-b6 Qc4-b5 Bc8-e6 Qb5xb6 a7xb6
Bg2-d5 Be6xd5 Rd1xd5 Rf8-d8 Ra1-d1 Rd8xd5 Rd1xd5 f7-f6 Kg1-f1 Kg8-f7 Kf1-e1
h7-h5 Ke1-d2 Nc6-e5 f2-f3 Kf7-e8 Rd5-b5 Ke8-d7 Rb5-b3 Kd7-c6 Na3-c2 Ra8-a5
e2-e4 Ra5-a4 a2-a3 Ra4-c4 Nc2-b4+ Kc6-d6 h2-h4 f6-f5 f3-f4 Rc4-d4+ Nb4-d3
f5xe4 f4xe5+ Kd6-d5) +0.44/39 25} Nxc4 {(Nd6xc4 Qd4xc4 Nb8-c6 Rf1-d1 Qd8-b6
Qc4-b5 Bc8-e6 Qb5xb6 a7xb6 Bg2-d5 Be6xd5 Rd1xd5 Rf8-d8 Ra1-d1 Rd8xd5 Rd1xd5
f7-f6 Kg1-f1 Kg8-f7 Kf1-e1 Ra8-a5 Rd5-d7 Ra5-a7 Ke1-d2 Kf7-e8 Rd7-d5 Nc6-e5
f2-f3 Ra7-a5 Rd5xa5 b6xa5 Kd2-c3 Ke8-d7 Na3-c4 Ne5xc4 Kc3xc4 Kd7-c6 f3-f4
b7-b5+ Kc4-d4 h7-h5 e2-e4 Kc6-d6) -0.19/34 31} 13. Qxc4 {(Qd4xc4 Nb8-c6
Rf1-d1 Qd8-b6 Qc4-b5 Bc8-e6 Qb5xb6 a7xb6 Bg2-d5 Be6xd5 Rd1xd5 Rf8-d8 Ra1-d1
Rd8xd5 Rd1xd5 f7-f6 Kg1-f1 Kg8-f7 Kf1-e1 Ra8-a5 Rd5-d7 Ra5-a7 Ke1-d2 Kf7-e8
Rd7-d5 Nc6-e5 Rd5-b5 Ra7-a6 h2-h4 Ke8-d7 Rb5-b3 Ne5-f7 f2-f3 h7-h5 Kd2-c3
Nf7-d6 Rb3-b4) +0.36/33 32} Nc6 {(Nb8-c6 Rf1-d1 Qd8-b6 Qc4-b5 Bc8-e6 Qb5xb6
a7xb6 Bg2-d5 Be6xd5 Rd1xd5 Rf8-d8 Ra1-d1 Rd8xd5 Rd1xd5 Kg8-f8 Kg1-f1 Kf8-e8
Kf1-e1 Ra8-a5 Rd5-b5 Ra5-a6 Ke1-d2 Nc6-d4 Rb5-d5 Nd4-e6 Kd2-c1 Ne6-c7
Rd5-d3 b6-b5 Rd3-b3 Ke8-d8 f2-f3 Kd8-d7 e2-e4 Kd7-c6 Kc1-b1 Kc6-c5 Rb3-c3+
Kc5-d6 Rc3-c1 h7-h5 f3-f4 f7-f6 Rc1-d1+ Kd6-e6) -0.30/36 22} 14. Rfd1
{(Rf1-d1 Qd8-b6 Qc4-b5 Bc8-e6 Qb5xb6 a7xb6 Bg2-d5 Be6xd5 Rd1xd5 Ra8-a5
Ra1-d1 Rf8-a8 Rd5-d7 Ra8-a7 Rd7-c7 b6-b5 Rd1-d5 b5-b4 Na3-b5 Ra7-a6 Rc7xb7
Ra5xa2 Rd5-d2 Kg8-f8 Rb7-c7 Kf8-g7 Rc7-d7 b4-b3 Kg1-g2 Ra6-a4 f2-f3 Ra2-a1
Rd7-c7 Ra4-b4 Nb5-c3 Rb4-c4 g3-g4 h7-h5 g4xh5 g6xh5 Kg2-f2) +0.32/38 38}
Qb6 {(Qd8-b6 Qc4-b5 Bc8-e6 Qb5xb6 a7xb6 Bg2-d5 Be6xd5 Rd1xd5 Rf8-d8 Ra1-d1
Rd8xd5 Rd1xd5 f7-f6 Kg1-f1 Kg8-f7 Kf1-e1 Kf7-e8 Rd5-b5 Ra8-a6 Ke1-d2 Nc6-d4
Rb5-d5 Nd4-e6 f2-f3 h7-h5 Rd5-d3 Ne6-c7 Kd2-c1 b6-b5 Rd3-b3 Ke8-d7 e2-e4
Kd7-c8 Kc1-b1 Ra6-b6 h2-h4 Kc8-d7 Na3-c2 Kd7-c8) -0.21/36 26} 15. Qb5
{(Qc4-b5 Bc8-e6 Qb5xb6 a7xb6 Bg2-d5 Be6xd5 Rd1xd5 Rf8-d8 Rd5-b5 Ra8-a6
Kg1-f1 Rd8-d4 Ra1-c1 Rd4-d7 e2-e3 h7-h5 Kf1-e1 f7-f6 Ke1-e2 Kg8-g7 f2-f4
Nc6-a7 Rb5-b4 Kg7-f7 Rc1-c3 b6-b5 Rc3-d3 Rd7-c7 e3-e4 Ra6-a4 Rb4xa4 b5xa4
Rd3-c3 Rc7xc3 b2xc3) +0.36/37 35} Be6 {(Bc8-e6 Qb5xb6 a7xb6 Bg2-d5 Be6xd5
Rd1xd5 Rf8-d8 Rd5-b5 Ra8-a6 Kg1-f1 Rd8-d4 Ra1-c1 f7-f6 e2-e3 Rd4-d7 Kf1-e2
Nc6-a7 Rb5-b3 Kg8-f7 h2-h3 h7-h5 Rc1-c2 b6-b5 Rb3-b4 Kf7-e6 Rc2-c3 Ke6-f7
h3-h4 Ra6-d6 Rc3-c2 Rd6-a6 Rc2-c1 Kf7-e8 Rb4-b3 Ke8-f7 Rc1-c5 Ra6-c6 Rc5-c3
Rc6xc3 b2xc3 Rd7-c7) -0.14/38 26} 16. Qxb6 {(Qb5xb6 a7xb6 Bg2-d5 Be6xd5
Rd1xd5 Rf8-d8 Rd5-b5 Ra8-a6 Kg1-f1 Rd8-d4 Ra1-c1 Rd4-d7 e2-e3 f7-f6 Kf1-e2
Kg8-f7 f2-f4 Nc6-a7 Rb5-b3 b6-b5 g3-g4 Ra6-a4 Na3xb5 Na7xb5 Rb3xb5 Ra4xa2
h2-h4 Ra2-a6 Rc1-d1 Rd7xd1 Ke2xd1 Ra6-a1+ Kd1-e2 Ra1-g1 Ke2-f3 Rg1-f1+
Kf3-e4 Rf1-g1 g4-g5 Rg1-h1 Rb5xb7 Rh1xh4 g5xf6 Kf7xf6 Rb7-b8 Rh4-h1 b2-b4
h7-h5) +0.22/39 29} axb6 {(a7xb6 Bg2-d5 Be6xd5 Rd1xd5 Rf8-d8 Rd5-b5 Ra8-a6
Kg1-f1 Rd8-d4 Ra1-c1 f7-f6 e2-e3 Rd4-d7 Kf1-e2 Kg8-f7 Rc1-c2 h7-h5 h2-h4
Rd7-c7 Rc2-c4 Rc7-d7 Rc4-c1 Rd7-d6 Rc1-d1 Nc6-a7 Rb5-b4 Na7-c6 Rb4-b3
Rd6xd1 Ke2xd1 Nc6-e5 Kd1-e2 Ne5-d7 Rb3-b4 Kf7-e6 f2-f4 f6-f5 e3-e4 f5xe4
Rb4xe4+ Ke6-f7 Re4-b4 Ra6-a5 Ke2-e3 Nd7-f6 Rb4-b5) -0.17/39 19} 17. Bd5
{(Bg2-d5 Be6xd5 Rd1xd5 Rf8-d8 Rd5-b5 Ra8-a6 Kg1-f1 Rd8-d4 Ra1-c1 Rd4-d7
e2-e3 f7-f6 Kf1-e2 Kg8-f7 h2-h3 Rd7-c7 Rc1-c4 Rc7-d7 Rc4-c2 Nc6-a7 Rb5-b4
Na7-c6 Rb4-b3 h7-h5 f2-f4 Nc6-a5 Rb3-b5 Na5-c6 Rc2-d2 Kf7-e8 Rb5-b3 Rd7xd2+
Ke2xd2 Nc6-d8 e3-e4 Ra6-a4 Kd2-e3 Nd8-f7 Ke3-d3 g6-g5 Rb3xb6 h5-h4 g3xh4)
+0.31/40 30} Bxd5 {(Be6xd5 Rd1xd5 Rf8-d8 Rd5-b5 Ra8-a6 Kg1-f1 Rd8-d4 Ra1-c1
Rd4-d7 Kf1-e1 f7-f6 e2-e3 Kg8-f7 Ke1-e2 h7-h5 Rb5-b3 Nc6-a5 Rb3-b4 Na5-c6
Rb4-b5 Rd7-d6 Rc1-d1 Rd6xd1 Ke2xd1 Nc6-e5 Kd1-e2 Kf7-e8 f2-f4 Ne5-d7 Rb5-d5
Nd7-c5 Rd5-d4 Ra6-a8 Ke2-d2 g6-g5 f4xg5 f6xg5 Rd4-b4 Nc5-d7 Na3-c4 Ra8xa2
Nc4xb6 Nd7xb6 Rb4xb6) -0.12/39 33} 18. Rxd5 {(Rd1xd5 Rf8-d8 Rd5-b5 Ra8-a6
Kg1-f1 Rd8-d4 Ra1-c1 f7-f6 e2-e3 Rd4-d7 Kf1-e2 Kg8-f7 f2-f4 h7-h5 Rc1-c3
Nc6-a7 Rb5-b3 b6-b5 Rc3-d3 Rd7-c7 Ke2-d2 Kf7-e6 e3-e4 Rc7-c5 Rd3-d4 Na7-c6
Rd4-d5 Rc5xd5+ e4xd5+ Ke6xd5 Na3xb5 Ra6xa2 Nb5-c3+ Kd5-c4 Rb3xb7 Ra2-a6
Rb7-c7 f6-f5 Kd2-e2 Ra6-b6) +0.21/37 25} Rfd8 {(Rf8-d8 Rd5-b5 Ra8-a6 Kg1-f1
Rd8-d4 Ra1-c1 f7-f6 e2-e3 Rd4-d7 Kf1-e2 Kg8-f7 Rc1-c2 h7-h5 Rc2-c3 Rd7-d6
Rc3-d3 Rd6xd3 Ke2xd3 Nc6-e5+ Kd3-e2 Kf7-e8 Rb5-b4 Ke8-d7 f2-f4 Ne5-f7 e3-e4
Kd7-c6 Rb4-c4+ Kc6-d7 Rc4-d4+ Kd7-c6 Ke2-d3 e7-e5 Rd4-c4+ Kc6-d7 Na3-b5
Ra6xa2 Rc4-c7+ Kd7-e8 Nb5-c3 Ra2xb2 Nc3-d5) -0.06/38 39} 19. Rb5 {(Rd5-b5
Ra8-a6 Kg1-f1 Rd8-d4 Kf1-e1 f7-f6 e2-e3 Rd4-a4 Ra1-d1 Kg8-f7 Rd1-d3 Nc6-e5
Rd3-c3 Ne5-d7 Ke1-d2 g6-g5 f2-f4 Nd7-c5 Rc3-c4 Nc5-e4+ Kd2-d3 Ra4xa3+
Kd3xe4 Ra3xa2 f4xg5 f6xg5 Rc4-b4 Ra6-a5 Rb5xb6 Ra2-a1 Rb4-b5 Ra5xb5 Rb6xb5
Ra1-h1 Ke4-f5 Rh1xh2 Kf5xg5) +0.28/37 58} Ra6 {(Ra8-a6 Kg1-f1 Rd8-d4 Kf1-e1
f7-f6 e2-e3 Rd4-b4 Rb5xb4 Nc6xb4 Ke1-d2 Ra6-a5 Kd2-c3 Nb4-d5+ Kc3-c4 Nd5-c7
e3-e4 Ra5-c5+ Kc4-d3 Nc7-a6 Na3-c2 b6-b5 f2-f3 Kg8-f7 Kd3-d2 Rc5-c6 Ra1-d1
e7-e5 Nc2-e1 Kf7-e7 Ne1-d3 Rc6-d6 Kd2-e2 Rd6-c6 h2-h4 Rc6-c2+ Rd1-d2 Rc2-c4
Ke2-e3 b5-b4 g3-g4 Na6-c5 Nd3xc5 Rc4xc5) -0.16/40 25} 20. Kf1 {(Kg1-f1
Rd8-d4 Ra1-c1 Rd4-d7 e2-e3 f7-f6 Kf1-e2 Nc6-a7 Rb5-b3 Kg8-f7 Rc1-c2 h7-h5
Rc2-c1 Kf7-e6 h2-h3 b6-b5 Rc1-a1 Ra6-c6 Ra1-d1 Rc6-c5 Rd1xd7 Ke6xd7 Ke2-d2
Kd7-c7 Na3-c2 Na7-c6 Nc2-e1 b5-b4 Ne1-d3 Rc5-a5 Nd3xb4 Nc6xb4 Rb3xb4 Ra5xa2
g3-g4) +0.23/36 27} Rd4 {(Rd8-d4 Ra1-c1 Rd4-d7 e2-e3 f7-f6 Kf1-e2 Kg8-f7
h2-h3 Nc6-a7 Rb5-b3 b6-b5 Rc1-c5 Ra6-c6 Rc5-c3 Rc6xc3 b2xc3 Rd7-c7 f2-f4
Rc7-c5 e3-e4 g6-g5 f4-f5 Kf7-e8 Ke2-d3 b7-b6 h3-h4 h7-h5 Rb3-b1 g5xh4 g3xh4
Ke8-f7 Rb1-b4 Na7-c6 Rb4xb5 Nc6-e5+ Kd3-c2 Rc5xb5 Na3xb5 Ne5-f3) -0.04/35
27} 21. Rc1 {(Ra1-c1 f7-f6 e2-e3 Rd4-d7 Kf1-e2 Kg8-f7 Rc1-c2 h7-h5 h2-h3
Nc6-a7 Rb5-b3 b6-b5 Rc2-c3 Ra6-a4 Na3xb5 Na7xb5 Rb3xb5 Ra4xa2 Rc3-b3 Rd7-c7
Ke2-f3 Rc7-c2 Rb5xb7 Ra2-a1 Rb3-c3 Rc2xc3 b2xc3 Ra1-c1 Rb7-c7 g6-g5 e3-e4
Kf7-e6 h3-h4 g5xh4 g3xh4 Ke6-d6 Rc7-c8 Kd6-d7 Rc8-c5 e7-e5 Kf3-e2 Rc1-h1
Rc5-d5+ Kd7-e6) +0.22/38 36} f6 {(f7-f6 e2-e3 Rd4-d7 Kf1-e2 h7-h5 Rc1-c3
Nc6-a7 Rb5-b3 Kg8-f7 Rb3-b4 Na7-c6 Rb4-b5 Nc6-a7 Rb5-b3 Na7-c6 Rc3-d3
Rd7xd3 Ke2xd3 Nc6-e5+ Kd3-e2 g6-g5 f2-f4 Ne5-d7 Rb3-b4 Kf7-g6 f4xg5 f6xg5
Na3-c4 Ra6xa2 Nc4xb6 Nd7xb6 Rb4xb6+ Kg6-f5 h2-h3 Ra2-a1 Rb6xb7 Ra1-c1
Rb7-b4 Rc1-c2+ Ke2-f3 g5-g4+ h3xg4+ h5xg4+ Rb4xg4 Rc2xb2 Rg4-f4+ Kf5-g6
Rf4-e4 Kg6-f6 Kf3-g4) -0.03/39 99} 22. e3 {(e2-e3 Rd4-d7 Kf1-e2 Kg8-f7
Rc1-c2 h7-h5 h2-h3 Nc6-a7 Rb5-b3 b6-b5 h3-h4 Ra6-a4 Rc2-c5 b7-b6 Rc5-c2
b5-b4 Rc2-c4 Na7-c6 Rc4xc6 b4xa3 b2xa3 Rd7-a7 Rc6-c3 Ra7-a5 Rc3-d3 g6-g5
h4xg5 f6xg5 Ke2-f3 b6-b5 Rd3-c3 Kf7-f6 Kf3-g2 Kf6-f7 Rc3-c5 Kf7-f6 Rc5-c6+
e7-e6 Rc6-b6) +0.13/38 23} Rd7 {(Rd4-d7 Rc1-c3 Kg8-f7 Kf1-e2 Nc6-a7 Rb5-b3
h7-h5 Rc3-d3 Kf7-e8 e3-e4 Na7-c6 Rd3xd7 Ke8xd7 Ke2-e3 Nc6-e5 f2-f3 Ra6-a4
h2-h3 g6-g5 h3-h4 g5xh4 g3xh4 e7-e6 f3-f4 Ne5-g4+ Ke3-f3 Ng4-h2+ Kf3-g3
Nh2-g4 Kg3-f3) 0.00/32 23} 23. Rc3 {(Rc1-c3 h7-h5 Kf1-e2 Kg8-f7 h2-h4
Nc6-a7 Rb5-b3 Na7-c6 Rc3-c2 Rd7-d5 f2-f4 Rd5-d7 Rc2-d2 Rd7xd2+ Ke2xd2
Kf7-e6 Kd2-e2 Ke6-d7 Ke2-f2 Nc6-a7 Na3-b1 Ra6xa2 Rb3xb6 Kd7-c7 Rb6-b3
Ra2-a1 Nb1-c3 Na7-c6 e3-e4 Ra1-h1 Nc3-d5+ Kc7-c8 Rb3-c3 Kc8-d7 Nd5-b6+
Kd7-c7) +0.15/36 22} h5 {(h7-h5 Kf1-e2 Kg8-f7 f2-f3 Nc6-a7 Rb5-b3 b6-b5
h2-h4 Ra6-a5 Rc3-c5 Ra5-a4 Na3xb5 Ra4xa2 Rc5-c7 Rd7xc7 Nb5xc7 Na7-c8 Rb3xb7
Nc8-d6 Rb7-b4 Ra2-a5 Rb4-b3 Ra5-a2 Rb3-b4) 0.00/36 30} 24. Ke2 {(Kf1-e2
