EN-Test 2022 - new testsuite

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

Moderator: Ras

DrEinstein
Posts: 75
Joined: Wed Sep 15, 2021 8:50 pm
Full name: Albert Einstein

Re: EN-Test 2022 - new testsuite

Post by DrEinstein »

So the answer to my question would probably be:
SFdevs are not so much interested in solving chess puzzles than they are interested in winning games and matches.
I think with this answer I can sleep quite well tonight. Thus, good night and good bye to all :D
Edit: Forgot in my answer: SF won all extremelyLTC games matches in TCEC some days ago quite convincingly with these 'bullet game training and optimizations' on fishtest.
Last edited by DrEinstein on Sat Nov 19, 2022 7:12 pm, edited 1 time in total.
Eduard
Posts: 1439
Joined: Sat Oct 27, 2018 12:58 am
Location: Germany
Full name: N.N.

Re: EN-Test 2022 - new testsuite

Post by Eduard »

My test is not just about chess puzzles. Several positions are even from correspondence chess games (the positions are solved by almost all clones, except for Stockfish dev). :(

For example, the following position:
[fen]5b2/5k1p/1pb2pp1/1r1p4/3P1B1P/P7/1P3PP1/2R3K1 w - - 0[/fen]


Another example: Here played with Stockfish myself (I losed), live on the server and Stockfish was blind against Lc0 and lost because he didn't see b4 coming. All clones solve the position, only Stockfish dev. unfortunately not.

Sleep well. :mrgreen:
Uri Blass
Posts: 10912
Joined: Thu Mar 09, 2006 12:37 am
Location: Tel-Aviv Israel

Re: EN-Test 2022 - new testsuite

Post by Uri Blass »

Eduard wrote: Sat Nov 19, 2022 6:57 pm My test is not just about chess puzzles. Several positions are even from correspondence chess games (the positions are solved by almost all clones, except for Stockfish dev). :(

For example, the following position:
[fen]5b2/5k1p/1pb2pp1/1r1p4/3P1B1P/P7/1P3PP1/2R3K1 w - - 0[/fen]


Another example: Here played with Stockfish myself (I losed), live on the server and Stockfish was blind against Lc0 and lost because he didn't see b4 coming. All clones solve the position, only Stockfish dev. unfortunately not.

Sleep well. :mrgreen:
Stockfish with 7 cores can find b4 after some minutes of search
It is relatively slow relative to engines that are not based on stockfish and even old fruit2.1 with a single core get it faster.

[fen]5b2/5k1p/1pb2pp1/1r1p4/3P1B1P/P7/1P3PP1/2R3K1 w - - 0 1[/fen]