Kg8-f7 Rc3-b3 Rd7-d6 Rb3-d3 Rd6xd3 Ke2xd3 Kf7-e8 f2-f4 Nc6-a7 Rb5-b3 Ke8-d7
Kd3-c3 b6-b5 Kc3-b4 Kd7-e6 Kb4-c5 Ke6-f5 Na3xb5 Na7xb5 Rb3xb5 Kf5-g4 Rb5xb7
Ra6xa2 Kc5-d4 Kg4-h3 Rb7xe7 Kh3xh2 Re7-g7 Ra2-a4+ Kd4-d3 Kh2xg3 Rg7xg6+
Kg3-f3 Rg6xf6 Ra4-e4 f4-f5 Re4xe3+ Kd3-c4 Re3-e4+ Kc4-c5 Re4-f4 Rf6-f8
h5-h4 f5-f6 Rf4-f5+ Kc5-d6) +0.23/38 44} Kf7 {(Kg8-f7 h2-h3 Nc6-a7 Rb5-b3
Na7-c6 Rc3-d3 Rd7xd3 Ke2xd3 Nc6-e5+ Kd3-e2 Ne5-d7 Rb3-b4 Nd7-c5 Ke2-d2
Kf7-e8 f2-f3 Ke8-d7 Rb4-d4+ Kd7-e8 Rd4-d5 g6-g5 Rd5-d4 Ra6-a8 Rd4-b4 Ke8-f7
Rb4-b5 Ra8-a6 Kd2-c3 Kf7-e8 h3-h4 g5xh4 g3xh4 Ke8-d7 Na3-c2 Ra6xa2 b2-b4
Nc5-a4+ Kc3-b3 Ra2-b2+ Kb3xa4 Rb2xc2 Rb5xh5 Kd7-c6 b4-b5+ Kc6-d6) -0.02/39
37} 25. h3 {(h2-h3 Nc6-a7 Rb5-b3 Na7-c6 Rc3-d3 Rd7xd3 Ke2xd3 Nc6-e5+ Kd3-e2
Kf7-e8 h3-h4 Ke8-d8 e3-e4 Ne5-f7 f2-f3 Nf7-d6 Ke2-d2 Kd8-d7 Kd2-c2 g6-g5
h4xg5 f6xg5 Na3-b5 Ra6xa2 Nb5xd6 e7xd6 Rb3xb6 Ra2-a1 Rb6xb7+ Kd7-c6 Rb7-g7
Ra1-f1 Rg7xg5 Rf1xf3 Kc2-c1 Rf3-f1+ Kc1-d2 Rf1-f3 Kd2-e2 Rf3-b3 Ke2-f1
Kc6-d7 Kf1-g1 Kd7-e6 Kg1-g2 Rb3xb2+ Kg2-h3) +0.11/36 61} Na7 {(Nc6-a7
Rb5-b3 Na7-c6 f2-f4 Rd7-d5 h3-h4 Rd5-d7 Rc3-d3 Kf7-e6 Rd3xd7 Ke6xd7 Ke2-f3
Nc6-a7 Rb3-d3+ Kd7-e6 Rd3-c3 Ke6-d7 e3-e4 Na7-c6 Kf3-e3 b6-b5 Rc3-d3+
Kd7-e8 Rd3-b3 b5-b4 Na3-c2 Ra6xa2 Nc2xb4 Ra2-a1 Nb4-d3 Ra1-g1 Ke3-f2 Rg1-d1
Kf2-e2 Rd1-g1 Ke2-f2) 0.00/42 76} 26. Rbb3 {(Rb5-b3 Na7-c6 Na3-c4 Ra6xa2
Nc4xb6 Rd7-d6 Nb6-c4 Rd6-d7 Rb3-b5 Nc6-a7 Rb5-b6 Ra2-a1 h3-h4 Na7-c6 Rb6-b3
Ra1-g1 Rc3-c2 Rg1-h1 Rb3-b5 Nc6-a7 Rb5-b4 Na7-c6 Rb4-b6 Rh1-a1 Nc4-d2
Nc6-e5 Nd2-f3 Ne5xf3 Ke2xf3 Ra1-d1 Rb6-b5 Rd1-e1 Rc2-c4 Re1-d1 Rb5-b3
Rd1-a1 Rc4-c2 Ra1-d1 Kf3-e2 Rd1-d6 Rb3-b5 Rd6-d5 Rb5-b4 Rd5-d3 Ke2-f3)
+0.10/45 74} Nc6 {(Na7-c6 Na3-c4 Ra6xa2 Nc4xb6 Rd7-d6 Nb6-c4 Rd6-d7 h3-h4
Kf7-e8 Nc4-d2 Ra2-a5 Nd2-e4 Ra5-e5 f2-f3 Rd7-c7 Rb3-b6 Re5-a5 Ke2-f2 Ra5-a1
f3-f4 Rc7-d7 Rc3-c2 Rd7-c7 Kf2-g2 Ra1-b1 Kg2-f2 Ke8-f7 Ne4-c5 Nc6-d8 Kf2-f3
Kf7-e8 Rc2-c3 Ke8-f7 Kf3-e2 Rb1-h1 Rb6-b4 Rh1-b1 Rb4-b5 Rb1-g1 Ke2-f3
Rg1-f1+ Kf3-e2) 0.00/41 17} 27. Nc4 {(Na3-c4 Ra6xa2 Nc4xb6 Rd7-d6 Nb6-c4
Rd6-d7 Nc4-d2 Ra2-a1 Nd2-f3 g6-g5 Nf3-e1 Ra1-b1 h3-h4 g5xh4 g3xh4 Nc6-e5
Ne1-d3 Ne5xd3 Rc3xd3 Rd7-c7 Rb3-b4 Rb1-h1 Rd3-c3 Rc7-c6 Rb4xb7 Rc6xc3 b2xc3
Rh1xh4 Rb7-b8 Rh4-h1 Ke2-f3 Rh1-c1 Rb8-h8 Rc1xc3 Rh8xh5 Rc3-c2 Rh5-b5
Kf7-e6 Rb5-b6+ Ke6-d7 Rb6-b7+ Kd7-e6 Rb7-b4 Ke6-d6 Rb4-d4+ Kd6-e6) +0.03/38
40} Rxa2 {(Ra6xa2 Nc4xb6 Rd7-d6 Nb6-c4 Rd6-d7 Nc4-d2 Ra2-a1 Nd2-f3 g6-g5
Rc3-d3 Rd7-c7 Rd3-c3 Rc7-d7) 0.00/44 22} 28. Nxb6 {(Nc4xb6 Rd7-d6 Nb6-c4
Rd6-d7 Rb3-b5 Nc6-a7 Rb5-b6 Ra2-a1 Rc3-a3 Ra1xa3 b2xa3 Na7-c6 g3-g4 h5xg4
h3xg4 g6-g5 a3-a4 Rd7-d5 Rb6xb7 Rd5-c5 Nc4-d6+ Kf7-f8 Nd6-f5 Rc5-c2+ Ke2-f3
Rc2-a2 a4-a5 Ra2xa5 Kf3-g3 Ra5-a4 f2-f3 Ra4-a7 Rb7-b6 Nc6-e5 Rb6-b8+ Kf8-f7
Rb8-h8 Ne5-c6 Rh8-h7+ Kf7-g8 Rh7-h1 Nc6-e5 Rh1-c1 Kg8-f7 f3-f4) +0.08/40
18} Rd6 {(Rd7-d6 Nb6-c4 Rd6-d7 Rb3-b5 Nc6-a7 Rb5-b6 Ra2-a1 Rc3-a3 Ra1xa3
b2xa3 Na7-c6 Rb6-b5 Nc6-d8 Nc4-a5 Rd7-c7 Ke2-d2 Kf7-e8 a3-a4 Rc7-d7+ Kd2-c3
Rd7-c7+ Kc3-d3 Rc7-d7+ Kd3-c3) 0.00/44 18} 29. Nc4 {(Nb6-c4 Rd6-d7 Rc3-c1)
+0.07/43 92} Rd7 {(Rd6-d7 Rc3-c1 Rd7-c7 Ke2-d1 Rc7-d7+ Kd1-e1 Nc6-a5 Nc4xa5
Ra2xa5 Ke1-e2 Ra5-d5 h3-h4 Rd5-d2+ Ke2-e1 Rd2-d5 Ke1-f1 Rd5-d6 Kf1-e2
Rd6-d2+) 0.00/41 20} 30. Rc1 {(Rc3-c1 Rd7-c7 Ke2-d1 Rc7-d7+ Kd1-e1 Nc6-a5
Nc4xa5 Ra2xa5 h3-h4 Ra5-d5 Ke1-f1 Rd5-d6 Rb3-b4 Rd6-c6 Rc1xc6 b7xc6 Kf1-e2
Rd7-d5 Rb4-c4 Rd5-b5 b2-b4 e7-e5 f2-f4 e5xf4 g3xf4 Kf7-e7 Ke2-d3 Ke7-d6
e3-e4 Kd6-d7 Kd3-e3 Kd7-c7 Rc4-d4 c6-c5 b4xc5 Rb5xc5) +0.08/36 17} Rc7
{(Rd7-c7 Ke2-d1 Rc7-d7+ Kd1-e1 Nc6-a5 Nc4xa5 Ra2xa5 h3-h4 Ra5-d5 Ke1-f1
Rd5-d6 Rb3-b4 b7-b6 Rc1-c2 Rd7-a7 Kf1-e2 Ra7-d7 Rb4-b5 Rd7-a7 Rb5-b4)
0.00/41 21} 31. Rb5 {(Rb3-b5 Nc6-a7 Rb5-b6 Na7-c8 Rb6-b4 Kf7-e6 Rc1-c3
Ra2-a6 f2-f4 Ra6-c6 g3-g4 h5xg4 h3xg4 b7-b5 Nc4-a3 Rc6xc3 b2xc3 Rc7xc3
Na3xb5 Rc3-c1 Ke2-f3 Nc8-d6 Nb5xd6 Ke6xd6 Rb4-b8 Kd6-e6 Rb8-g8 Ke6-f7
Rg8-h8 Rc1-f1+ Kf3-g2 Rf1-c1 Rh8-b8 Kf7-e6 Kg2-f3 Rc1-c7 Rb8-b6+) +0.05/36
16} Na7 {(Nc6-a7 Rb5-b4 Na7-c6 Rb4-b5) 0.00/40 18} 32. Rb6 {(Rb5-b6 Na7-c8
Rb6-b4 Kf7-e6 Rc1-c3 Ra2-a6 Rc3-b3 Ra6-c6 Nc4-a3 Nc8-d6 Na3-b5 Nd6xb5
Rb4xb5 Rc6-c2+ Ke2-f3 Rc7-d7 Rb5-b6+ Ke6-f7 Rb3-c3 Rc2xc3 b2xc3 Rd7-c7
c3-c4 Rc7xc4 Rb6xb7 Rc4-c1 Rb7-b4 Rc1-h1 h3-h4 Kf7-e6 Rb4-b7 Rh1-c1 Rb7-b6+
Ke6-f7 Rb6-b2 Rc1-d1 Rb2-b4 Rd1-d2 g3-g4 Rd2-a2 g4xh5 g6xh5) +0.04/38 17}
Nc8 {(Na7-c8 Rb6-b4 Kf7-e6 Rc1-c3 Nc8-d6 Ke2-d3 Ra2-a1 Nc4xd6 Rc7xc3+ b2xc3
Ke6xd6 Rb4xb7 Ra1-a2 f2-f4 Ra2-g2 g3-g4 h5xg4 h3xg4 Rg2xg4 c3-c4 g6-g5
f4-f5 Rg4-g1 c4-c5+ Kd6xc5 Rb7xe7 Kc5-d5 Re7-e6 Rg1-g3 Re6xf6 Kd5-e5 Rf6-f8
g5-g4 f5-f6 Ke5-e6 Rf8-b8 Ke6xf6 Rb8-b6+ Kf6-f5 Rb6-b5+ Kf5-g6 Rb5-b6+
Kg6-f7 Rb6-b7+ Kf7-g6) +0.04/37 18} 33. Rb4 {(Rb6-b4 Kf7-e6 Ke2-d3 Ra2-a6
Rc1-c3 Rc7-d7+ Kd3-e2 Rd7-c7 e3-e4 Ra6-c6 Ke2-d3 Rc7-d7+ Kd3-e3 Rd7-c7
Ke3-d3) 0.00/38 24} Ke6 {(Kf7-e6 Rc1-c3 Nc8-d6 Ke2-d3 Ra2-a6 e3-e4 Nd6xc4
Rc3xc4 Ke6-d7 Rc4xc7+ Kd7xc7 Rb4-c4+ Kc7-d6 Rc4-d4+ Kd6-c6 Rd4-c4+ Kc6-d6)
0.00/39 23} 34. Rc3 {(Rc1-c3 Ra2-a6 Rc3-b3 Ra6-c6 Nc4-a3 Nc8-d6 Na3-b5
Nd6xb5 Rb4xb5 Rc6-c2+ Ke2-f3 Rc7-d7 Rb5-b6+ Ke6-f7 Rb3-c3 Rc2xc3 b2xc3
Rd7-c7 c3-c4 Rc7xc4 Rb6xb7 h5-h4 g3xh4 Rc4xh4 Kf3-g2 Rh4-h5 Rb7-b4 Kf7-e6
h3-h4 Rh5-h8 Rb4-e4+ Ke6-f7 Re4-d4 f6-f5 Rd4-b4 Rh8-c8 Kg2-g3 Kf7-e6 f2-f3)
+0.04/39 60} Nd6 {(Nc8-d6 Ke2-d3 Ra2-a1 Nc4xd6 Rc7xc3+ Kd3xc3 Ke6xd6 Kc3-d3
Ra1-f1 Kd3-e2 Rf1-h1 h3-h4 Kd6-c6 Rb4-c4+ Kc6-d6 Rc4-d4+ Kd6-c7 Rd4-c4+
Kc7-d8 Rc4-d4+ Kd8-c7) 0.00/43 18} 35. Kd3 {(Ke2-d3 Ra2-a1 Nc4xd6 Rc7xc3+
b2xc3 Ke6xd6 Rb4xb7 Ra1-a2 f2-f4 Ra2-g2 g3-g4 h5xg4 h3xg4 Rg2xg4 c3-c4
Rg4-g1 c4-c5+ Kd6-e6 Kd3-c4 g6-g5 f4xg5 f6xg5 Kc4-d4 g5-g4 Kd4-e4 Rg1-c1
Rb7-b6+ Ke6-f7 Ke4-d5 g4-g3 Rb6-b2 Rc1-d1+ Kd5-e4 Rd1-c1 Rb2-g2 Rc1xc5
Rg2xg3 Rc5-b5 Rg3-h3 Kf7-e6 Rh3-h6+ Ke6-d7 Rh6-h8 Rb5-b4+ Ke4-f5 Rb4-b5+
Kf5-f4 Rb5-b2) +0.04/43 24} Ra6 {(Ra2-a6 e3-e4 Nd6xc4 Rc3xc4 Ke6-d7 Rc4-d4+
Kd7-e6 h3-h4 Ra6-a1 Rb4-b6+ Ke6-f7 Rb6-b4 Ra1-d1+ Kd3-e3 Rd1-e1+ Ke3-d3
Re1-c1 Rd4-c4 Rc7-d7+ Kd3-e2 Rc1-h1 Rc4-c3 Rh1-d1 Rb4-b6 Rd7-d2+ Ke2-e3
Rd2-d7 Ke3-e2) 0.00/38 19} 36. Nxd6 {(Nc4xd6 Ra6xd6+ Kd3-e2 Ke6-d7 g3-g4
h5xg4 Rb4xg4 Rc7xc3 b2xc3 Rd6-c6 Ke2-d3 Rc6-d6+ Kd3-c4 Rd6-c6+ Kc4-b3 g6-g5
Rg4-d4+ Kd7-e6 h3-h4 Rc6-b6+ Kb3-c2 g5xh4 Rd4xh4 Ke6-f5 Rh4-c4 Kf5-e6
Rc4-d4 Ke6-f5 Rd4-h4 Kf5-e6 Rh4-e4+ Ke6-d7 Re4-d4+ Kd7-c6 Rd4-g4 Rb6-a6
Rg4-e4 Kc6-d7 Kc2-b3 Kd7-d6 Re4-b4 Kd6-c7 Rb4-c4+ Kc7-d6) +0.04/41 26}
Rxd6+ {(Ra6xd6+ Kd3-e2 Ke6-d7 g3-g4 h5xg4 h3xg4 Rc7xc3 b2xc3 Kd7-c6 Rb4-c4+
Kc6-d7 Rc4-b4) 0.00/39 33} 37. Ke2 {(Kd3-e2 Rc7xc3 b2xc3 Rd6-c6 Ke2-d3
Rc6-d6+ Kd3-e4 Rd6-c6 g3-g4 h5xg4 h3xg4 Rc6xc3 Rb4xb7 Rc3-c4+ Ke4-f3 g6-g5
Kf3-g3 Rc4-c3 Rb7-b1 Ke6-d6 Rb1-b5 Kd6-e6 Rb5-b4 Ke6-d5 Rb4-d4+ Kd5-e5
Rd4-a4 Rc3-b3 Kg3-f3 Rb3-b7 Ra4-d4 Rb7-c7 Rd4-e4+ Ke5-d5 Re4-a4 Rc7-c1
Ra4-d4+ Kd5-e6 Kf3-g3 Rc1-g1+ Kg3-h3 Rg1-h1+) +0.04/37 30} Rxc3 {(Rc7xc3
b2xc3 Rd6-c6 Ke2-d3 Rc6-d6+ Kd3-e4 Rd6-c6 Ke4-d3) 0.00/43 18} 38. bxc3
{(b2xc3 Rd6-c6 Ke2-d3 Rc6-a6 Rb4xb7 Ra6-a2 f2-f4 Ra2-g2 g3-g4 h5xg4 h3xg4
Rg2xg4 Rb7-b6+ Ke6-d7 Rb6-b1 g6-g5 f4xg5 Rg4xg5 Rb1-a1 Kd7-c7 Kd3-e4 Kc7-b7
Ke4-d4 Rg5-g4+ Kd4-d3 Rg4-g5 Ra1-a4 Kb7-c6 Ra4-a6+ Kc6-d7 c3-c4 Rg5-g1
Kd3-d4 Rg1-e1 Ra6-a7+ Kd7-d6 c4-c5+ Kd6-e6 Ra7-b7 Re1-d1+ Kd4-c4 Rd1-c1+
Kc4-b5 Rc1-e1 Kb5-b6 Re1xe3) +0.04/40 37} Rc6 {(Rd6-c6 Ke2-d3 Rc6-a6 Rb4xb7