Stockfish_22111908_x64_avx2:
NNUE evaluation using nn-ad9b42354671.nnue enabled
1/1 00:00 235 118k +0.27 Rc1xc6 Rb5xb2
2/2 00:00 671 336k +0.27 Rc1xc6 Rb5xb2
3/3 00:00 911 456k +0.43 Rc1xc6 Rb5xb2 Rc6-c7+ Bf8-e7
4/4 00:00 1k 537k +0.43 Rc1xc6 Rb5xb2 Rc6-c7+ Bf8-e7
5/5 00:00 1k 692k +0.43 Rc1xc6 Rb5xb2 Rc6-c7+ Bf8-e7
6/5 00:00 3k 1,601k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Bf8-e7
7/8 00:00 8k 2,797k 0.00 Rc1xc6 Rb5xb2 Bf4-d6 Bf8xd6 Rc6xd6 Rb2-d2
8/10 00:00 16k 3,894k -0.02 Rc1xc6 Rb5xb2 Bf4-d6 Bf8xd6 Rc6xd6 Kf7-e7
9/10 00:00 26k 5,255k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8xa3 Rc7xh7 Rb2-b4 Rh7-b7 Rb4xd4 Rb7xb6+ Ke6-f7
10/13 00:00 41k 5,099k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8-e7 Kg1-f1 h7-h5 Rc7-c6+ Ke6-d7
11/15 00:00 58k 5,762k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8-e7 Kg1-f1 h7-h5 Rc7-c6+ Ke6-d7 Rc6-c7+ Kd7-e6
12/13 00:00 62k 5,655k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8-e7 Rc7-c6+ Ke6-d7 Rc6-c7+
13/15 00:00 79k 5,664k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8-e7 Rc7-c6+ Ke6-d7 Rc6-c7+
14/12 00:00 100k 6,236k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8-e7 Rc7-c6+ Ke6-d7 Rc6-c7+ Kd7-e6
15/20 00:00 194k 6,470k -0.05 Rc1xc6 Rb5xb2 Bf4-d6 Bf8xd6 Rc6xd6 Kf7-e7 Rd6xd5 Rb2-b3 a3-a4 Rb3-b4 Kg1-h2 Ke7-e6 Rd5-d8 Rb4xa4 Kh2-g3
16/20 00:00 264k 6,434k 0.00 Rc1xc6 Rb5xb2 g2-g4 Bf8xa3 Rc6-c7+ Kf7-e6 Rc7-c6+
17/23 00:00 370k 6,611k 0.00 Rc1xc6 Rb5xb2 g2-g4 Bf8xa3 Rc6-c7+ Kf7-e6 Rc7-c6+
18/8 00:00 381k 6,682k 0.00 Rc1xc6 Rb5xb2 g2-g4 Bf8xa3 Rc6-c7+ Kf7-e6 Rc7-c6+
19/8 00:00 400k 6,778k 0.00 Rc1xc6 Rb5xb2 g2-g4 Bf8xa3 Rc6-c7+ Kf7-e6 Rc7-c6+
20/8 00:00 418k 6,750k 0.00 Rc1xc6 Rb5xb2 g2-g4 Bf8xa3 Rc6-c7+ Kf7-e6 Rc7-c6+
21/11 00:00 493k 6,752k 0.00 Rc1xc6 Rb5xb2 g2-g4 Bf8xa3 Rc6-c7+ Kf7-e6 Rc7-c6+
22/24 00:00 733k 7,047k 0.00 Rc1xc6 Rb5xb2 g2-g4 Bf8xa3 Rc6-c7+ Kf7-e6 Rc7-c6+
23/20 00:00 956k 7,132k 0.00 Rc1xc6 Rb5xb2 g2-g4 Bf8xa3 Rc6-c7+ Kf7-e6 Rc7-c6+
24/25 00:00 1,393k 7,179k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7xh7 Bf8xa3 Rh7-b7 Rb2-b1+ Kg1-h2 Ke6-f5 Bf4-e3 Kf5-e4 Rb7-g7 Ke4-f5
25/28 00:00 1,770k 7,285k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8xa3 Rc7-c6+ Ke6-f7 Rc6-c7+
26/11 00:00 1,872k 7,282k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8xa3 Rc7-c6+ Ke6-f7 Rc6-c7+ Kf7-e6
27/10 00:00 2,119k 7,307k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8xa3 Rc7-c6+ Ke6-f7 Rc6-c7+
28/28 00:00 2,424k 7,345k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Rb2-b1+ Kg1-g2 h7-h5 Rc7-c6+ Ke6-f7 g4-g5 f6xg5 h4xg5 Bf8-e7 a3-a4 Rb1-b4 Bf4-e3 Rb4xa4 Rc6xb6 Ra4-a5 Be3-f4 Ra5-a7 Kg2-g1 Ra7-a2 Kg1-g2
29/33 00:00 3,128k 7,536k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Rb2-b1+ Kg1-g2 h7-h5 Rc7-c6+ Ke6-f7 g4-g5 f6xg5 h4xg5 Bf8-e7 a3-a4 Rb1-b4 Bf4-e3 Rb4xa4 Rc6xb6 Ra4-a5 Be3-f4 Ra5-a7 Kg2-g1 Ra7-a2 Kg1-g2 Ra2-c2 Bf4-e3 Rc2-c3 Kg2-g1 Rc3-a3 Kg1-g2 Ra3-a1
30/21 00:00 3,773k 7,653k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Rb2-b1+ Kg1-h2 h7-h5 Rc7-c6+ Ke6-f7 Rc6-c7+ Kf7-e6
31/13 00:00 4,254k 7,692k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Rb2-b1+ Kg1-h2 h7-h5 Rc7-c6+ Ke6-f7 Rc6-c7+ Kf7-e6
32/23 00:00 4,629k 7,651k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Rb2-b1+ Kg1-h2 h7-h5 Rc7-c6+ Ke6-f7 Rc6-c7+
33/12 00:00 5,697k 7,516k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Rb2-b1+ Kg1-h2 Bf8-e7 Rc7-c6+ Ke6-d7 Rc6-c7+
34/12 00:00 7,344k 7,366k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8xa3 Rc7-c6+ Ke6-f7 Rc6-c7+ Kf7-e6
35/31 00:01 10,756k 7,048k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7-c6+
36/20 00:01 12,008k 7,006k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8xa3 Rc7-c6+ Ke6-e7 Kg1-g2 b6-b5 Rc6-c7+ Ke7-e6 Rc7xh7 Rb2-b4 Rh7-b7 Rb4xd4 Rb7-b6+ Ke6-f7 Rb6-b7+
37/21 00:01 13,549k 7,006k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8xa3 Rc7-c6+ Ke6-e7 Kg1-g2 b6-b5 Rc6-c7+ Ke7-e6 Rc7xh7 Rb2-b4 Rh7-b7 Rb4xd4 Rb7-b6+ Ke6-f7 Rb6-b7+
38/21 00:02 15,248k 6,982k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8xa3 Rc7-c6+ Ke6-e7 Kg1-g2 b6-b5 Rc6-c7+ Ke7-e6 Rc7xh7 Rb2-b4 Rh7-b7 Rb4xd4 Rb7-b6+ Ke6-f7 Rb6-b7+ Kf7-e6
39/6 00:02 18,863k 6,887k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7-c6+
40/44 00:05 39,236k 6,857k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7-c6+ Ke6-f7
41/11 00:06 41,712k 6,856k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7-c6+
42/7 00:08 58,369k 6,884k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7-c6+ Ke6-f7
43/13 00:08 59,578k 6,895k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Rb2-b1+ Kg1-h2 h7-h5 Rc7-c6+ Ke6-f7 Rc6-c7+
44/21 00:11 78,816k 6,948k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7-c6+
45/53 00:18 129,416k 6,988k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8xa3 Rc7-c6+ Ke6-e7 Kg1-g2 b6-b5 Rc6-c7+ Ke7-e6 Rc7xh7 Rb2-b4 Rh7-b7 Rb4xd4 Rb7-b6+ Ke6-f7 Rb6-b7+ Kf7-g8 Rb7-b8+ Kg8-f7
46/25 00:18 130,690k 6,986k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8xa3 Rc7-c6+ Ke6-e7 Kg1-g2 b6-b5 Rc6-c7+ Ke7-e6 Rc7xh7 Rb2-b4 Rh7-b7 Rb4xd4 Rb7-b6+ Ke6-f7 Rb6-b7+ Kf7-g8 Rb7-b8+ Kg8-h7 Rb8-b7+ Kh7-g8
47/24 00:19 134,126k 6,986k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8xa3 Rc7-c6+ Ke6-e7 Kg1-g2 b6-b5 Rc6-c7+ Ke7-e6 Rc7xh7 Rb2-b4 Rh7-b7 Rb4xd4 Rb7-b6+ Ke6-f7 Rb6-b7+
48/24 00:19 138,435k 7,000k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8xa3 Rc7-c6+ Ke6-e7 Kg1-g2 b6-b5 Rc6-c7+ Ke7-e6 Rc7xh7 Rb2-b4 Rh7-b7 Rb4xd4 Rb7-b6+ Ke6-f7 Rb6-b7+
49/24 00:23 162,843k 7,000k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7-c6+
50/10 00:25 175,236k 6,991k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8xa3 Rc7-c6+ Ke6-f7 Rc6-c7+
51/10 00:26 188,504k 6,990k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8xa3 Rc7-c6+ Ke6-f7 Rc6-c7+
52/10 00:30 211,375k 7,004k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8xa3 Rc7-c6+ Ke6-f7 Rc6-c7+
53/10 00:39 269,471k 6,901k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8xa3 Rc7-c6+ Ke6-f7 Rc6-c7+
54/10 00:40 281,142k 6,889k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8xa3 Rc7-c6+ Ke6-f7 Rc6-c7+
55/10 00:51 350,775k 6,843k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8xa3 Rc7-c6+ Ke6-f7 Rc6-c7+
56/66 03:45 1,479,376k 6,574k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 g2-g4 Bf8xa3 Rc7-c6+ Ke6-f7 Rc6-c7+
57/55 03:45 1,480,760k 6,575k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7xh7 Bf8xa3 g2-g3 Rb2-b4 Rh7-g7 Ke6-f5 Bf4-e3 b6-b5 Kg1-g2 Ba3-b2 Kg2-f3 Bb2xd4 g3-g4+ Kf5-e5 Rg7xg6 Bd4xe3 f2xe3 f6-f5 g4xf5 Rb4xh4 Rg6-b6 Ke5xf5 Rb6xb5 Kf5-e5 Rb5-b8 Rh4-h5 Kf3-e2 Ke5-d6 Rb8-c8 Rh5-h2+ Ke2-d3 Kd6-d7 Rc8-g8 Kd7-d6 Rg8-d8+ Kd6-c6 Rd8-f8 Kc6-d7
58/62 03:45 1,481,341k 6,574k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7xh7 Bf8xa3 g2-g3 Rb2-b4 Rh7-g7 Ke6-f5 Bf4-e3 b6-b5 Kg1-g2 Ba3-b2 Kg2-f3 Bb2xd4 g3-g4+ Kf5-e5 Rg7xg6 Bd4xe3 f2xe3 f6-f5 g4xf5 Rb4xh4 Rg6-b6 Ke5xf5 Rb6xb5 Kf5-e5 Rb5-b8 Rh4-h5 Kf3-e2 Ke5-d6 Rb8-c8 Rh5-h2+ Ke2-d3 Kd6-d7 Rc8-g8 Kd7-d6 Rg8-d8+ Kd6-c6 Rd8-f8 Kc6-d7 Kd3-d4 Kd7-e6 Rf8-e8+ Ke6-d6 e3-e4 d5xe4 Re8xe4 Rh2-a2 Re4-f4 Ra2-a4+ Kd4-e3 Ra4xf4 Ke3xf4 Kd6-d5 Kf4-f5 Kd5-d4 Kf5-g4 Kd4-e4 Kg4-g5 Ke4-f3
59/51 03:45 1,482,104k 6,574k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7xh7 Bf8xa3 g2-g3 Rb2-b4 Rh7-g7 Ke6-f5 Bf4-e3 b6-b5 Kg1-g2 Ba3-b2 Kg2-f3 Bb2xd4 g3-g4+ Kf5-e5 Rg7xg6 Bd4xe3 f2xe3 f6-f5 g4xf5 Rb4xh4 Rg6-b6 Ke5xf5 Rb6xb5 Kf5-e5 Rb5-b8 Rh4-h5 Kf3-e2 Ke5-d6 Rb8-c8 Rh5-h2+ Ke2-d3 Kd6-d7 Rc8-g8 Kd7-d6
60/55 03:50 1,512,558k 6,568k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7xh7 Bf8xa3 g2-g3 Rb2-b4 Rh7-g7 Ke6-f5 Bf4-e3 b6-b5 Kg1-g2 Ba3-b2 Kg2-f3 Bb2xd4 g3-g4+ Kf5-e5 Rg7xg6 Bd4xe3 f2xe3 f6-f5 g4xf5 Rb4xh4 Rg6-b6 Ke5xf5 Rb6xb5 Kf5-e5 Rb5-b8 Rh4-h5 Kf3-e2 Ke5-d6 Rb8-c8 Rh5-h2+ Ke2-d3 Kd6-d7 Rc8-g8 Kd7-d6 Rg8-d8+ Kd6-c6 Rd8-f8 Kc6-d7
61/54 04:00 1,579,460k 6,555k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7xh7 Bf8xa3 g2-g3 Rb2-b4 Rh7-g7 Ke6-f5 Bf4-e3 b6-b5 Kg1-g2 Ba3-b2 Kg2-f3 Bb2xd4 g3-g4+ Kf5-e5 Rg7xg6 Bd4xe3 f2xe3 f6-f5 g4xf5 Rb4xh4 Rg6-b6 Ke5xf5 Rb6xb5 Kf5-e5 Rb5-b8 Rh4-h5 Kf3-e2 Ke5-d6 Rb8-c8 Rh5-h2+ Ke2-d3 Kd6-d7 Rc8-g8 Kd7-d6 Rg8-d8+ Kd6-c6 Rd8-f8 Kc6-d7
62/59 04:04 1,603,842k 6,548k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7xh7 Bf8xa3 g2-g3 Rb2-b4 Rh7-g7 Ke6-f5 Bf4-e3 b6-b5 Kg1-g2 Ba3-b2 Kg2-f3 Bb2xd4 g3-g4+ Kf5-e5 Rg7xg6 Bd4xe3 f2xe3 f6-f5 g4xf5 Rb4xh4 Rg6-b6 Ke5xf5 Rb6xb5 Kf5-e5 Rb5-b8 Rh4-h5 Kf3-e2 Ke5-d6 Rb8-c8 Rh5-h2+ Ke2-d3 Kd6-d7 Rc8-g8 Kd7-d6 Rg8-a8 Kd6-c6 Ra8-d8 Rh2-h4 Rd8-b8 Kc6-c5 Rb8-c8+ Kc5-d6 Kd3-c3 Rh4-h2 Rc8-a8 Kd6-c6 Kc3-d3 Kc6-d6 Ra8-d8+
63/49 04:25 1,734,718k 6,536k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7xh7 Bf8xa3 g2-g3 Rb2-b4 Rh7-g7 Ke6-f5 Bf4-e3 b6-b5 Kg1-g2 Ba3-b2 Kg2-f3 Bb2xd4 g3-g4+ Kf5-e5 Rg7xg6 Bd4xe3 f2xe3 f6-f5 g4xf5 Rb4xh4 Rg6-b6 Ke5xf5 Rb6xb5 Kf5-e5 Rb5-b8 Rh4-h5 Kf3-e2 Ke5-d6 Rb8-c8 Rh5-h2+ Ke2-d3 Kd6-d7 Rc8-g8 Kd7-d6
64/59+ 06:11 2,374,583k 6,391k +0.02 b2-b4
64/59+ 06:17 2,412,587k 6,395k +0.05 b2-b4
64/59+ 06:55 2,662,539k 6,412k +0.60 b2-b4