Ra6-a2 Rb7-b6+ Ke6-d7 Rb6-b7+ Kd7-e6) 0.00/43 19} 39. Rxb7 {(Rb4xb7 Rc6xc3
Rb7-b6+ Ke6-d7 Ke2-f3 Rc3-c2 Rb6-b7+ Kd7-d6 Rb7-b8 Kd6-e6 Rb8-b5 Rc2-c7
h3-h4 Rc7-c1 Rb5-b8 Rc1-d1 Rb8-h8 Ke6-f7 g3-g4 h5xg4+ Kf3xg4 Rd1-h1 Rh8-h7+
Kf7-e6 Rh7-g7 Rh1-h2 Kg4-g3 Rh2-h1 Rg7-g8 Rh1-g1+ Kg3-f3 Ke6-f7 Rg8-c8
Kf7-g7 Rc8-c4 f6-f5 Rc4-b4 Rg1-c1 Kf3-g2 Rc1-c2 Kg2-g3 Rc2-c1 Rb4-b7)
+0.04/43 85} Rxc3 {(Rc6xc3 Rb7-b6+ Ke6-f5 Rb6-b5+ Kf5-e6 Rb5-b6+) 0.00/38
23} 40. Rb6+ {(Rb7-b6+ Ke6-f5 Rb6-b5+ Kf5-e6 Rb5-a5 Rc3-c4 Ke2-f3 Rc4-b4
g3-g4 h5xg4+ h3xg4 g6-g5 Ra5-a3 Rb4-b1 Kf3-g2 Rb1-c1 Ra3-b3 Ke6-f7 Kg2-f3
Rc1-c6 e3-e4 Rc6-c1 Rb3-b6 Rc1-c4 Kf3-e3 Rc4-c1 Ke3-d3 Rc1-f1 Kd3-e2 Rf1-g1
Ke2-f3 Rg1-f1 Rb6-b7 Rf1-c1 Rb7-b3 Kf7-e6 Rb3-b6+ Ke6-e5 Rb6-b5+ Ke5-e6
Rb5-f5) +0.03/39 19} Kd7 {(Ke6-d7 Ke2-f3 Rc3-c2 Rb6-b8 Kd7-e6 Rb8-b5 Ke6-f7
g3-g4 h5xg4+ h3xg4 Rc2-d2 Rb5-b4 Rd2-a2 Rb4-b5 Ra2-a4 Kf3-g3 g6-g5 Kg3-f3
Ra4-a7 Rb5-b4 Kf7-g7 Rb4-b8 Ra7-a5 Rb8-b3 f6-f5 g4xf5 Ra5xf5+ Kf3-g3 Rf5-e5
Kg3-g4 Kg7-f6 Rb3-b8 Re5-e4+ Kg4-h3 Re4-h4+ Kh3-g3 Rh4-h1 Kg3-f3 Rh1-e1
Rb8-b6+ Kf6-f5 Rb6-b5+ Kf5-g6) +0.05/41 146} 41. Kf3 {(Ke2-f3 Rc3-c2 Rb6-b5
Kd7-e8 Rb5-b8+ Ke8-f7 Rb8-b5 Rc2-d2 h3-h4 Rd2-d6 Rb5-c5 Rd6-d1 e3-e4 Rd1-b1
e4-e5 f6-f5 Kf3-f4 Rb1-b2 Kf4-f3 Rb2-b1) 0.00/32 10} Rc4 {(Rc3-c4 Rb6-b5
Kd7-e6 h3-h4 Rc4-c2 e3-e4 Rc2-c3+ Kf3-e2 Rc3-c7 Ke2-e3 Rc7-a7 Rb5-b6+
Ke6-f7 Rb6-b5 Ra7-a3+ Ke3-e2 Kf7-g7 f2-f3 Kg7-f7 Ke2-f2 Ra3-a6 Kf2-f1
Ra6-a2 Kf1-g1 Ra2-a1+ Kg1-f2 Ra1-a7 Kf2-e2 Ra7-a2+ Ke2-f1 Ra2-a4 Kf1-f2
Ra4-a3 Kf2-g2 Kf7-e6 Kg2-f2 Ra3-a2+ Kf2-g1 Ra2-a6 Kg1-f1 Ke6-f7 Kf1-f2
Ra6-a1) +0.05/41 33} 42. Ra6 {(Rb6-a6 Rc4-c5 Kf3-g2 Rc5-b5 Ra6-a8 Kd7-e6
Ra8-h8 Rb5-c5 Rh8-h7 f6-f5 Rh7-h6 Ke6-f7 Rh6-h8 Kf7-g7 Rh8-a8 Rc5-c4 Kg2-f3
Kg7-f7 Ra8-a2 Kf7-e6 Ra2-b2 Ke6-f7 Rb2-d2 Kf7-e6 Rd2-a2 Ke6-f6 Kf3-g2
Kf6-g7 Ra2-a6 Kg7-f7 Ra6-a7 Kf7-e6 Kg2-f3 Ke6-f7 Ra7-a1 Kf7-e6 Ra1-g1
Ke6-f6 Rg1-e1) +0.03/39 42} Rc5 {(Rc4-c5 Kf3-g2 Rc5-b5 Ra6-a8 Kd7-e6 Ra8-h8
Ke6-f5 Kg2-f3 e7-e5 e3-e4+ Kf5-e6 h3-h4 Rb5-b3+ Kf3-e2 f6-f5 e4xf5+ Ke6xf5
Rh8-a8 Rb3-b6 Ra8-a4 Kf5-f6 Ke2-f3 Rb6-c6 Ra4-b4 Kf6-f5 Rb4-b7 Rc6-d6
Rb7-b5 Rd6-a6 Rb5-b7 Ra6-c6 Rb7-b5 Kf5-e6 Rb5-a5 Rc6-c4 Ra5-b5 Ke6-f6)
+0.04/36 17} 43. Kg2 {(Kf3-g2 Rc5-b5 Ra6-a8 Kd7-e6 Ra8-a4 Rb5-b2 Ra4-h4
Ke6-f7 g3-g4 h5xg4 h3xg4 g6-g5 Rh4-h7+ Kf7-e6 Rh7-h1 Ke6-e5 Rh1-c1 Rb2-b3
Rc1-c5+ Ke5-d6 Rc5-c2 Kd6-e5 Kg2-f3 Ke5-e6 Kf3-g3 Ke6-d5 Rc2-c7 Kd5-d6
Rc7-c4 Kd6-e5 Kg3-f3 e7-e6 Rc4-c2 Rb3-b4 Rc2-c1 Ke5-d6 Rc1-c3 Kd6-e5)
+0.03/36 10} g5 {(g6-g5 Ra6-a8 Kd7-e6 Ra8-h8 g5-g4 Rh8-h7 Rc5-e5 Rh7-g7
Ke6-f5 Rg7-g8 Re5-a5 Rg8-g7 Ra5-a7 Rg7-g8 Ra7-a2 Rg8-g7 e7-e6 h3xg4+ h5xg4
e3-e4+ Kf5xe4 Rg7xg4+ Ke4-e5 Rg4-g7 f6-f5 Rg7-g8 Ra2-c2 g3-g4 f5xg4 Rg8xg4
Ke5-f5 Rg4-g8 Rc2-c7 Kg2-f3 Rc7-d7 Kf3-g3 Rd7-d3+ Kg3-g2) +0.06/37 26} 44.
Kf3 {(Kg2-f3 Rc5-f5+ Kf3-g2 Rf5-c5) 0.00/37 12} Rf5+ {(Rc5-f5+ Kf3-g2 g5-g4
Ra6-a4 g4xh3+ Kg2-f1 Kd7-e8 Ra4-h4 Ke8-f7 Rh4xh3 Kf7-g6 Rh3-h4 e7-e6 Rh4-e4
Kg6-f7 Re4-h4 Rf5-b5 Kf1-g2 Kf7-g6 Rh4-c4 Rb5-b2 Rc4-h4 Rb2-c2 Kg2-f3 e6-e5
Kf3-g2 Kg6-g5 Rh4-a4 Rc2-d2 Ra4-h4 Kg5-g6 Rh4-c4 Rd2-d3 Rc4-c8 Rd3-d1
Rc8-h8 Rd1-d2 Rh8-c8 Kg6-f7 Rc8-h8) +0.04/38 22} 45. Kg2 {(Kf3-g2 Rf5-c5
Kg2-f3 Rc5-f5+) 0.00/37 9} Rd5 {(Rf5-d5 Ra6-a8 Kd7-e6 Ra8-h8 g5-g4 Rh8-h6
Rd5-a5 e3-e4 Ra5-a4 Rh6xh5 Ra4xe4 h3xg4 Re4xg4 Kg2-f3 Rg4-g8 Kf3-f4 Rg8-b8
Kf4-g4 Rb8-b2 f2-f4 Rb2-b7 Rh5-h7 Rb7-b5 f4-f5+ Ke6-d7 Rh7-h2 e7-e6 Rh2-h5
Rb5-b2 Kg4-f3 Kd7-d6 Rh5-h6 Kd6-e5 f5xe6 Ke5xe6 Kf3-e3) +0.06/37 35} 46.
Kf3 {(Kg2-f3 Rd5-f5+ Kf3-g2 Rf5-c5) 0.00/39 16} Rf5+ {(Rd5-f5+ Kf3-g2 g5-g4
h3xg4 h5xg4 Ra6-a4 Rf5-g5 Kg2-f1 Kd7-d6 Ra4-e4 Rg5-g7 Re4-c4 Kd6-e5 Rc4-c5+
Ke5-e4 Rc5-c4+ Ke4-d5 Rc4-b4 f6-f5 Kf1-e2 Rg7-f7 Rb4-f4 Kd5-e5 e3-e4 e7-e6
e4xf5 e6xf5 Rf4-b4 Rf7-a7 Rb4-b6 Ra7-c7 Rb6-b8 Rc7-c3 Rb8-e8+ Ke5-d4
Re8-d8+ Kd4-e4 Rd8-e8+ Ke4-d5 Re8-f8 Kd5-e6 Rf8-e8+ Ke6-f7 Re8-a8 Rc3-b3
Ra8-a6 Rb3-b5) +0.05/44 43} 47. Kg2 {(Kf3-g2 Rf5-b5 Ra6-a8 Kd7-e6 Ra8-h8
g5-g4 Rh8-h7 Rb5-e5 h3-h4 Ke6-d6 f2-f4 Re5xe3 Rh7xh5 Re3-e2+ Kg2-g1 Re2-e3
Kg1-f2 Re3-f3+ Kf2-g2 Rf3-b3 Rh5-a5 Rb3-b2+ Kg2-f1 Rb2-b3 Kf1-f2 Rb3-b2+
Kf2-e3 Rb2-b3+ Ke3-e4 Kd6-e6 f4-f5+ Ke6-f7 Ke4-f4 Rb3-b4+ Kf4-e3 Rb4-b3+
Ke3-f4) 0.00/39 29} Rb5 {(Rf5-b5 Ra6-a8 Kd7-e6 Ra8-h8 g5-g4 Rh8-h7 Rb5-e5
h3-h4 Ke6-d6 Kg2-f1 Kd6-d5 Kf1-e1 Kd5-c6 Ke1-f1 Kc6-d7 Rh7-h8 Re5-a5 Kf1-g2
e7-e6 Kg2-f1 Kd7-e7 Rh8-h7+ Ke7-f8 f2-f4 g4xf3/ep Kf1-f2 Ra5-c5 Kf2xf3
Rc5-f5+ Kf3-e4 Rf5-e5+ Ke4-f4 Re5-a5 Rh7-b7 Ra5-c5 Kf4-e4 Kf8-e8 Ke4-f3
Rc5-f5+ Kf3-e2 Ke8-f8 Rb7-h7 Rf5-b5 Ke2-f3 Rb5-c5) +0.01/43 11} 48. Ra8
{(Ra6-a8 Kd7-e6 f2-f3 Rb5-b2+ Kg2-f1 Rb2-b1+ Kf1-e2 Rb1-b2+ Ke2-f1) 0.00/39
7} Ke6 {(Kd7-e6 Ra8-h8 g5-g4 Rh8-h7 Rb5-e5 h3-h4 Ke6-d6 Kg2-f1 Kd6-d5
Kf1-e1 Kd5-c6 Ke1-f1 Kc6-d7 Rh7-h8 e7-e6 Kf1-e2 Re5-a5 Rh8-h7+ Kd7-e8 f2-f4
Ke8-f8 Ke2-f1 Kf8-g8 Rh7-e7 Ra5-a1+ Kf1-f2 Ra1-a6 Kf2-e2 Ra6-b6 Ke2-f2
Rb6-c6 Kf2-f1 Kg8-f8 Re7-h7 Rc6-c3) +0.01/34 25} 49. Rh8 {(Ra8-h8 g5-g4
Rh8-h7 Rb5-e5 h3-h4 Ke6-d6 e3-e4 Kd6-d7 f2-f3 g4xf3+ Kg2xf3 Re5-c5 Kf3-f4
Kd7-e8 g3-g4 h5xg4 Kf4xg4 Rc5-e5 Kg4-f3 Re5-c5 Kf3-g4) 0.00/43 13} g4
{(g5-g4 Rh8-h7 Rb5-e5 h3-h4 Ke6-d6 Kg2-f1 Kd6-d5 Kf1-e1 Kd5-e6 Ke1-f1
Re5-c5 Kf1-g2 Rc5-f5 Kg2-f1 Ke6-d7 Rh7-h8 Kd7-e6 Rh8-h7) 0.00/48 9} 50. Rh7
{(Rh8-h7 Rb5-e5 h3-h4 Ke6-d6 Kg2-f1 Kd6-d5 Kf1-e1 Kd5-e6 Ke1-f1 Re5-c5
Kf1-g2 Rc5-f5 Kg2-f1 Ke6-d7 Rh7-h8 Kd7-e6 Rh8-h7) 0.00/48 41} Re5 {(Rb5-e5
h3-h4 Re5-b5 Kg2-f1 Rb5-c5 Kf1-g2 Rc5-f5 Kg2-f1 Ke6-d7 Rh7-h8 e7-e6 Rh8-h7+
Kd7-e8 Kf1-g2 Rf5-c5 Kg2-f1 Ke8-f8 f2-f4 Rc5-f5 e3-e4 Rf5-a5 Kf1-g2 Kf8-e8
Kg2-f1 Ra5-b5 Kf1-f2 Rb5-b2+ Kf2-f1 Rb2-b1+ Kf1-f2 Rb1-b5 Kf2-f1 Rb5-c5
Kf1-g2 Ke8-f8 Kg2-f2 Rc5-b5 Kf2-g2 Rb5-b2+ Kg2-f1 Rb2-b3 Kf1-g2 Rb3-b5
Kg2-f1 Kf8-g8 Rh7-e7 Rb5-b6 Kf1-g2) +0.01/47 27} 51. h4 {(h3-h4 Ke6-d7
Rh7-h8 e7-e6 Rh8-h7+ Kd7-e8 Rh7-g7 Re5-b5 Rg7-c7 Ke8-f8 f2-f4 Rb5-d5 Kg2-f1
e6-e5 Rc7-h7 e5xf4 e3xf4 Rd5-c5 Rh7-d7 Rc5-c1+ Kf1-g2 Rc1-c2+ Kg2-f1
Rc2-c1+) 0.00/47 8} Kd7 {(Ke6-d7 Rh7-h8 e7-e6 Rh8-h7+ Kd7-e8 Rh7-g7 Re5-b5
Rg7-c7 Ke8-f8 f2-f4 Rb5-d5 Kg2-f1 e6-e5 Rc7-h7 e5xf4 e3xf4 Rd5-c5 Rh7-d7
Rc5-c1+ Kf1-g2 Rc1-c2+ Kg2-f1 Rc2-c1+) 0.00/51 13} 52. Rh8 {(Rh7-h8 e7-e6
Rh8-h7+ Kd7-e8 Rh7-g7 Re5-b5 Rg7-h7 Ke8-f8 f2-f4 Rb5-c5 Kg2-f1 Kf8-g8
Rh7-e7 Rc5-c1+ Kf1-f2 Rc1-c6 Kf2-g2 f6-f5 Kg2-f2 Rc6-c2+ Kf2-f1 Rc2-c6
Kf1-f2) 0.00/53 10} e6 {(e7-e6 Rh8-h6 Kd7-e7 Rh6-h7+ Ke7-f8 Rh7-c7 Kf8-g8
Rc7-e7 Re5-e4 Re7-c7 Re4-e5) 0.00/52 52} 53. Rg8 {(Rh8-g8 Re5-b5 Rg8-g7+
Kd7-d6 Rg7-h7 Rb5-a5 Rh7-h8 Kd6-e7 Rh8-h7+ Ke7-d6) 0.00/62 7} Rb5 {(Re5-b5
Rg8-g7+ Kd7-d6 Rg7-h7 Rb5-a5 Rh7-h8 Kd6-e7 Rh8-h7+ Ke7-d6) 0.00/63 10} 54.
Rg7+ {(Rg8-g7+ Kd7-d6 Rg7-h7 Rb5-a5 Rh7-h8 Kd6-e7 Rh8-h7+ Ke7-d6) 0.00/59
21} Kd6 {(Kd7-d6 Rg7-h7 Rb5-a5 Rh7-h8 Kd6-e7 Rh8-h7+ Ke7-d6) 0.00/62 7} 55.
Rh7 {(Rg7-h7 Rb5-a5 Rh7-h8 Kd6-e7 Rh8-h7+ Ke7-d6) 0.00/64 10} Ra5 {(Rb5-a5
Rh7-h8 Kd6-e7 Rh8-h7+ Ke7-d6) 0.00/64 7} 56. Rh8 {(Rh7-h8 Kd6-e7 Rh8-h7+
Ke7-d6) 0.00/62 8} Ke7 {(Kd6-e7 Rh8-h7+ Ke7-d6 Rh7-h8) 0.00/66 10} 57. Rh7+
{(Rh8-h7+ Ke7-d6 Rh7-h8 Kd6-e7) 0.00/65 11} Kd6 {(Ke7-d6 Rh7-h8 Kd6-e7
Rh8-h7+) 0.00/69 10} 58. Rh8 {(Rh7-h8 Kd6-e7 Rh8-h7+ Ke7-d6) 0.00/68 10}
Ke7 {(Kd6-e7 Rh8-h7+ Ke7-d6) 0.00/69 7} 59. Rh7+ {(Rh8-h7+ Ke7-d6) 0.00/69
6} Kd6 {(Ke7-d6) 0.00/73 10 3-fold repetition} *[/pgn]
Uri Blass
Posts: 11150
Joined: Thu Mar 09, 2006 12:37 am
Location: Tel-Aviv Israel