FEN: 5b2/5k1p/1pb2pp1/1r1p4/3P1B1P/P7/1P3PP1/2R3K1 w - - 0 1

Fruit_21:
1/2 00:00 3 14 +0.04 Rc1xc6 Rb5xb2
2/6 00:00 172 851 +0.23 Rc1xc6 Rb5xb2 Rc6-c7+ Bf8-e7
3/6 00:00 417 2k +0.23 Rc1xc6 Rb5xb2 Rc6-c7+ Bf8-e7
4/10 00:00 1k 6k +0.28 Rc1xc6 Rb5xb2 Rc6-c7+ Bf8-e7 Bf4-d6
5/16 00:00 4k 22k +0.13 Rc1xc6 Rb5xb2 a3-a4 Rb2-b1+ Kg1-h2 Rb1-b2 Rc6-c7+ Bf8-e7
6/16 00:00 13k 63k +0.06 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7xh7 Rb2-b1+ Kg1-h2 Bf8xa3
7/18 00:00 33k 162k -0.01 Rc1xc6 Rb5xb2 Bf4-c1 Rb2-a2 Rc6xb6 Bf8xa3 Bc1-f4
8/18 00:00 77k 2,475k 0.00 Rc1xc6 Rb5xb2 Bf4-c1 Rb2-b3 Rc6-c7+ Kf7-g8 Rc7-c8 Kg8-f7 Rc8-c7+
9/20 00:00 170k 3,614k -0.09 Rc1xc6 Rb5xb2 Bf4-c1 Rb2-b3 a3-a4 Rb3-b4 Rc6-c7+ Kf7-g8 Bc1-f4 Rb4xa4
10/23 00:00 383k 3,484k -0.15 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7xh7 Bf8xa3 Kg1-f1 Ke6-f5 Rh7-a7 Rb2-c2 Ra7-c7
11/25 00:00 666k 3,541k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7xh7 Bf8xa3 g2-g4 Rb2-b4 Rh7-b7 b6-b5 Rb7-b6+ Ke6-f7 Rb6-b7+ Kf7-e6
12/26 00:00 1,131k 3,613k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7xh7 Bf8xa3 g2-g4 Rb2-e2 Rh7-b7 Re2-e4 Rb7xb6+ Ke6-f7 Rb6-b7+ Kf7-e6 Rb7-b6+
13/27 00:00 2,253k 3,693k -0.02 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7xh7 Bf8xa3 g2-g4 Rb2-b4 Bf4-e3 Ba3-b2 Rh7-g7 b6-b5 Kg1-g2 Rb4-c4 Rg7xg6 Bb2xd4 Be3xd4 Rc4xd4
14/28 00:01 4,176k 3,712k -0.12 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7xh7 Bf8xa3 g2-g4 Rb2-e2 Rh7-b7 Re2-e4 Rb7xb6+ Ke6-f7 Rb6-b7+ Ba3-e7 Bf4-d6 Re4xg4+ Kg1-f1 Rg4-e4
15/33 00:03 12,024k 3,629k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7-c6+ Ke6-f7
16/33 00:04 17,375k 3,669k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7-c6+ Ke6-f7
17/41 00:12 47,433k 3,688k -0.07 Rc1xc6 Rb5xb2 Bf4-d6 Bf8xd6 Rc6xd6 Rb2-b3 g2-g3 h7-h5 Rd6xd5 Kf7-e7 Kg1-g2 Rb3xa3 Rd5-b5 Ra3-a4 Rb5xb6 Ra4xd4 Rb6-b7+ Ke7-e6 Kg2-f3
18/41 00:21 80,855k 3,723k -0.02 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7-c6+ Ke6-f5 Bf4-c7 b6-b5 f2-f3 h7-h5 Bc7-d8 Bf8-g7 Rc6-d6 b5-b4 Rd6xd5+ Kf5-e6 Rd5-a5 b4-b3 Ra5-a6+ Ke6-f5 Ra6-a5+ Kf5-f4 Ra5-c5
19/41 00:33 123,993k 3,708k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7-c6+ Ke6-f5 Bf4-c7 b6-b5 f2-f3 h7-h5 Kg1-f1 Bf8-e7 Bc7-d6 Be7-d8 Bd6-b4 Kf5-f4 Rc6-e6 Kf4-g3 Bb4-e1+ Kg3-f4 Be1-b4
20/45 01:00 224,263k 3,696k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7-c6+ Ke6-f5 Bf4-c7 b6-b5 f2-f3 h7-h5 Kg1-f1 Bf8-e7 Bc7-d6 Be7-d8 Bd6-b4 Kf5-f4 Rc6-e6 Kf4-g3 Bb4-e1+ Kg3-f4 Be1-b4
21/45 02:02 440,024k 3,600k 0.00 Rc1xc6 Rb5xb2 Rc6-c7+ Kf7-e6 Rc7-c6+ Ke6-f5 Bf4-c7 b6-b5 f2-f3 h7-h5 Kg1-f1 Bf8-e7 Bc7-d6 Be7-d8 Bd6-b4 Kf5-f4 Rc6-e6 Kf4-g3 Bb4-e1+ Kg3-f4 Be1-b4
21/45 03:15 697,251k 3,559k +0.29 b2-b4 Bc6-d7 Rc1-b1 Kf7-e6 Bf4-d2 Bf8-c5 Bd2-c3 h7-h5 Rb1-b2 g6-g5 g2-g3 g5xh4 g3xh4 Bc5-d6 a3-a4 Ke6-f5 f2-f3 Bd6-g3 a4xb5 Bd7xb5 Kg1-g2 Bg3xh4 Rb2-a2
22/45 04:26 938,055k 3,524k +0.31 b2-b4 Bc6-d7 Rc1-b1 Kf7-e6 Bf4-d2 Bf8-c5 Bd2-c3 f6-f5 Rb1-b2 f5-f4 b4xc5 Rb5xb2 Bc3xb2 b6xc5 d4xc5 d5-d4 Bb2xd4 Ke6-d5 Bd4-c3 Kd5xc5 Bc3-e5 f4-f3 g2xf3 Bd7-e6
23/45 07:26 1,561,295k 3,497k +0.37 b2-b4 Bc6-d7 Rc1-b1 Kf7-e6 Bf4-d2 Bf8-c5 Bd2-c3 h7-h5 f2-f3 f6-f5 Kg1-f2 Ke6-f7 Rb1-b2 Bc5-e7 g2-g3 f5-f4 a3-a4 g6-g5 h4xg5 f4xg3+ Kf2xg3 Be7xg5 Rb2-g2 h5-h4+ Kg3-f2 Bg5-f4 a4xb5 Bd7xb5
Eduard
Posts: 1439
Joined: Sat Oct 27, 2018 12:58 am
Location: Germany
Full name: N.N.