Re: Testing depend largely on good Openings

Post by Uri Blass »

I see score of more than +2 by stockfish even for opening with no material winning.

For example 1.d4 c5 2.d5 d6 3.e4 e5 and now
[fen]rnbqkbnr/pp3ppp/3p4/2pPp3/4P3/8/PPP2PPP/RNBQKBNR w KQkq e6 0 4[/fen]

Stockfish_22070513_x64_avx2:
NNUE evaluation using nn-3c0054ea9860.nnue enabled
1/1 00:00 340 170k +2.13 a2-a4
2/2 00:00 858 429k +2.13 a2-a4 Bf8-e7
3/3 00:00 1k 647k +2.13 a2-a4 Bf8-e7 h2-h3
4/4 00:00 2k 813k +2.17 a2-a4 Bf8-e7 h2-h3 Ng8-f6
5/5 00:00 2k 1,013k +2.17 a2-a4 Bf8-e7 h2-h3 Ng8-f6 Nb1-d2
6/6 00:00 3k 1,305k +2.17 a2-a4 Bf8-e7 h2-h3 Ng8-f6 Nb1-d2 Nb8-a6
7/11 00:00 12k 4,095k +1.97 Ng1-f3 Bf8-e7 h2-h3 Nb8-a6 a2-a4 Bc8-d7 Nb1-c3 Na6-b4
8/13 00:00 16k 5,349k +2.13 a2-a4 Bf8-e7 h2-h3 Ng8-f6 Qd1-f3 Nb8-a6 Nb1-d2
9/13 00:00 19k 6,305k +2.02 a2-a4 Bf8-e7 h2-h3 Ng8-f6 Nb1-d2 Nb8-a6 Qd1-f3 Na6-c7 g2-g4 O-O Ng1-e2 Ra8-b8 Ne2-g3
10/13 00:00 23k 7,573k +2.06 a2-a4 Ng8-f6 Qd1-f3 Bf8-e7 h2-h3 Nb8-a6 Nb1-d2 Na6-c7 Ng1-e2 O-O g2-g4 Ra8-b8 Nd2-c4 Nf6-e8
11/15 00:00 39k 7,842k +2.08 a2-a4 Ng8-f6 Nb1-d2 Bf8-e7 h2-h3 Nb8-a6 Qd1-f3 O-O g2-g4 Na6-c7 Ng1-e2 h7-h6 Qf3-g2 Bc8-d7 Ne2-g3 Nf6-h7
12/18 00:00 44k 8,743k +2.08 a2-a4 Bf8-e7 Nb1-d2 Ng8-f6 h2-h3 Nb8-a6 Qd1-f3 O-O g2-g4 Na6-c7 Ng1-e2 h7-h6 Qf3-g2 Bc8-d7 Ne2-g3 Nf6-h7 Nd2-f3 b7-b5 a4xb5
13/18 00:00 49k 9,731k +2.08 a2-a4 Bf8-e7 Nb1-d2 Ng8-f6 h2-h3 Nb8-a6 Qd1-f3 O-O g2-g4 Na6-c7 Ng1-e2 h7-h6 Qf3-g2 Bc8-d7 Ne2-g3 Nf6-h7 Nd2-f3 b7-b5 a4xb5
14/18 00:00 54k 10,722k +2.08 a2-a4 Bf8-e7 Nb1-d2 Ng8-f6 h2-h3 Nb8-a6 Qd1-f3 O-O g2-g4 Na6-c7 Ng1-e2 h7-h6 Qf3-g2 Bc8-d7 Ne2-g3 Nf6-h7 Nd2-f3 b7-b5 a4xb5
15/17 00:00 61k 10,202k +2.08 a2-a4 Bf8-e7 Nb1-d2 Ng8-f6 h2-h3 Nb8-a6 Qd1-f3 O-O g2-g4 Na6-c7 Ng1-e2 h7-h6 Qf3-g2 Bc8-d7 Ne2-g3 Nf6-h7 Nd2-f3
16/18 00:00 76k 9,555k +2.08 a2-a4 Bf8-e7 Nb1-d2 Ng8-f6 h2-h3 Nb8-a6 Qd1-f3 O-O g2-g4 Na6-c7 Ng1-e2 h7-h6 Qf3-g2 Bc8-d7 Ne2-g3 Nf6-h7 Nd2-f3 b7-b5 a4xb5
17/18 00:00 84k 10,447k +2.08 a2-a4 Bf8-e7 Nb1-d2 Ng8-f6 h2-h3 Nb8-a6 Qd1-f3 O-O g2-g4 Na6-c7 Ng1-e2 h7-h6 Qf3-g2 Bc8-d7 Ne2-g3 Nf6-h7 Nd2-f3 b7-b5
18/32 00:00 301k 8,137k +2.01 a2-a4 Ng8-f6 Nb1-d2 Bf8-e7 h2-h3 Nb8-a6 Qd1-f3 O-O Bf1xa6 b7xa6 g2-g4 a6-a5 Ng1-e2 Nf6-e8 Ne2-c3 Be7-g5 h3-h4 Bg5-f4 Nc3-e2 h7-h5
19/29 00:00 541k 7,613k +2.04 a2-a4 Ng8-f6 Nb1-d2 Bf8-e7 h2-h3 Nb8-a6 Qd1-f3 O-O g2-g4 Na6-c7 Ng1-e2 Ra8-b8 Nd2-c4 h7-h6 Ne2-g3 Nf6-h7 Qf3-g2 Nc7-e8 Bf1-e2 Be7-h4 Ng3-f5
20/25 00:00 684k 7,681k +2.04 a2-a4 Ng8-f6 Nb1-d2 Bf8-e7 h2-h3 Nb8-a6 Qd1-f3 O-O g2-g4 Na6-c7 Ng1-e2 Ra8-b8 Nd2-c4 h7-h6 Ne2-g3 Nf6-h7 Qf3-g2 Nc7-e8 Bf1-e2 Be7-h4 Ng3-f5
21/28 00:00 810k 7,713k +1.99 a2-a4 Ng8-f6 Nb1-d2 Bf8-e7 h2-h3 Nb8-a6 Qd1-f3 O-O g2-g4 Na6-c7 Ng1-e2 Bc8-d7 Nd2-c4 h7-h6 Ne2-g3 Nf6-h7 h3-h4 b7-b5 a4xb5 Nc7xb5 g4-g5 Nb5-d4 Qf3-d1 Ra8-b8 g5xh6
22/32 00:00 1,793k 7,933k +1.96 Ng1-e2 Ng8-f6 Ne2-c3 Bf8-e7 Bf1-b5+ Nb8-d7 a2-a4 O-O Qd1-e2 Nf6-e8 Bb5-d3 a7-a6 Nb1-d2 Ra8-b8 b2-b3 g7-g6 a4-a5 Be7-g5 Ke1-d1 f7-f5 h2-h4 Bg5xd2 Bc1xd2 Nd7-f6 Kd1-c1 Ne8-g7 Bd2-g5 Ng7-h5 Kc1-b2 Qd8-c7
23/26 00:00 1,909k 7,955k +1.97 Ng1-e2 Ng8-f6 Ne2-c3 Bf8-e7 Bf1-b5+ Nb8-d7 a2-a4 O-O Qd1-e2 Nf6-e8 Bb5-d3 Be7-g5 Nb1-d2 Nd7-f6 g2-g3 Ne8-c7 f2-f3 a7-a6 h2-h4 Bg5-h6 g3-g4 Bh6-f4 Qe2-f2 Nf6-d7
24/28 00:00 2,010k 7,883k +2.02 Ng1-e2 Ng8-f6 Ne2-c3 Bf8-e7 Bf1-b5+ Nb8-d7 a2-a4 O-O Qd1-e2 Nf6-e8 Bb5-d3 Be7-g5 Nb1-d2 Nd7-f6 g2-g3 a7-a6 a4-a5 Ne8-c7 f2-f3 Nc7-b5 Bd3xb5 a6xb5 Qe2xb5 Bc8-d7 Qb5xb7 Ra8xa5 Ra1xa5 Qd8xa5
25/33 00:00 2,671k 7,719k +2.01 Ng1-e2 Ng8-f6 Ne2-c3 Bf8-e7 Bf1-b5+ Nb8-d7 a2-a4 O-O Qd1-e2 Nf6-e8 Bb5-d3 Be7-g5 Nb1-d2 Nd7-f6 b2-b3 Ne8-c7 Bc1-b2 Nf6-e8 h2-h4 Bg5xd2+ Qe2xd2 f7-f5 h4-h5 h7-h6 O-O-O f5xe4 Nc3xe4
26/32 00:00 3,779k 7,713k +2.00 Ng1-e2 Ng8-f6 Ne2-c3 Bf8-e7 Bf1-b5+ Nb8-d7 a2-a4 O-O Qd1-e2 Nf6-e8 Bb5-d3 Be7-g5 Nb1-d2 Nd7-f6 b2-b3 Ne8-c7 Bc1-b2 Bg5-h6 f2-f3 b7-b6 Nd2-c4 Nf6-e8 O-O f7-f5 e4xf5 Bc8xf5 Bd3xf5
27/38 00:00 5,326k 7,620k +2.02 a2-a4 Ng8-f6 Nb1-d2 Bf8-e7 h2-h3 Nb8-a6 Qd1-f3 O-O g2-g4 Na6-c7 Bf1-d3 b7-b6 Nd2-c4 Ra8-b8 g4-g5 Nf6-e8 h3-h4 a7-a6 a4-a5 b6-b5 Nc4-b6 f7-f6 Qf3-g2 c5-c4 Bd3-e2 f6-f5 Ng1-h3 f5-f4 Nb6xc8 Qd8xc8 Be2-g4
28/37 00:00 6,462k 7,549k +2.12 a2-a4 Ng8-f6 Nb1-d2 Bf8-e7 h2-h3 Nb8-a6 Ng1-e2 O-O Ne2-g3 Nf6-e8 Nd2-c4 Be7-g5 Bc1xg5 Qd8xg5 h3-h4 Qg5-d8 Nc4-e3 Ra8-b8 Qd1-d2 Bc8-d7 Bf1-e2 Kg8-h8 f2-f3 Na6-c7 b2-b3 b7-b6 Ke1-d1 a7-a6
29/38 00:01 10,747k 7,356k +2.14 a2-a4 Ng8-f6 Nb1-d2 Bf8-e7 h2-h3 Nb8-a6 Qd1-f3 O-O g2-g4 Na6-c7 Nd2-c4 Bc8-d7 Ng1-e2 b7-b5 a4xb5 Nc7xb5 Ne2-g3 a7-a5 c2-c3 a5-a4 g4-g5 Nf6-e8 h3-h4 a4-a3 b2xa3 f7-f6 Qf3-e3 Ra8-a4 Bf1-e2 g7-g6 Qe3-d2 f6-f5 e4xf5 g6xf5
30/41 00:01 14,112k 7,286k +2.13 a2-a4 Ng8-f6 Nb1-d2 Bf8-e7 h2-h3 Nb8-a6 Qd1-f3 Na6-c7 g2-g4 h7-h6 Nd2-c4 O-O Bf1-d3 Bc8-d7 Qf3-g2 b7-b5 a4xb5 Nc7xb5 f2-f4 e5xf4 Bc1xf4 a7-a5 c2-c3 a5-a4 Ng1-f3 a4-a3 O-O a3xb2 Ra1xa8 Qd8xa8 Qg2xb2 Qa8-b7 Kg1-h1 Rf8-b8 Bf4-h2 Nf6-h7 Qb2-d2
31/45- 00:08 56,685k 6,439k +2.00 a2-a4 Bf8-e7
31/49+ 00:09 63,358k 6,393k +2.13 a2-a4
31/49 00:10 69,701k 6,368k +2.22 a2-a4 Bf8-e7 Nb1-d2 Ng8-f6 h2-h3 Nb8-a6 Qd1-f3 O-O g2-g4 Na6-c7 Nd2-c4 Bc8-d7 Ng1-e2 b7-b5 a4xb5 Nc7xb5 c2-c3 h7-h6 Ne2-g3 Nf6-h7 Ng3-f5 a7-a5 Nf5xe7+ Qd8xe7 Nc4-b6 Ra8-b8 Nb6xd7 Qe7xd7 h3-h4 a5-a4 g4-g5 f7-f5
32/40- 00:11 74,442k 6,342k +2.05 a2-a4 Bf8-e7
32/48+ 00:22 141,104k 6,185k +2.18 Nb1-d2
32/48 00:24 147,578k 6,147k +2.14 Nb1-d2 Bf8-e7 a2-a4 Ng8-f6 h2-h3 Nb8-a6 Qd1-f3 Ra8-b8 g2-g4 O-O Nd2-c4 Bc8-d7 c2-c3 Na6-c7 Nc4-e3 g7-g6 Bf1-d3 b7-b6 h3-h4 a7-a6 Qf3-g2 b6-b5 h4-h5 Kg8-h8 Ng1-f3 c5-c4 Bd3-c2 b5-b4 c3xb4 Rb8xb4
33/42 00:24 152,117k 6,131k +2.15 Nb1-d2 Bf8-e7 a2-a4 Ng8-f6 h2-h3 Nb8-a6 Qd1-f3 Ra8-b8 g2-g4 O-O Nd2-c4 Bc8-d7 Ng1-e2 Na6-c7 Ne2-g3 Nf6-e8 Bf1-e2 Be7-h4 Rh1-g1 Bh4-g5 Bc1xg5 Qd8xg5 Qf3-e3 Qg5-d8 Ke1-d2 a7-a6 Nc4-b6 Nc7-a8 Nb6xd7 Qd8xd7 h3-h4 b7-b5 a4xb5 a6xb5 Ng3-f5
34/43- 00:28 170,859k 6,056k +2.02 Nb1-d2 Bf8-e7
34/43+ 00:32 198,138k 6,034k +2.15 Nb1-d2
34/43 00:34 207,883k 6,006k +2.05 Nb1-d2 Bf8-e7 a2-a4 Ng8-f6 h2-h3 O-O Bf1-d3 Nb8-a6 Ng1-e2 Na6-b4 Bd3-c4 Kg8-h8 Ne2-g3 Nf6-g8 O-O Be7-h4 Nd2-f3 Bh4-e7 Nf3-h2 Be7-g5 f2-f4 Bg5xf4 Bc1xf4 e5xf4 Rf1xf4 Ng8-e7 Rf4-f1 Ne7-g6 Nh2-f3 f7-f6 Qd1-d2 Nb4-a6 a4-a5 Bc8-d7 b2-b3 Na6-b4