Re: EN-Test 2022 - new testsuite

Post by Eduard »

Thanks!

My engine takes less than 30s.

54: Bc0 - Stockfish Line 0.0, b4!
5b2/5k1p/1pb2pp1/1r1p4/3P1B1P/P7/1P3PP1/2R3K1 w - - 0 1

Analysis by Solista Attack v5:

28.b4
= (0.05++) Depth: 51/44 00:00:25 492MN, tb=623654
The position is equal

Stockfish dev. now calculates more selectively than before. A few more moves will be skipped faster. That's good for Bullet but bad for me, sorry. That's why Stockfish now only solves 99 positions.
Eduard
Posts: 1439
Joined: Sat Oct 27, 2018 12:58 am
Location: Germany
Full name: N.N.

Re: EN-Test 2022 - new testsuite

Post by Eduard »

Stockfish dev and a new Stockfish-clone are BLIND!

Live from PlayChess.com:

[fen]r3r2k/p1pq3p/R3bppN/2P5/1p1pP1P1/3P4/2P2P1P/3Q2RK b - - 0 1[/fen]


Analysis by Stockfish dev-20221119-41c6a74d:

30...Qb5 31.Qf3 Kg7 32.Rxe6 Rxe6 33.Nf5+ Kf7 34.Qh3 Kg8 35.Nxd4 Qe8 36.Nxe6 Qxe6 37.Qe3 a5 38.Qd4 Rb8 39.Qd5 Qxd5 40.exd5 a4 41.Kg2 Rd8 42.Rb1 Rxd5 43.Rxb4 Rxc5 44.Rxa4 Rxc2 45.Kf3 Kf7 46.h4 Rc1 47.Ra5 Ke7 48.Ra7 Kd6 49.Ra6+ Ke7 50.Kf4 Rh1 51.Rc6 Rxh4 52.Rxc7+ Kd6 53.Ra7 h5 54.f3 Ke6 55.Ra6+ Ke7 56.Ra4
= (0.00) Depth: 44/51 00:00:20 393MN, tb=26327
The position is equal

Stockfish is blind, after 20s on 20 Threads -->0.00??

Analysis by Deutronius 201122:

30...Qe7 31.Qd2 Qxc5 32.h4 Re7 33.Rga1 Rf8 34.Rxa7 Kg7 35.Kg1 Qc3 36.Qf4 Bc8 37.Kg2 Qc5 38.Qd2 Qc3 39.Qxc3 dxc3 40.g5 fxg5 41.hxg5 b3 42.cxb3 Rd8 43.Ra8 Red7 44.d4 Rxd4 45.R8a7 Be6 46.R1a6 Rxe4 47.Rxc7+ Rd7 48.Rxc3 Re5 49.f4 Re4 50.Rb6
+/= (0.54 --) Depth: 32/40 00:00:10 181MN, tb=1294

My Engine Deutronius sees the loss already after 3s, and need to play very soon Qe7. White is better but is not won.

PGN:

[Event "5 min, rated"]
[Site "Engine Room"]
[Date "2022.11.20"]
[Round "?"]
[White "A2-a3, Deutronius 131122-avx2"]
[Black "Blañk Queen, Tvar'Darkodev-20221"]
[Result "1-0"]
[ECO "C65"]
[WhiteElo "2940"]
[BlackElo "2947"]
[Annotator "0.10;0.19"]
[PlyCount "101"]
[EventDate "2022.11.20"]
[EventType "blitz"]
[TimeControl "300"]