35/47 00:50 300,650k 6,002k +2.19 Nb1-d2 Bf8-e7 a2-a4 Ng8-f6 Bf1-d3 O-O Ng1-e2 Nf6-e8 O-O Nb8-a6 h2-h3 Be7-g5 c2-c3 Na6-b8 a4-a5 Nb8-d7 Nd2-c4 Ra8-b8 b2-b4 g7-g6 b4-b5 Qd8-e7 Ne2-g3 Bg5-f4 Qd1-f3 Qe7-f6 Ng3-e2 Bf4xc1 Qf3xf6 Nd7xf6 Ne2xc1 Nf6-d7 Nc1-b3 f7-f5
36/47- 00:53 319,477k 6,013k +2.02 Nb1-d2 Bf8-e7
36/47+ 00:53 323,574k 6,014k +2.15 Nb1-d2
36/47 00:57 344,992k 5,975k +2.01 Nb1-d2 Bf8-e7 a2-a4 Ng8-f6 Bf1-d3 O-O Ng1-e2 Nf6-e8 O-O Be7-g5 Nd2-c4 Bg5xc1 Qd1xc1 b7-b6 Ne2-c3 a7-a6 f2-f4 Nb8-d7 f4-f5 Ra8-b8 Qc1-e3 Nd7-f6 Nc4-d2 g7-g6 Rf1-f3 Kg8-h8 f5xg6 f7xg6 h2-h3 Kh8-g7 Ra1-f1 Nf6-d7 Rf3xf8 Nd7xf8 Qe3-e2
37/49+ 01:06 394,337k 5,938k +2.16 Nb1-d2
37/49 01:18 464,346k 5,931k +2.04 a2-a4 Ng8-f6 Bf1-b5+ Nb8-d7 Nb1-c3 Bf8-e7 Ng1-f3 O-O Bb5-d3 Nf6-e8 Nf3-d2 g7-g6 g2-g4 Be7-g5 h2-h3 Kg8-h8 h3-h4 Bg5xh4 Nd2-f3 Bh4-e7 Qd1-d2 Nd7-f6 Qd2-h6 Rf8-g8 Nf3-g5 Rg8-g7 f2-f3 Bc8-d7 Qh6-h2 a7-a6 a4-a5
38/50+ 01:52 647,002k 5,756k +2.13 a2-a4
38/50 02:20 802,614k 5,702k +2.10 a2-a4 Ng8-f6 Qd1-f3 Bf8-e7 Nb1-d2 Bc8-g4 Qf3-e3 Bg4-c8 h2-h3 h7-h6 Ng1-f3 Nf6-h5 Nf3-h2 O-O Qe3-f3 Nh5-f4 g2-g3 Nf4-g6 Nh2-g4 Nb8-a6 Ng4-e3 Be7-g5 Nd2-c4 Ng6-e7 g3-g4 Ne7-g6 h3-h4 Ng6xh4 Qf3-g3 Nh4-g6 Ne3-f5 Bg5-f4 Bc1xf4 e5xf4 Qg3-g2
39/50- 02:24 824,755k 5,707k +1.98 a2-a4 Ng8-f6
39/50+ 02:26 835,706k 5,710k +2.10 a2-a4
39/50 02:27 840,459k 5,712k +2.06 a2-a4 Ng8-f6 Nb1-d2 Bf8-e7 Bf1-b5+ Nb8-d7 Ng1-e2 O-O Ne2-c3 Nf6-e8 h2-h4 a7-a6 Bb5-d3 b7-b6 Nd2-c4 Nd7-f6 Nc4-e3 g7-g6 g2-g4 h7-h5 g4-g5 Nf6-h7 Ne3-f5 g6xf5 e4xf5 Ne8-g7 Qd1-f3 e5-e4 Bd3xe4 Rf8-e8 Bc1-e3 Be7-f8 f5-f6 Bc8-g4 Qf3-f4 Re8xe4 Nc3xe4 Ng7-f5 f2-f3 Nf5xe3 Qf4xe3 Bg4-f5 c2-c4 Qd8-d7 Ke1-f2 Ra8-e8 Qe3-f4 Bf5xe4
40/40+ 02:33 875,129k 5,719k +2.18 a2-a4
40/50- 03:08 1,086,122k 5,763k +1.94 a2-a4 Ng8-f6
40/55 03:24 1,183,918k 5,786k +1.94 a2-a4 Ng8-f6 Nb1-d2 Bf8-e7 Bf1-d3 O-O Ng1-e2 Nf6-e8 O-O Be7-g5 a4-a5 Nb8-a6 f2-f4 e5xf4 Ne2xf4 Qd8-e7 Nd2-c4 Na6-b4 Rf1-e1 Nb4xd3 Nf4xd3 Bg5xc1 Qd1xc1 Bc8-d7 e4-e5 Bd7-b5 Qc1-f4 Bb5xc4 Qf4xc4 Ra8-d8 Ra1-d1 b7-b6 Qc4-a4 Qe7-d7 Qa4xd7 Rd8xd7 c2-c4 b6xa5 b2-b3 h7-h6
41/50+ 04:17 1,472,913k 5,710k +2.08 Ng1-f3
41/57+ 05:28 1,845,909k 5,625k +2.19 a2-a4
41/57 05:31 1,861,627k 5,613k +2.22 a2-a4 Ng8-f6 Nb1-c3 Bf8-e7 Bf1-b5+ Nb8-d7 Ng1-f3 O-O Nf3-d2 Nf6-e8 h2-h4 a7-a6 Bb5-d3 b7-b6 Nd2-c4 Ra8-b8 b2-b3 Nd7-f6 Nc4-e3 Nf6xe4 Nc3xe4 f7-f5 Ne4-g5 e5-e4 Bd3xa6 Bc8xa6 Ng5-e6 Qd8-d7 Ne6xf8 Kg8xf8 g2-g3 Be7-f6 Ra1-a2 Bf6-c3+ Bc1-d2 Bc3-e5 c2-c4 Kf8-g8 Ne3-g2 Ne8-f6 Bd2-f4 h7-h6 Bf4xe5 d6xe5 O-O Rb8-f8 f2-f4 Nf6-g4 f4xe5 Ng4xe5 Ng2-f4
42/57- 07:50 2,608,682k 5,546k +2.06 a2-a4 Bf8-e7
42/57+ 07:55 2,640,302k 5,548k +2.18 a2-a4
42/59 10:27 3,418,958k 5,449k +2.24 Ng1-e2 Bf8-e7 Ne2-c3 Ng8-f6 Bf1-b5+ Nb8-d7 a2-a4 O-O Nb1-d2 Nf6-e8 h2-h4 Ra8-b8 Bb5-d3 Ne8-c7 Nd2-c4 a7-a6 b2-b3 Nd7-f6 Nc4-b6 Bc8-g4 f2-f3 Bg4-d7 Nb6xd7 Nf6xd7 g2-g3 b7-b5 a4xb5 a6xb5 Qd1-e2 b5-b4 Nc3-d1 Rb8-a8 Ra1xa8 Nc7xa8 Nd1-e3
43/58- 10:35 3,453,019k 5,435k +2.04 Ng1-e2 Bf8-e7
43/59+ 10:37 3,461,625k 5,433k +2.17 Ng1-e2
43/59 10:52 3,530,350k 5,407k +2.22 Ng1-e2 Bf8-e7 Ne2-c3 Ng8-f6 Bf1-b5+ Nb8-d7 a2-a4 O-O Nb1-d2 a7-a6 Bb5-d3 Nf6-e8 h2-h4 Ra8-b8 Nd2-c4 b7-b6 b2-b3 Nd7-f6 Nc4-e3 Nf6xe4 Nc3xe4 f7-f5 Ne4-g5 e5-e4 Bd3xa6 Bc8xa6 Ng5-e6 Qd8-d7 Ne6xf8 Kg8xf8 g2-g3 Be7-f6 Ra1-b1 Bf6-c3+ Bc1-d2 Bc3-e5 c2-c4 Kf8-g8 Qd1-c2 Ne8-f6 Bd2-c3 Rb8-f8 Ke1-d1 Nf6-h5 Bc3xe5 d6xe5 Kd1-c1 f5-f4 Ne3-f1 Qd7-g4 Qc2xe4 Nh5-f6
44/51 11:35 3,743,944k 5,384k +2.24 Ng1-e2 Ng8-f6 Ne2-c3 Bf8-e7 Bf1-b5+ Nf6-d7 a2-a4 Nb8-a6 Nb1-d2 O-O O-O Na6-c7 Bb5-d3 b7-b6 Nd2-c4 Bc8-a6 Qd1-g4 Ba6xc4 Bd3xc4 Nd7-f6 Qg4-e2 Nf6-d7 Nc3-d1 Be7-g5 Nd1-e3 Bg5-h6 Bc4-d3 Kg8-h8 Qe2-f3 a7-a6 Qf3-e2 Bh6xe3
45/59 12:40 4,087,543k 5,376k +2.21 Ng1-e2 Ng8-f6 Ne2-c3 Bf8-e7 Bf1-b5+ Nb8-d7 a2-a4 O-O Nb1-d2 a7-a6 Bb5-d3 Nf6-e8 h2-h4 Ra8-b8 Nd2-c4 Ne8-c7 b2-b3 h7-h6 Bc1-d2 b7-b5 Nc4-a5 Qd8-e8 Na5-c6 b5-b4 Nc3-e2 f7-f5 Ne2-g3 Rb8-b7 Qd1-h5 Nd7-f6 Qh5xe8 Rf8xe8 Nc6xe7+ Re8xe7 Ng3xf5 Bc8xf5 e4xf5 Kg8-f8 O-O-O Nf6xd5 g2-g4 Nd5-c3 Bd2xc3 b4xc3 Bd3-c4 Re7-d7
46/52- 12:52 4,145,906k 5,365k +2.08 Ng1-e2 Ng8-f6
46/54+ 13:10 4,228,040k 5,348k +2.21 Ng1-e2
46/66 15:03 4,834,484k 5,349k +2.07 a2-a4 Ng8-f6 Nb1-d2 Bf8-e7 Bf1-b5+ Nb8-d7 Ng1-e2 a7-a6 Bb5-d3 g7-g6 Ne2-c3 Nf6-h5 Bd3-e2 Nh5-g7 a4-a5 f7-f5 O-O O-O Nd2-c4 f5-f4 Nc3-a4 Ra8-b8 Qd1-d3 g6-g5 Bc1-d2 Ng7-e8 Be2-g4 Nd7-f6 Bg4xc8 Qd8xc8 f2-f3 Be7-d8 b2-b4 Nf6-d7 Kg1-h1 Rf8-f7 Ra1-b1 h7-h5 b4-b5 a6xb5 Rb1xb5 Rf7-g7 Rf1-b1 g5-g4 f3xg4 h5xg4 g2-g3
47/62+ 15:05 4,842,164k 5,350k +2.18 a2-a4
47/62 15:26 4,942,391k 5,333k +2.19 a2-a4 Ng8-f6 Nb1-d2 Bf8-e7 Bf1-b5+ Nb8-d7 Ng1-e2 O-O Ne2-c3 Nf6-e8 h2-h4 Ra8-b8 Bb5-d3 a7-a6 Nd2-c4 Ne8-c7 b2-b3 Nd7-f6 Nc4-b6 Bc8-g4 f2-f3 Bg4-d7 Nb6xd7 Nf6xd7 g2-g3 b7-b5 a4xb5 a6xb5 Qd1-e2 c5-c4 b3xc4 b5-b4 Nc3-b5 Nc7xb5 c4xb5 Nd7-c5 Bc1-e3 Qd8-c8 Bd3-c4 Be7-d8 Ke1-f2 Bd8-b6 Ra1-b1 Bb6-a5 Be3-d2 Rb8-a8 Kf2-g2 f7-f5 e4xf5 Qc8xf5 Bd2xb4 Ba5xb4 Rb1xb4 Ra8-a3
48/63 16:11 5,168,608k 5,320k +2.07 Nb1-d2 Ng8-f6 a2-a4 Bf8-e7 Bf1-b5+ Nb8-d7 Ng1-e2 O-O Ne2-c3 Nf6-e8 h2-h4 Ra8-b8 Bb5-d3 Ne8-c7 Nd2-c4 Nd7-f6 b2-b3 a7-a6 Nc4-b6 Bc8-g4 f2-f3 Bg4-d7 Nb6xd7 Nf6xd7 g2-g3 b7-b5 a4xb5 a6xb5 Qd1-e2 c5-c4 b3xc4 b5-b4 Nc3-b5 Nd7-c5 Bc1-e3 Nc7xb5 c4xb5 Qd8-c8 Be3xc5 Qc8xc5 Ra1-a6 Rb8-a8 Qe2-f2 Qc5-c3+ Qf2-d2 Qc3-d4 Ke1-f1 g7-g6 Kf1-g2 Ra8xa6 b5xa6 Rf8-a8
49/54 17:10 5,493,741k 5,332k +2.17 Nb1-d2 Ng8-f6 a2-a4 Bf8-e7 Bf1-b5+ Nb8-d7 Ng1-e2 O-O Ne2-c3 Nf6-e8 h2-h4 Ra8-b8 Bb5-d3 Ne8-c7 Nd2-c4 Nd7-f6 b2-b3 a7-a6 Nc4-b6 Bc8-g4 f2-f3 Bg4-d7 Nb6xd7 Nf6xd7 g2-g3 b7-b5 a4xb5 a6xb5 Qd1-e2 c5-c4 b3xc4 b5-b4 Nc3-b5 Nd7-c5 Bc1-e3 Nc7xb5 c4xb5 Qd8-c8 Be3xc5 Qc8xc5 Ra1-a6 Rb8-a8 Qe2-f2 Qc5-c3+ Qf2-d2 Qc3-d4 Ke1-f1 Ra8xa6 b5xa6 Be7-d8 Qd2-f2 Qd4-b2 Kf1-g2 Bd8-a5
50/55- 21:44 6,883,720k 5,276k +2.01 Nb1-d2 Ng8-f6
50/66+ 23:25 7,406,501k 5,270k +2.13 Nb1-d2
50/66- 38:01 11,889,182k 5,211k +1.88 Nb1-d2 Ng8-f6
50/66+ 38:28 12,037,347k 5,214k +2.09 Nb1-d2
50/67 49:01 15,074,393k 5,125k +2.13 Nb1-d2 Ng8-f6 a2-a4 Bf8-e7 h2-h3 Nb8-a6 Qd1-f3 Na6-c7 g2-g4 h7-h6 Nd2-c4 Bc8-d7 Ng1-e2 Ke8-f8 Ne2-c3 Nf6-e8 Bf1-e2 Kf8-g8 Bc1-e3 g7-g6 Qf3-g2 Be7-g5 Qg2-h2 a7-a6 h3-h4 Bg5xe3 Nc4xe3 Kg8-g7 Rh1-f1 b7-b5 b2-b3 Ne8-f6 f2-f3 Ra8-b8 Ke1-d2 Rh8-f8 Qh2-g3 Qd8-e7 Be2-d3 b5xa4 Nc3xa4 Nc7-b5 f3-f4 e5xf4 Qg3xf4 Rf8-e8 Na4-b2 Nf6xe4+ Bd3xe4 Qe7xe4 Qf4xf7+ Kg7-h8 Qf7-f6+ Kh8-g8
51/65- 54:19 16,641,780k 5,105k +1.98 Nb1-d2 Ng8-f6
51/65+ 1:04:37 19,802,333k 5,107k +2.10 Nb1-d2
51/72 1:13:41 22,485,493k 5,085k +2.06 Nb1-d2 Ng8-f6 a2-a4 Bf8-e7 h2-h3 Nb8-a6 Bf1-d3 O-O Ng1-e2 b7-b6 O-O g7-g6 c2-c3 Na6-c7 f2-f4 e5xf4 Nd2-f3 Nf6-d7 Bc1xf4 f7-f6 Qd1-d2 Nd7-e5 Rf1-f2 Bc8-d7 b2-b3 a7-a6 Ra1-f1 Ne5-f7 Bf4-e3 Nc7-e8 c3-c4 Nf7-e5 Ne2-c3 Rf8-f7 Be3-f4 Rf7-g7 Bf4-h6 Rg7-f7 g2-g4 Ra8-a7 Qd2-e3 Be7-f8 Bh6xf8 Rf7xf8 Nf3xe5 d6xe5 Qe3-h6
52/63- 1:21:05 24,659,264k 5,068k +1.94 Nb1-d2 Ng8-f6
52/69+ 1:21:23 24,758,124k 5,069k +2.06 Nb1-d2
52/69 1:38:35 30,379,544k 5,136k +2.05 Nb1-d2 Ng8-f6 a2-a4 Bf8-e7 Qd1-f3 h7-h5 h2-h3 h5-h4 Bf1-d3 Nf6-h7 Nd2-c4 Nb8-d7 Qf3-d1 Nd7-b6 Nc4-e3 a7-a5 b2-b3 Be7-g5 Bd3-b5+ Ke8-f8 Ng1-e2 Nh7-f6 Ne2-c3 Nf6-h5 Bc1-d2 g7-g6 Bb5-e2 Nh5-f6 Be2-d3 Nf6-h5 Qd1-e2 Nh5-f4 Qe2-f1 Nf4xd3+ Qf1xd3 Bc8-d7 O-O Bg5-f4 Nc3-b5 Qd8-g5 Kg1-h1 Kf8-e7 Nb5-c7 Ra8-c8 Bd2xa5 Bf4xe3 f2xe3 Rc8xc7 Ba5xb6 Rc7-c8 a4-a5 f7-f5 a5-a6
53/69 1:51:19 34,135,945k 5,110k +2.05 a2-a4 Ng8-f6 Qd1-f3 Bf8-e7 h2-h3 h7-h5 Bc1-e3 b7-b6 Nb1-d2 Nb8-d7 Bf1-d3 Nd7-f8 Bd3-b5+ Bc8-d7 Bb5xd7+ Nf8xd7 Qf3-e2 g7-g6 Ng1-f3 a7-a6 O-O O-O Be3-h6 Rf8-e8 Ra1-a3 Qd8-c7 Rf1-a1 Be7-f8 Bh6-g5 Nf6-h7 Bg5-e3 Nd7-f6 Qe2-d3 Nf6-d7 Ra3-b3 Re8-b8 g2-g3 Bf8-e7 Kg1-g2 Ra8-a7 Nf3-e1 h5-h4 g3-g4 Nd7-f8 Ne1-f3 g6-g5 Rb3-a3 f7-f6 Ra3-b3 Nf8-g6 Nd2-c4
54/57 1:57:10 35,804,751k 5,093k +2.08 a2-a4 Ng8-f6 Qd1-f3 Bf8-e7 h2-h3 h7-h5 Bc1-e3 b7-b6 Nb1-d2 Nb8-d7 Bf1-d3 Nd7-f8 Bd3-b5+ Bc8-d7 Bb5xd7+ Nf8xd7 Qf3-e2 g7-g6 Ng1-f3 a7-a6 O-O O-O Be3-h6 Rf8-e8 Ra1-a3 Qd8-c7 Rf1-a1 Be7-f8 Bh6-g5 Nf6-h7 Bg5-e3 Nd7-f6 Qe2-d3 Nf6-d7 Ra3-b3 Re8-b8 g2-g3 Bf8-e7 Kg1-g2 Ra8-a7 Nf3-e1 Rb8-f8 f2-f4 f7-f5 Ne1-f3 g6-g5 e4xf5 g5-g4 Nf3-g5
55/76- 2:10:26 39,782,629k 5,083k +1.94 a2-a4 Ng8-f6
55/76+ 2:11:47 40,142,607k 5,076k +2.06 a2-a4
55/76+ 2:42:13 49,043,730k 5,039k +2.22 a2-a4
55/76 2:54:05 52,546,784k 5,031k +2.10 a2-a4 Bf8-e7 Qd1-f3 Nb8-a6 Nb1-d2 Na6-b4 Qf3-d1 Bc8-d7 Nd2-c4 f7-f5 e4xf5 Bd7xf5 Nc4-e3 Bf5-d7 Ng1-e2 Nb4-a6 Ne2-g3 Ng8-h6 Bf1-d3 g7-g6 c2-c3 Nh6-f7 Qd1-c2 Be7-g5 Ng3-e4 O-O a4-a5 Ra8-c8 b2-b3 Na6-c7 Ne4xg5 Nf7xg5 O-O Ng5-f7 f2-f3 Nc7-e8 Ne3-d1 Ne8-f6 c3-c4 Nf6-h5 Nd1-f2 Nh5-f4 Kg1-h1 Rc8-b8 Qc2-d2 Nf4xd3 Qd2xd3
56/77- 3:30:36 63,616,146k 5,034k +2.00 a2-a4 Bf8-e7
56/77+ 3:32:07 64,061,610k 5,033k +2.12 a2-a4
56/77 3:42:07 66,952,069k 5,023k +2.12 a2-a4 Ng8-f6 Qd1-f3 Bf8-e7 h2-h3 h7-h5 Nb1-a3 Nb8-d7 Bf1-d3 Nd7-f8 Bd3-b5+ Bc8-d7 Ng1-e2 Bd7xb5 Na3xb5 a7-a6 Nb5-a3 Nf8-g6 O-O h5-h4 Na3-c4 b7-b5 Nc4-e3 b5-b4 Ne3-c4 O-O Bc1-e3 Nf6-h7 Qf3-g4 Qd8-c7 b2-b3 Nh7-f6 Qg4-f3 Ra8-e8 Ra1-e1 Nf6-h7 Ne2-c1 Be7-g5 Nc1-d3 Bg5xe3 f2xe3 Qc7-d8 Re1-d1 Ng6-e7 Qf3-g4 Ne7-c8 Rf1-f5 g7-g6 Rd1-f1 Kg8-g7 Rf5-f2 Qd8-g5
57/68 3:46:07 68,177,506k 5,025k +2.14 a2-a4 Ng8-f6 Qd1-f3 Bf8-e7 Nb1-d2 h7-h5 Bf1-d3 h5-h4 h2-h3 Nf6-h7 Nd2-c4 Nb8-d7 Qf3-d1 Nd7-b6 Nc4-e3 c5-c4 Ne3xc4 Nb6xc4 Bd3xc4 Be7-g5 Ng1-f3 Bg5xc1 Bc4-b5+ Bc8-d7 Qd1xc1 Nh7-f6 Qc1-e3 Qd8-a5+ Qe3-d2 Qa5xd2+ Nf3xd2 Ra8-c8 c2-c3 Nf6-h5 O-O Ke8-e7 Bb5xd7 Ke7xd7 Ra1-a3 Nh5-f4 a4-a5 g7-g6 Rf1-e1 Rc8-c7 c3-c4 Rh8-b8 b2-b4 f7-f5 Kg1-h2 Rb8-c8 e4xf5 g6xf5 g2-g3 h4xg3+ f2xg3 Nf4-h5 Ra3-f3
58/57 3:48:39 68,883,844k 5,021k +2.12 a2-a4 Ng8-f6 Qd1-f3 Bf8-e7 Nb1-d2 h7-h5 Bf1-d3 h5-h4 h2-h3 Nf6-h7 Nd2-c4 Nb8-d7 Qf3-d1 Nd7-b6 Nc4-e3 c5-c4 Ne3xc4 Nb6xc4 Bd3xc4 Be7-g5 Ng1-f3 Bg5xc1 Bc4-b5+ Bc8-d7 Qd1xc1 Nh7-f6 Qc1-e3 Qd8-a5+ Qe3-d2 Qa5xd2+ Nf3xd2 Ra8-c8 c2-c3 Nf6-h5 O-O Ke8-e7 Ra1-a3 Rc8-c7 Bb5xd7 Ke7xd7 a4-a5 Nh5-f4 Kg1-h2 g7-g6 c3-c4 f7-f5 Rf1-e1 Nf4-h5 Ra3-b3 Rh8-b8 g2-g4 h4xg3/ep+ f2xg3
59/70- 3:59:20 72,038,265k 5,017k +2.00 a2-a4 Ng8-f6
59/76+ 4:35:35 82,735,742k 5,004k +2.12 a2-a4
59/76 5:11:11 93,338,519k 4,999k +2.06 a2-a4 Ng8-f6 Qd1-f3 Bf8-e7 h2-h3 O-O Nb1-d2 Nb8-a6 g2-g4 Na6-c7 Nd2-c4 Ra8-b8 Nc4-e3 Bc8-d7 Bf1-d3 b7-b5 a4xb5 Nc7xb5 c2-c3 a7-a5 Ng1-e2 a5-a4 Bd3-c2 Rb8-a8 Ke1-f1 Nb5-c7 Ne2-g3 Nf6-e8 Qf3-d1 g7-g6 Bc2xa4 Be7-h4 Kf1-g2 Ne8-f6 Qd1-c2 Ra8-a6 Rh1-f1 Nc7-e8 Bc1-d2 Ra6xa4 Ra1xa4 Bd7xa4 Qc2xa4 h7-h5 g4xh5 Ne8-g7 h5xg6 f7xg6 Rf1-g1 Nf6-h5
60/57- 5:58:10 107,317,493k 4,994k +1.95 a2-a4 Ng8-f6
60/57+ 6:01:57 108,586,396k 5,000k +2.07 a2-a4
60/80 6:32:32 118,056,984k 5,012k +2.06 a2-a4 Ng8-f6 Qd1-f3 Bf8-e7 h2-h3 O-O g2-g4 Nb8-a6 Nb1-d2 Na6-c7 Nd2-c4 Bc8-d7 Bc1-d2 Bd7xa4 Ra1xa4 b7-b5 Ra4-a2 b5xc4 Bf1xc4 Nf6-d7 h3-h4 a7-a5 Ke1-f1 a5-a4 Bc4-e2 Qd8-c8 Qf3-d3 Qc8-b7 Kf1-g2 Nc7-a6 Ra2xa4 c5-c4 Ra4xc4 Qb7xb2 Qd3-b3 Ra8-b8 Qb3xb2 Rb8xb2 Bd2-c1 Rb2-a2 Bc1-e3 Na6-c5 g4-g5 Rf8-b8 Ng1-f3 Rb8-b2 Be2-d3 f7-f6 g5xf6 Be7xf6 h4-h5 h7-h6 Rc4-c3 Kg8-f8 Kg2-g3 Bf6-d8 Nf3-h4 Bd8xh4+ Rh1xh4
61/60 6:52:16 124,068,139k 5,016k +2.03 a2-a4 Ng8-f6 Qd1-f3 Nb8-a6 Bf1-b5+ Nf6-d7 Ng1-e2 Bf8-e7 h2-h3 Na6-c7 Bb5-d3 a7-a6 Nb1-d2 b7-b6 b2-b3 Ra8-b8 Nd2-c4 Nd7-f6 O-O O-O Ne2-g3 Nf6-e8 a4-a5 b6-b5 Nc4-b6 Be7-g5 c2-c4 Bg5xc1 Ra1xc1 g7-g6 Qf3-e3 Nc7-a8 Nb6xc8 Rb8xc8 f2-f4 Qd8xa5 f4xe5 d6xe5 Qe3-h6 Qa5-c7 c4xb5 a6xb5 Bd3xb5 Ne8-d6 Bb5-c6 f7-f6 h3-h4 Na8-b6 h4-h5 Qc7-g7 Qh6xg7+ Kg8xg7 Rc1xc5 Nb6-d7 Rc5-a5
62/85+ 9:21:10 172,918,934k 5,136k +2.16 Ng1-e2
62/85+ 10:44:05 200,865,396k 5,198k +2.28 Ng1-e2
62/85- 11:02:15 207,038,063k 5,210k +1.92 Ng1-e2 Ng8-f6
62/88 13:37:51 257,093,318k 5,239k +1.98 Ng1-e2 Ng8-f6 Ne2-c3 Bf8-e7 Bf1-b5+ Nb8-d7 a2-a4 O-O Qd1-e2 Nf6-e8 Bb5-d3 Ra8-b8 b2-b3 Nd7-b6 a4-a5 Nb6-d7 f2-f3 Ne8-c7 Bc1-d2 Nd7-f6 Nb1-a3 Bc8-d7 O-O Nf6-h5 g2-g3 g7-g6 Kg1-h1 Kg8-h8 Nc3-b5 Nc7xb5 Na3xb5 a7-a6 Nb5-a3 Be7-g5 Bd2xg5 Qd8xg5 Na3-c4 Qg5-e7 f3-f4 e5xf4 g3xf4 Bd7-h3 Rf1-f2 Rb8-d8 Qe2-e3 Qe7-h4 Ra1-g1 f7-f5 e4-e5 d6xe5 Nc4xe5 Nh5-f6 Qe3xc5 Nf6-e4 Bd3xe4 f5xe4
63/74+ 13:42:20 258,622,028k 5,242k +2.10 Ng1-e2
63/74+ 13:43:07 258,893,152k 5,242k +2.22 Ng1-e2
63/82 13:44:31 259,372,117k 5,243k +2.10 Ng1-e2 Ng8-f6 Ne2-c3 Bf8-e7 Bf1-b5+ Nb8-d7 a2-a4 O-O Bb5-d3 Nf6-e8 Nb1-a3 Be7-g5 Bc1xg5 Qd8xg5 Qd1-f3 a7-a6 h2-h4 Qg5-e7 g2-g4 Ne8-c7 Na3-c4 b7-b5 a4xb5 a6xb5 Ra1xa8 b5xc4 Ra8xc8 Rf8xc8 Bd3xc4 Nc7-a8 Ke1-e2 Nd7-f8 h4-h5 h7-h6 Rh1-a1 Na8-b6 Bc4-b5 Nf8-h7 Bb5-c6 Nh7-g5 Qf3-g3 Rc8-b8 b2-b3 Qe7-d8 Qg3-e3 Qd8-c8 f2-f3 Ng5-h3 Nc3-b5 Qc8-d8 Ke2-d2 Nb6-c8 Kd2-c1 Nh3-f4 Nb5-a3 Nc8-b6 Qe3-d2 Qd8-h4 Kc1-b2 c5-c4 b3-b4
64/94+ 13:51:04 261,644,469k 5,247k +2.24 Ng1-e2
Marcus91
Posts: 58
Joined: Sun Jul 10, 2022 11:00 am
Full name: Marco Giorgio