1. e4 {B 0} e5 {B 0} 2. Nf3 {B 0} Nc6 {B 0} 3. Bb5 {B 0} Nf6 {B 0} 4. d3 {B 0} Bc5 {B 0} 5. Bxc6 {B 0} dxc6 {B 0} 6. O-O {B 0} Nd7 {B 0} 7. Nbd2 {B 0} O-O {B 0} 8. Nc4 {B 0} Re8 {B 0} 9. Bd2 {B 0} f6 {B 0} 10. Nh4 {B 0} g6 {B 0} 11. Kh1 {B 0} Nb6 {B 0} 12. Na5 {B 0} Bf8 {B 0} 13. a4 {B 0} Nd7 {B 0} 14. g3 {B 0} Nc5 {B 0} 15. Nc4 {B 0} Ne6 {B 0} 16. Ng2 {B 0} b5 {B 0} 17. Na5 {B 0} Bd7 {B 0} 18. Ne3 {0.10/36 46} Bg7 {0.19/24 8 (Rb8)} 19. Bc3 {0.24/31 4} Nd4 {0.12/27 0} 20. axb5 {0.21/31 7} cxb5 {0.15/30 0} 21. Nb7 {0.19/33 7} Qc8 {0.14/30 0} 22. Nc5 {0.25/33 6} Bh3 {0.19/32 3} 23. Rg1 {0.16/32 4} Bf8 {0.10/31 0} 24. b4 {0.16/36 4} Bxc5 {0.09/29 0} 25. bxc5 {0.11/34 5} Be6 {0.12/32 0} 26. g4 {0.06/35 5} Qd7 {0.20/30 8} 27. Ra6 {0.04/40 14} b4 {0.00/28 6 (Qe7)} 28. Bxd4 {0.43/29 4} exd4 {0.00/33 0} 29. Nf5 {0.49/30 4} Kh8 {0.00/35 0} 30. Nh6 {0.49/31 4} Qb5 {0.00/35 3 (Qe7)} 31. Qf3 {3.62/26 6} Kg7 {0.10/34 1 (Qd7)} 32. e5 {5.66/26 6} Qxa6 {1.05/47 24} 33. Qxf6+ {7.21/29 0} Kxh6 {4.39/21 0} 34. g5+ {7.50/27 4} Kh5 {5.18/23 0} 35. Qf3+ {7.52/29 6} Kh4 {5.79/23 3} 36. Qf4+ {7.74/31 10} Kh5 {5.66/23 3} 37. Qf3+ {7.79/29 0} Kh4 {5.68/23 2} 38. h3 {7.85/28 4} Rf8 {5.98/27 6} 39. Qe4+ {7.92/28 0} Kh5 {5.92/27 4} 40. Qe2+ {8.05/33 11} Kh4 {6.04/25 5} 41. Qe4+ {8.11/29 0} Kh5 {6.09/24 3} 42. Kh2 {8.19/30 1} Bc8 {6.08/25 5} 43. Qe2+ {8.52/31 4} Kh4 {6.35/24 0} 44. Rg4+ {10.32/31 23} Kh5 {6.39/30 0} 45. Rg2+ {10.77/28 5} Kh4 {6.75/29 8} 46. Rg4+ {10.99/29 1} Kh5 {6.83/25 2} 47. Rg1+ {11.97/30 21} Kh4 {6.04/23 5} 48. Qe4+ {12.72/25 0} Kh5 {6.61/28 13} 49. f4 {12.99/28 0} Qe6 {6.69/25 6} 50. Qe2+ {13.95/24 2} Kh4 {6.72/24 4} 51. Qf2+ {14.98/25 1 Blaсk Queen,Tvar'Darkodev-202211 Сдается (Lag: Av=0.37s, max=0.7s)} 1-0

Impressive tactical win by Deutronius! :-)
Last edited by Eduard on Sun Nov 20, 2022 6:32 am, edited 2 times in total.
Eduard
Posts: 1439
Joined: Sat Oct 27, 2018 12:58 am
Location: Germany
Full name: N.N.

Re: EN-Test 2022 - new testsuite

Post by Eduard »

What a blind engine!

r3r2k/p1pq3p/R3bppN/2P5/1p1pP1P1/3P4/2P2P1P/3Q2RK b - - 0 1

Analysis by Stockfish dev-20221119-41c6a74d:

30...Qe7 31.Qd2 Qxc5 32.f4 Qf8 33.f5 Bc8 34.fxg6 Bxa6 35.Nf7+ Kg8 36.Nh6+
The position is equal: = (0.00) Depth: 10/14 00:00:00
30...Qe7 31.Qd2 Qxc5 32.f4 Qf8 33.f5 Bc8 34.fxg6 Bxa6 35.Nf7+ Kg8 36.Nh6+ Kh8
The position is equal: = (0.00) Depth: 11/14 00:00:00
30...Qe7 31.Qd2 Qxc5 32.f4 Qf8 33.f5 Bc8 34.fxg6 Bxa6 35.Nf7+ Kg8 36.Nh6+ Kh8
The position is equal: = (0.00) Depth: 12/21 00:00:00
30...Qe7 31.Qd2 Bc8 32.Ra5 Kg7 33.f3 Rf8 34.Kg2 a6 35.Rb1 Rb8 36.h4
The position is equal: = (0.21) Depth: 13/21 00:00:00 1008kN
30...Qb5 31.Qa1 Qxc5 32.f4 c6 33.f5 Bd5 34.Ra5 Qxc2 35.Qxd4
The position is equal: = (0.16) Depth: 14/24 00:00:00 1277kN
30...Qb5 31.Qf3 Kg7 32.Rxe6 Rxe6 33.Nf5+ Kf7 34.Nxd4 Qxc5 35.Nxe6 Kxe6 36.g5 Rf8 37.Qh3+ Kd6 38.gxf6
The position is equal: = (0.00) Depth: 15/24 00:00:00 1593kN, tb=1
30...Qb5 31.Qf3 Kg7 32.e5 Qxa6 33.Qxf6+ Kxh6 34.g5+ Kh5 35.Qf3+ Kh4 36.Qe4+ Kh5
The position is equal: = (0.00) Depth: 16/25 00:00:00 1759kN, tb=1
30...Qb5 31.Qf3 Kg7 32.e5 Qxa6 33.Qxf6+ Kxh6 34.g5+ Kh5 35.Qf3+ Kh4 36.Qe4+ Kh5 37.Qe2+ Kh4
The position is equal: = (0.00) Depth: 17/16 00:00:00 2012kN, tb=2
30...Qb5 31.Qf3 Kg7 32.Rxe6 Rxe6 33.Nf5+ Kf7 34.Nxd4 Qxc5 35.Nxe6 Kxe6 36.g5 Rf8 37.d4 Qxc2 38.gxf6 Rxf6 39.Qh3+ Ke7 40.Qxh7+ Kf8 41.Qh8+ Kf7
The position is equal: = (0.00) Depth: 18/23 00:00:00 2223kN, tb=2
(----)
30...Qb5 31.Qf3 Kg7 32.Rxe6 Rxe6 33.Nf5+ Kf7 34.Qh3 Kg8 35.Nxd4 Qe8 36.Nxe6 Qxe6 37.f4 a5 38.f5 Qe7 39.Qh6 b3 40.Qc1 bxc2 41.Qxc2 Qe5 42.g5 fxg5 43.Qa2+ Kh8 44.fxg6 hxg6 45.Qf7 Rg8 46.Rg3 Qa1+ 47.Kg2 Qb2+ 48.Kh3 a4 49.Kg4 Qe2+ 50.Qf3 Qb2 51.Qf7
White is slightly better: +/= (0.33 --) Depth: 48/64 00:00:53 1012MN, tb=102045
30...Qb5 31.Qf3 Kg7 32.Rxe6 Rxe6 33.Nf5+ Kf7 34.Qh3 Kg8 35.Nxd4 Qe8 36.Nxe6 Qxe6 37.f4 a5 38.f5 Qe7 39.Qh6 b3 40.Qc1 bxc2 41.Qxc2 Qe5 42.g5 fxg5 43.Qa2+ Kh8 44.fxg6 hxg6 45.Qf7 Rg8 46.Rg3 Qa1+ 47.Kg2 Qb2+ 48.Kh3 a4 49.Kg4 Qe2+ 50.Qf3 Qb2 51.Qf7
White is slightly better: +/= (0.45 --) Depth: 48/64 00:01:05 1219MN, tb=104172
30...Qe7
White is slightly better: +/= (0.35 ++) Depth: 48/64 00:01:20 1498MN, tb=121400

Stockfish dev needs 1:20 Min to see the Blunder! :roll:

Analysis by Deutronius 201122:

30...Qb5 31.Qf3 Kg7 32.e5 Qxa6 33.Qxf6+ Kxh6 34.g5+ Kh5 35.Qf3+ Kh4 36.Qe4+ Kh5 37.h3 Qc8 38.Qf3+ Kh4 39.Qf4+ Kh5 40.Qf3+
The position is equal: = (0.00) Depth: 23/28 00:00:00 5465kN, tb=5
30...Qb5 31.Qf3 Kg7 32.e5 Qxa6 33.Qxf6+ Kxh6 34.g5+ Kh5 35.Qf3+ Kh4 36.Qe4+ Kh5 37.h3 Qc8 38.Qf3+ Kh4 39.Qf4+ Kh5 40.Qf3+
The position is equal: = (0.00) Depth: 24/18 00:00:00 7237kN, tb=8
30...Qe7 31.Qf3 Rab8 32.Qf4 g5 33.Qf3 Rb5 34.h4 Rxc5 35.Nf5 Bxf5 36.gxf5 gxh4 37.Qh5 Rf8 38.Qh6 Rg8 39.Rxg8+ Kxg8 40.Rxa7 Qf8 41.Qd2 Rc3 42.Ra4 Kh8 43.Kh2 c5 44.Ra6 c4 45.e5 Qb8 46.Qh6 Qxe5+ 47.Kg1
White is slightly better: +/= (0.49) Depth: 25/37 00:00:02 31602kN, tb=9
30...Qe7 31.Qf3 Rf8 32.Qf4 Bd7 33.Rga1 b3 34.cxb3 Bb5 35.Rxa7 Bxd3 36.f3 Rxa7 37.Rxa7 Qxc5 38.Qxc7 Qxc7 39.Rxc7 Bb5 40.Rb7 Bc6 41.Rb6 Rc8 42.Kg2 Kg7 43.g5 fxg5 44.Ng4 h5 45.Ne5 Be8 46.b4 Rc2+ 47.Kg3 Rb2 48.h4 gxh4+ 49.Kxh4 Rg2
White is slightly better: +/= (0.42) Depth: 26/45 00:00:03 54603kN, tb=33
30...Qe7 31.Qd2 Qxc5 32.h4 Re7 33.Rga1 Rf8 34.Kg1 Kg7 35.g5 fxg5 36.hxg5 Qe5 37.Rxa7 Rxf2 38.Qxf2 Qxg5+ 39.Kh1 Qxh6+ 40.Qh2 Qe3 41.Rf1 Bg4 42.Qf2 Qh3+ 43.Kg1 Qe3 44.Rb7 h5 45.Rxb4 c5 46.Rc4 Re5 47.Qxe3 dxe3
White is slightly better: +/= (0.38) Depth: 27/45 00:00:03 70425kN, tb=48
30...Qe7 31.Qd2 Qxc5 32.h4 Re7 33.Rga1 Rf8 34.Kg1 Kg7 35.g5 fxg5 36.hxg5 Qe5 37.Rxa7 Rxf2 38.Qxf2 Qxg5+ 39.Kh1 Qxh6+ 40.Qh2 Qe3 41.Rf1 Bg4 42.Qf2 Qh3+ 43.Kg1 Qe3 44.Rb7 h5 45.Rxb4 c5 46.Rc4 Re5 47.Qxe3 dxe3
White is slightly better: +/= (0.42 --) Depth: 28/33 00:00:04 74649kN, tb=51
30...Qe7 31.Qd2 Qxc5 32.h4 Re7 33.Rga1 Rf8 34.Kg1 Kg7 35.g5 fxg5 36.hxg5 Qe5 37.Rxa7 Rxf2 38.Qxf2 Qxg5+ 39.Kh1 Qxh6+ 40.Qh2 Qe3 41.Rf1 Bg4 42.Qf2 Qh3+ 43.Kg1 Qe3 44.Rb7 h5 45.Rxb4 c5 46.Rc4 Re5 47.Qxe3 dxe3
White is slightly better: +/= (0.45 --) Depth: 28/33 00:00:04 78981kN, tb=51
30...Qe7 31.Qd2 Qxc5 32.h4 Re7 33.Rga1 Rf8 34.Kg1 Kg7 35.g5 fxg5 36.hxg5 Qe5 37.Rxa7 Rxf2 38.Qxf2 Qxg5+ 39.Kh1 Qxh6+ 40.Qh2 Qe3 41.Rf1 Bg4 42.Qf2 Qh3+ 43.Kg1 Qe3 44.Rb7 h5 45.Rxb4 c5 46.Rc4 Re5 47.Qxe3 dxe3
White is slightly better: +/= (0.50 --) Depth: 28/33 00:00:04 83733kN, tb=54

Only 2 Seconds!