Re: Testing depend largely on good Openings

Post by Marcus91 »

lkaufman wrote: Fri Jul 08, 2022 5:49 am This is precisely why so many are switching to unbalanced openings (with color reversal). If you look at the current CCRL Complete Rapid list, the top five engines (four stockfish versions and latest Komodo Dragon) are all within four elo points (!), while the next six (five Stockfish versions or derivatives plus previous Komodo Dragon) are within another dozen Elo of the top five, using reasonably balanced openings. The best engines just don't lose games against each other anymore with good openings (with 4 cpu and Rapid tc). With unbalanced openings, the average result of a two game match between equal opponents is likely to be one win and one draw, or at least close to that. If one engine is just slightly stronger, it will tend to be the one that scores the 1.5. So instead of a four elo gap, we might see a forty elo gap or more when unbalanced openings are used. With good openings, I think it is unlikely that any engine will ever reach 3600 (top now is 3540) on that CCRL Rapid list; at about 2 elo per year, it would take 30 years, but progress will surely get less each year. With unbalanced openings, it is true that there is a practical upper limit of 192 elo on the result between two super engines, assuming that they are strong enough never to lose from the nearly winning side. But as long as engine pairings are limited to maybe 150 elo apart or less, there is no obvious limit to how high ratings might go with unbalanced openings, 5000, 6000, who knows? You may argue that this is artificial, that only "balanced" (i.e. typical GM) openings should be used or else it's not real chess, and that is a valid argument, but unless we go with unbalanced openings computer chess is dying; there isn't enough room for further gains from the normal starting position, the path to a draw is too clear now. We just have to learn to think in terms of pairs of games; as long as each side gets White once from the same opening, it is fair, it's just not how chess is played between humans (well, so far anyway).
I agree on using unbalanced openings to highlight small differences between top engines, but there is no real evidence that the current best engines are 'unbeatable' with balanced openings, here are a couple of examples I noticed in a game analyzed with Dragon 2.6:

Here dragon prefers ..Qd8 (??) without seeing f5! as a continuation, at least in the first minute of analysis
[fen]rnb2rk1/1pqp1p1p/p3p1pQ/4P3/5P2/BN1B4/P1P3PP/b4RK1 b - - 1 2[/fen]


a few moves later, once again the winning f5!! it is not seen by Dragon 2.6
[fen]rn1q1rk1/1b3p1p/p3p1pQ/1p1pP1P1/5P2/BN1B4/P1P4P/b4R1K w - - 2 18[/fen]


What I want to show is that there could be many winning moves that are not seen by the current top engines
Uri Blass
Posts: 11150
Joined: Thu Mar 09, 2006 12:37 am
Location: Tel-Aviv Israel

Re: Testing depend largely on good Openings

Post by Uri Blass »

Marcus91 wrote: Sun Jul 10, 2022 1:50 pm
lkaufman wrote: Fri Jul 08, 2022 5:49 am This is precisely why so many are switching to unbalanced openings (with color reversal). If you look at the current CCRL Complete Rapid list, the top five engines (four stockfish versions and latest Komodo Dragon) are all within four elo points (!), while the next six (five Stockfish versions or derivatives plus previous Komodo Dragon) are within another dozen Elo of the top five, using reasonably balanced openings. The best engines just don't lose games against each other anymore with good openings (with 4 cpu and Rapid tc). With unbalanced openings, the average result of a two game match between equal opponents is likely to be one win and one draw, or at least close to that. If one engine is just slightly stronger, it will tend to be the one that scores the 1.5. So instead of a four elo gap, we might see a forty elo gap or more when unbalanced openings are used. With good openings, I think it is unlikely that any engine will ever reach 3600 (top now is 3540) on that CCRL Rapid list; at about 2 elo per year, it would take 30 years, but progress will surely get less each year. With unbalanced openings, it is true that there is a practical upper limit of 192 elo on the result between two super engines, assuming that they are strong enough never to lose from the nearly winning side. But as long as engine pairings are limited to maybe 150 elo apart or less, there is no obvious limit to how high ratings might go with unbalanced openings, 5000, 6000, who knows? You may argue that this is artificial, that only "balanced" (i.e. typical GM) openings should be used or else it's not real chess, and that is a valid argument, but unless we go with unbalanced openings computer chess is dying; there isn't enough room for further gains from the normal starting position, the path to a draw is too clear now. We just have to learn to think in terms of pairs of games; as long as each side gets White once from the same opening, it is fair, it's just not how chess is played between humans (well, so far anyway).
I agree on using unbalanced openings to highlight small differences between top engines, but there is no real evidence that the current best engines are 'unbeatable' with balanced openings, here are a couple of examples I noticed in a game analyzed with Dragon 2.6:

Here dragon prefers ..Qd8 (??) without seeing f5! as a continuation, at least in the first minute of analysis
[fen]rnb2rk1/1pqp1p1p/p3p1pQ/4P3/5P2/BN1B4/P1P3PP/b4RK1 b - - 1 2[/fen]


a few moves later, once again the winning f5!! it is not seen by Dragon 2.6
[fen]rn1q1rk1/1b3p1p/p3p1pQ/1p1pP1P1/5P2/BN1B4/P1P4P/b4R1K w - - 2 18[/fen]


What I want to show is that there could be many winning moves that are not seen by the current top engines
1)Dragon2.6 is not a current top engine so it needs time to find f5 and stockfish can easily find f5 faster.

2)It seems that at long time control and not at blitz time control Dragon2.6 almost never lose from balanced openings.

https://ccrl.chessdom.com/ccrl/4040/cgi ... 4-bit_4CPU
Marcus91
Posts: 58
Joined: Sun Jul 10, 2022 11:00 am
Full name: Marco Giorgio

Re: Testing depend largely on good Openings

Post by Marcus91 »

Uri Blass wrote: Sun Jul 10, 2022 2:32 pm
Marcus91 wrote: Sun Jul 10, 2022 1:50 pm
lkaufman wrote: Fri Jul 08, 2022 5:49 am This is precisely why so many are switching to unbalanced openings (with color reversal). If you look at the current CCRL Complete Rapid list, the top five engines (four stockfish versions and latest Komodo Dragon) are all within four elo points (!), while the next six (five Stockfish versions or derivatives plus previous Komodo Dragon) are within another dozen Elo of the top five, using reasonably balanced openings. The best engines just don't lose games against each other anymore with good openings (with 4 cpu and Rapid tc). With unbalanced openings, the average result of a two game match between equal opponents is likely to be one win and one draw, or at least close to that. If one engine is just slightly stronger, it will tend to be the one that scores the 1.5. So instead of a four elo gap, we might see a forty elo gap or more when unbalanced openings are used. With good openings, I think it is unlikely that any engine will ever reach 3600 (top now is 3540) on that CCRL Rapid list; at about 2 elo per year, it would take 30 years, but progress will surely get less each year. With unbalanced openings, it is true that there is a practical upper limit of 192 elo on the result between two super engines, assuming that they are strong enough never to lose from the nearly winning side. But as long as engine pairings are limited to maybe 150 elo apart or less, there is no obvious limit to how high ratings might go with unbalanced openings, 5000, 6000, who knows? You may argue that this is artificial, that only "balanced" (i.e. typical GM) openings should be used or else it's not real chess, and that is a valid argument, but unless we go with unbalanced openings computer chess is dying; there isn't enough room for further gains from the normal starting position, the path to a draw is too clear now. We just have to learn to think in terms of pairs of games; as long as each side gets White once from the same opening, it is fair, it's just not how chess is played between humans (well, so far anyway).
I agree on using unbalanced openings to highlight small differences between top engines, but there is no real evidence that the current best engines are 'unbeatable' with balanced openings, here are a couple of examples I noticed in a game analyzed with Dragon 2.6:

Here dragon prefers ..Qd8 (??) without seeing f5! as a continuation, at least in the first minute of analysis
[fen]rnb2rk1/1pqp1p1p/p3p1pQ/4P3/5P2/BN1B4/P1P3PP/b4RK1 b - - 1 2[/fen]


a few moves later, once again the winning f5!! it is not seen by Dragon 2.6
[fen]rn1q1rk1/1b3p1p/p3p1pQ/1p1pP1P1/5P2/BN1B4/P1P4P/b4R1K w - - 2 18[/fen]


What I want to show is that there could be many winning moves that are not seen by the current top engines
2)It seems that at long time control and not at blitz time control Dragon2.6 almost never lose from balanced openings.
For now
Chessqueen
Posts: 5685
Joined: Wed Sep 05, 2018 2:16 am
Location: Moving
Full name: Jorge Picado

Re: Testing depend largely on good Openings

Post by Chessqueen »

Marcus91 wrote: Sun Jul 10, 2022 3:57 pm
Uri Blass wrote: Sun Jul 10, 2022 2:32 pm
Marcus91 wrote: Sun Jul 10, 2022 1:50 pm
lkaufman wrote: Fri Jul 08, 2022 5:49 am This is precisely why so many are switching to unbalanced openings (with color reversal). If you look at the current CCRL Complete Rapid list, the top five engines (four stockfish versions and latest Komodo Dragon) are all within four elo points (!), while the next six (five Stockfish versions or derivatives plus previous Komodo Dragon) are within another dozen Elo of the top five, using reasonably balanced openings. The best engines just don't lose games against each other anymore with good openings (with 4 cpu and Rapid tc). With unbalanced openings, the average result of a two game match between equal opponents is likely to be one win and one draw, or at least close to that. If one engine is just slightly stronger, it will tend to be the one that scores the 1.5. So instead of a four elo gap, we might see a forty elo gap or more when unbalanced openings are used. With good openings, I think it is unlikely that any engine will ever reach 3600 (top now is 3540) on that CCRL Rapid list; at about 2 elo per year, it would take 30 years, but progress will surely get less each year. With unbalanced openings, it is true that there is a practical upper limit of 192 elo on the result between two super engines, assuming that they are strong enough never to lose from the nearly winning side. But as long as engine pairings are limited to maybe 150 elo apart or less, there is no obvious limit to how high ratings might go with unbalanced openings, 5000, 6000, who knows? You may argue that this is artificial, that only "balanced" (i.e. typical GM) openings should be used or else it's not real chess, and that is a valid argument, but unless we go with unbalanced openings computer chess is dying; there isn't enough room for further gains from the normal starting position, the path to a draw is too clear now. We just have to learn to think in terms of pairs of games; as long as each side gets White once from the same opening, it is fair, it's just not how chess is played between humans (well, so far anyway).
I agree on using unbalanced openings to highlight small differences between top engines, but there is no real evidence that the current best engines are 'unbeatable' with balanced openings, here are a couple of examples I noticed in a game analyzed with Dragon 2.6:

Here dragon prefers ..Qd8 (??) without seeing f5! as a continuation, at least in the first minute of analysis
[fen]rnb2rk1/1pqp1p1p/p3p1pQ/4P3/5P2/BN1B4/P1P3PP/b4RK1 b - - 1 2[/fen]


a few moves later, once again the winning f5!! it is not seen by Dragon 2.6
[fen]rn1q1rk1/1b3p1p/p3p1pQ/1p1pP1P1/5P2/BN1B4/P1P4P/b4R1K w - - 2 18[/fen]

What I want to show is that there could be many winning moves that are not seen by the current top engines
2)It seems that at long time control and not at blitz time control Dragon2.6 almost never lose from balanced openings.
But in that position....Qd8 is the only best move since .... f5 is a Blunder since it allows mate in one move Qxf8+, you probably meant ...d6, but still the White Bishop will take on d6 anyway

FEN: rn1q1rk1/1b3p1p/p3p1pQ/1p1pP1P1/5P2/BN1B4/P1P4P/b4R1K w - - 2 18

Dragon-2.6.1-64bit-avx2:
1 00:00 132 132k -1.22 Ba3xf8
2 00:00 203k 5,205k -1.54 Ba3xf8 Qd8xf8 Qh6xf8+ Kg8xf8 Rf1xa1 Nb8-d7 Kh1-g1
2 00:00 206k 5,174k -1.40 Nb3-a5 Ra8-a7 Qh6xf8+ Qd8xf8 Ba3xf8 Kg8xf8 Rf1xa1 Nb8-d7
3 00:00 207k 5,168k -1.40 Nb3-a5 Ra8-a7 Qh6xf8+ Qd8xf8 Ba3xf8 Kg8xf8
4 00:00 208k 5,164k -1.40 Nb3-a5 Ra8-a7 Qh6xf8+ Qd8xf8 Ba3xf8 Kg8xf8 Rf1xa1 Nb8-d7
5 00:00 209k 5,164k -1.40 Nb3-a5 Ra8-a7 Qh6xf8+ Qd8xf8 Ba3xf8
6 00:00 210k 5,161k -1.40 Nb3-a5 Ra8-a7 Qh6xf8+ Qd8xf8 Ba3xf8 Kg8xf8 Rf1xa1
7- 00:00 211k 5,161k -1.48 Nb3-a5 Bb7-c6
7 00:00 212k 5,160k -1.50 Nb3-a5 Bb7-c6 Ba3xf8 Qd8xf8 Qh6xf8+ Kg8xf8 Rf1xa1
8 00:00 214k 5,146k -1.46 Nb3-a5 Ra8-a7 Qh6xf8+ Qd8xf8 Ba3xf8 Kg8xf8 Rf1xa1 Nb8-d7 Kh1-g1
9- 00:00 218k 5,133k -1.54 Nb3-a5 Ra8-a7
9- 00:00 220k 5,124k -1.71 Nb3-a5 Bb7-c6
9- 00:00 223k 5,120k -2.03 Nb3-a5 Bb7-c6
9+ 00:00 225k 5,116k -1.82 Nb3-a5 Qd8-c8
9 00:00 226k 5,110k -1.69 Nb3-a5 Bb7-c6 Na5xc6 Nb8xc6 Rf1xa1 Rf8-e8 Ba3-c5 Ra8-c8
10- 00:00 229k 5,095k -1.78 Nb3-a5 Bb7-c6
10- 00:00 232k 5,095k -1.87 Nb3-a5 Bb7-c6
10- 00:00 248k 5,064k -2.15 Nb3-a5 Bb7-c6
10 00:00 259k 5,078k -2.12 Rf1xa1 Rf8-e8 Ba3-d6 Nb8-d7 Nb3-d4 Qd8-b6 c2-c3 Ra8-c8 Bd6-b4 Rc8-c4 Bd3xc4 d5xc4+ Kh1-g1
11- 00:00 272k 5,069k -2.19 Rf1xa1 Rf8-e8
11+ 00:00 309k 4,987k -2.17 Rf1xa1 Rf8-e8 Kh1-g1
11+ 00:00 314k 4,963k -2.04 Rf1xa1 Rf8-e8 Kh1-g1
11 00:00 317k 4,950k -1.99 Rf1xa1 Rf8-e8 Kh1-g1 Nb8-d7 Ra1-f1 b5-b4 Ba3-b2 Qd8-e7 h2-h4 Ra8-a7
12- 00:00 336k 4,954k -2.06 Rf1xa1 Rf8-e8
12- 00:00 429k 4,752k -2.15 Rf1xa1 Rf8-e8
12- 00:00 457k 4,720k -2.27 Rf1xa1 Rf8-e8
12+ 00:00 486k 4,727k -2.22 Rf1xa1 Rf8-e8 Kh1-g1
12 00:00 490k 4,732k -2.27 Rf1xa1 Rf8-e8 Kh1-g1 Nb8-d7 Ra1-f1 Qd8-b6+ Rf1-f2 d5-d4 Nb3-d2 b5-b4 Ba3-b2 Qb6-c5 Bd3-e4
13+ 00:00 507k 4,687k -2.20 Rf1xa1 Rf8-e8 Kh1-g1
13+ 00:00 509k 4,693k -2.11 Rf1xa1 Rf8-e8 Kh1-g1
13- 00:00 512k 4,696k -2.19 Rf1xa1 Rf8-e8
13 00:00 522k 4,698k -2.19 Rf1xa1 Rf8-e8 Kh1-g1 Nb8-d7 Ra1-f1 b5-b4 Ba3-b2 d5-d4 Rf1-f2 Ra8-c8 h2-h4 Nd7-c5 Nb3xc5 Rc8xc5 h4-h5
14- 00:00 678k 4,496k -2.26 Rf1xa1 Rf8-e8
14- 00:00 769k 4,461k -2.35 Rf1xa1 Rf8-e8
14+ 00:00 996k 4,373k -2.30 Rf1xa1 Rf8-e8 Nb3-c5
14- 00:00 1,019k 4,363k -2.39 Rf1xa1 Rf8-e8
14+ 00:00 1,201k 4,306k -2.14 Nb3-c5 Ra8-a7 Rf1xa1
14- 00:00 1,203k 4,308k -2.30 Nb3-c5 Ra8-a7
14+ 00:00 1,405k 4,303k -2.23 Nb3-c5 Ra8-a7 Rf1xa1
14 00:00 1,444k 4,288k -2.27 Nb3-c5 Ra8-a7 Rf1xa1 Rf8-e8 h2-h4 Qd8-b6 Ra1-b1 b5-b4 Rb1xb4 Qb6xc5 Rb4xb7 Qc5xa3 Rb7xa7 Qa3-c1+ Kh1-g2
15+ 00:00 1,504k 4,250k -2.20 Nb3-c5 Ra8-a7 Rf1xa1
15+ 00:00 1,520k 4,243k -2.11 Nb3-c5 Ra8-a7 Rf1xa1
15+ 00:00 1,538k 4,241k -1.99 Nb3-c5 Ra8-a7 Rf1xa1
15- 00:00 1,543k 4,242k -2.11 Nb3-c5 Ra8-a7
15+ 00:00 1,679k 4,180k -2.06 Nb3-c5 Ra8-a7 Rf1xa1
15- 00:00 1,706k 4,175k -2.19 Nb3-c5 Ra8-a7
15+ 00:00 1,733k 4,177k -2.09 Nb3-c5 Ra8-a7 Rf1xa1
15 00:00 1,789k 4,169k -2.13 Nb3-c5 Ra8-a7 Rf1xa1 Rf8-e8 h2-h4 Bb7-c6 Ba3-b2 d5-d4+ Kh1-h2 Nb8-d7 Nc5-b3 Nd7xe5 f4xe5 Qd8-d5 Ra1-g1 Qd5xe5+ Kh2-h3
16- 00:00 2,200k 4,132k -2.20 Nb3-c5 Ra8-a7
16+ 00:00 2,692k 4,062k -2.06 Rf1xa1 Rf8-e8 Nb3-d4
16- 00:00 2,708k 4,062k -2.14 Rf1xa1 Rf8-e8
16- 00:00 2,944k 4,052k -2.26 Rf1xa1 Rf8-e8
16+ 00:00 3,455k 4,082k -2.20 Rf1xa1 Rf8-e8 Nb3-d4
16+ 00:00 3,497k 4,087k -1.98 Rf1xa1 Rf8-e8 Nb3-d4
16 00:00 3,514k 4,087k -1.95 Rf1xa1 Rf8-e8 Nb3-d4 Qd8-c7 Ba3-b2 Nb8-c6 Ra1-f1 Qc7-e7 h2-h4 Qe7-b4 Nd4-b3 d5-d4 Kh1-h2 Nc6-e7 Bb2xd4 Re8-c8 h4-h5
17- 00:00 3,868k 4,091k -2.02 Rf1xa1 Rf8-e8
17- 00:00 4,027k 4,102k -2.11 Rf1xa1 Nb8-d7
17+ 00:01 4,119k 4,101k -2.08 Rf1xa1 Nb8-d7 Nb3-d4
17 00:01 4,170k 4,103k -2.06 Rf1xa1 Nb8-d7 Nb3-d4 Rf8-e8 Ra1-f1 Qd8-b6 Ba3-b2 Qb6-a5 h2-h4 Qa5-b4 Nd4-b3 d5-d4+ Kh1-g1 Qb4-f8 h4-h5 Qf8xh6 g5xh6 Ra8-c8 Kg1-h2
18- 00:01 4,451k 4,121k -2.13 Rf1xa1 Nb8-d7
18+ 00:01 4,792k 4,134k -2.11 Rf1xa1 Nb8-d7 Nb3-d4
18 00:01 5,066k 4,134k -2.04 Rf1xa1 Nb8-d7 Nb3-d4 Rf8-e8 Ra1-f1 Qd8-b6 Ba3-b2 Qb6-c5 a2-a3 Qc5-f8 h2-h4 Nd7-c5 h4-h5 Nc5-e4 Rf1-g1 Re8-c8 Kh1-h2 Rc8-c7 Rg1-g4 Ra8-c8 Qh6xf8+ Kg8xf8 Rg4-h4 Ne4-c5 h5xg6 f7xg6
19+ 00:01 5,865k 4,174k -1.97 Rf1xa1 Nb8-d7 Nb3-d4
19+ 00:01 6,470k 4,245k -1.88 Rf1xa1 Nb8-d7 Nb3-d4
19- 00:01 6,712k 4,269k -1.96 Rf1xa1 Rf8-e8
19+ 00:01 7,399k 4,329k -1.92 Rf1xa1 Rf8-e8 Nb3-d4
19- 00:01 8,284k 4,370k -2.02 Rf1xa1 d5-d4+
19- 00:02 10,107k 4,425k -2.32 Rf1xa1 d5-d4+
19+ 00:02 10,690k 4,431k -1.95 Nb3-c5 Bb7-c6 Bd3xg6
19 00:02 10,987k 4,424k -1.78 Nb3-c5 Ra8-a7 Rf1-f3 Bb7-c8 Nc5-b3 f7-f5 g5xf6/ep Rf8xf6 Nb3xa1 Rf6-f7 Na1-b3 d5-d4 Rf3-g3 Nb8-c6 Ba3-c5 Ra7-c7 Bc5-d6 Nc6-e7 Bd6xc7 Qd8xc7 Nb3xd4
20+ 00:02 12,224k 4,425k -1.71 Nb3-c5 Ra8-a7 Bd3xg6
20+ 00:02 12,379k 4,424k -1.62 Nb3-c5 Bb7-c6 Bd3xg6
20+ 00:02 12,501k 4,419k -1.50 Nb3-c5 Ra8-a7 Bd3xg6
20- 00:02 12,829k 4,420k -1.60 Nb3-c5 Bb7-c8
20+ 00:02 13,221k 4,414k -1.55 Nb3-c5 Bb7-c8 Rf1xa1
20+ 00:03 15,049k 4,410k -1.25 Nb3-c5 Bb7-c8 Nc5-b3
20- 00:03 15,360k 4,409k -1.46 Nb3-c5 Bb7-c8
20+ 00:03 17,374k 4,398k -1.32 Nb3-c5 Bb7-c8 Nc5-b3
20- 00:04 17,722k 4,398k -1.68 Nb3-c5 Bb7-c8
20+ 00:04 19,607k 4,402k -1.40 Nb3-c5 Bb7-c8 Rf1xa1
20 00:04 20,209k 4,389k -1.66 Nb3-c5 Bb7-c8 Rf1xa1 Rf8-e8 Ra1-g1 Qd8-a5 Ba3-b2 Qa5-b4 Nc5-b3 Bc8-b7 Bb2-d4 Qb4-f8 Qh6-h3 Nb8-c6 Bd4-c5 Nc6-e7 Nb3-a5 Re8-b8 Bc5-d6 d5-d4+ Na5xb7 Rb8xb7 Bd3-e4 Rb7-a7 Rg1-g3
21+ 00:04 21,591k 4,372k -1.59 Nb3-c5 Bb7-c8 Rf1xa1
21- 00:05 22,518k 4,371k -1.66 Nb3-c5 Bb7-c8
21+ 00:05 23,298k 4,376k -1.63 Nb3-c5 Bb7-c8 Rf1xa1
21- 00:05 23,542k 4,369k -1.70 Nb3-c5 Bb7-c8
21+ 00:05 23,650k 4,365k -1.65 Nb3-c5 Bb7-c8 Rf1xa1
21 00:05 24,264k 4,360k -1.60 Nb3-c5 Bb7-c8 Rf1xa1 Rf8-e8 Ra1-g1 Qd8-a5 Ba3-b2 Qa5-b4 Nc5-b3 Bc8-b7 Bb2-d4 Qb4-f8 h2-h4 Nb8-c6 Bd4-c5 Qf8xh6 g5xh6 d5-d4 Kh1-h2 Nc6-e7 Bc5-d6 Ne7-f5 Nb3-c5 Bb7-c6
22- 00:06 26,606k 4,327k -1.67 Nb3-c5 Bb7-c8
22+ 00:06 28,423k 4,325k -1.65 Nb3-c5 Bb7-c8 Rf1xa1
22- 00:06 28,528k 4,322k -1.70 Nb3-c5 Bb7-c8
22- 00:06 29,531k 4,316k -1.86 Nb3-c5 Bb7-c8
22+ 00:07 30,529k 4,304k -1.81 Nb3-c5 Bb7-c8 Rf1xa1
22 00:07 31,492k 4,295k -1.76 Nb3-c5 Bb7-c8 Rf1xa1 Rf8-e8 Ra1-g1 Qd8-a5 Ba3-b2 Qa5-b4 Nc5-b3 Nb8-c6 Rg1-g3 Qb4-f8 Qh6-h4 Qf8-g7 Qh4-g4 Ra8-a7 h2-h4 Ra7-c7 Nb3-d4 Nc6-b4 Bb2-a3 Nb4xd3 c2xd3 Bc8-d7 h4-h5 g6xh5 Qg4xh5
23- 00:08 35,049k 4,299k -1.83 Nb3-c5 Bb7-c8
23+ 00:08 35,884k 4,294k -1.81 Nb3-c5 Bb7-c8 Rf1xa1
23+ 00:08 38,181k 4,301k -1.69 Nb3-c5 Bb7-c8 Rf1xa1
23- 00:08 38,329k 4,298k -1.77 Nb3-c5 Bb7-c8
23+ 00:09 42,545k 4,277k -1.72 Nb3-c5 Bb7-c8 Rf1xa1
23- 00:10 43,487k 4,274k -1.85 Nb3-c5 Bb7-c8
23+ 00:10 44,408k 4,268k -1.42 f4-f5 e6xf5 Rf1-f4
23+ 00:11 48,550k 4,262k +3.10 f4-f5 e6xf5 Rf1-f4
23 00:11 49,293k 4,263k +2.71 f4-f5 e6xf5 Rf1-f4 f7-f6 Bd3xf5 Bb7-c8 e5xf6 Ra8-a7 Bf5xg6 b5-b4 Rf4-h4 Rf8-f7 Bg6xf7+ Ra7xf7 g5-g6 Qd8xf6 g6xf7+ Qf6xf7 Qh6-g5+ Kg8-h8 Ba3xb4