After Qb5:

r3r2k/p1p4p/R3bppN/1qP5/1p1pP1P1/3P4/2P2P1P/3Q2RK w - - 0 1

Analysis by Deutronius 201122:

31.Qf3 Kg7 32.e5 Qxa6 33.Qxf6+ Kxh6 34.g5+ Kh5 35.Qf3+ Kh4 36.h3 Rf8 37.Qe4+ Kh5 38.Kh2 Qc8
White is slightly better: +/= (0.64) Depth: 10/20 00:00:00
31.Qf3 Kg7 32.e5 Qxa6 33.Qxf6+ Kxh6 34.g5+ Kh5 35.Qf3+ Kh4 36.Qe4+ Kh5 37.h3 Rf8 38.Kh2 Rxf2+ 39.Kg3
White is slightly better: +/= (0.62) Depth: 11/20 00:00:00
31.Qf3 Kg7 32.e5 Qxa6 33.Qxf6+ Kxh6 34.g5+ Kh5 35.Qf3+ Kh4 36.Qe4+ Kh5 37.h3 Rf8 38.Kh2 Rxf2+ 39.Kg3
White is slightly better: +/= (0.62) Depth: 12/20 00:00:00
31.Qf3 Kg7 32.e5 Qxa6 33.Qxf6+ Kxh6 34.g5+ Kh5 35.Qf3+ Kh4 36.h3 Rf8 37.Qe4+ Kh5 38.Kh2 Rxf2+ 39.Kg3
White is slightly better: +/= (0.62) Depth: 13/20 00:00:00 114kN
31.Qf3 Kg7 32.e5 Qxa6 33.Qxf6+ Kxh6 34.g5+ Kh5 35.Qf3+ Kh4 36.h3 Rf8 37.Qe4+ Kh5 38.Kh2 Rxf2+ 39.Kg3
White is slightly better: +/= (0.62) Depth: 14/20 00:00:00 135kN
31.Qf3 Kg7 32.e5 Qxa6 33.Qxf6+ Kxh6 34.g5+ Kh5 35.Qf3+ Kh4 36.h3 Rf8 37.Qe4+ Kh5 38.Kh2 Qc8
White is better: +/- (0.83) Depth: 15/27 00:00:00 181kN
31.Qf3 Kg7 32.e5 Qxa6 33.Qxf6+ Kxh6 34.g5+ Kh5 35.Qf3+ Kh4 36.Qe4+ Kh5 37.h3 Rf8 38.Kh2 Rxf2+ 39.Kg3
White is slightly better: +/= (0.62) Depth: 16/27 00:00:00 359kN
31.Qf3 Kg7 32.e5 Qxa6 33.Qxf6+ Kxh6 34.g5+ Kh5 35.Qf3+ Kh4 36.Qe4+ Kh5 37.h3 Rf8 38.Kh2 Rxf2+ 39.Kg3
White is slightly better: +/= (0.62) Depth: 17/27 00:00:00 609kN
31.Qf3 Qd7 32.Qxf6+ Qg7 33.g5 Bc8 34.Rc6 Qxf6 35.gxf6 Rf8 36.f4 a5 37.Rxc7 Rxf6 38.f5 a4 39.Nf7+ Kg8 40.Nd6 Rf8 41.fxg6 hxg6 42.Rxg6+ Kh8 43.Rh6+ Kg8 44.Rhh7
White is clearly better: +- (2.45) Depth: 18/34 00:00:00 1305kN
31.Qf3 Qd7 32.Qxf6+ Qg7 33.g5 Bc8 34.Qxg7+ Kxg7 35.Ra5 Bh3 36.Rg3 Bc8 37.Kg2 Rf8 38.c6 Be6 39.Rf3 Rfd8 40.Kg3 Rd6 41.Ra6 Rf8 42.Rxf8
White is clearly better: +- (2.40) Depth: 19/34 00:00:00 3704kN, tb=29
31.Qf3 Qd7 32.Qxf6+ Qg7 33.g5 Qxf6 34.gxf6 Bc8 35.Rc6 Re6 36.f7 Kg7 37.Rxc7 Rf6 38.f8R+ Kxf8 39.f4 Ke8 40.f5 Bd7 41.Ng4 Rc6 42.Rb7
White has a decisive advantage: +- (2.71) Depth: 20/38 00:00:00 8810kN, tb=29
31.Qf3 Qd7 32.Qxf6+ Qg7 33.g5 Qxf6 34.gxf6 Bc8 35.Rc6 Re6 36.f7 Kg7 37.Rxc7 Kf8 38.f4 Re7 39.Rc6 Re6 40.Rd6 Rxd6 41.cxd6 Be6 42.f5 Bxf7 43.Rf1 Kg7 44.Nxf7 Kxf7 45.fxg6+ Kxg6 46.e5 Rd8 47.h4 a5 48.Ra1
White has a decisive advantage: +- (2.63) Depth: 21/38 00:00:00 10084kN, tb=29
31.Qf3 Qd7 32.Qxf6+ Qg7 33.g5 Bc8 34.Rc6 Qxf6 35.gxf6 Re6 36.f7 Kg7 37.Nf5+ Kf8 38.Rxc7 Rf6 39.Nd6 Bh3 40.e5 Rxf2 41.Ne4 Rxf7 42.Rxf7+ Kxf7 43.Ng5+ Ke7 44.Nxh3 a5 45.Nf4 Rc8 46.Ne2 Ke6 47.c6 a4 48.Rb1 a3
White has a decisive advantage: +- (3.03) Depth: 22/41 00:00:01 15521kN, tb=43
31.Qf3 Qd7 32.Qxf6+ Qg7 33.g5 Bc8 34.Rc6 Qxf6 35.gxf6 Re6 36.f7 Kg7 37.Rxc7 Rf6 38.f8Q+ Kxf8 39.f4 Be6 40.Rxh7 Re8 41.Rxa7 Re7 42.Ra5 Kg7 43.Ng4 Rxf4 44.Ne5 Rf2 45.Nc6 b3 46.Nxe7 bxc2
White has a decisive advantage: +- (3.26) Depth: 23/41 00:00:01 25400kN, tb=658

Fast and strong! :-))

Blue Marlin 15.4 is fast to, great!
Corchess needs 30s, cool.

Analysis by ShashChess 25.4:

30...Qe7
+/= (0.33 ++) Depth: 48/28 00:01:28 1696MN, tb=97715
White is slightly better

Shashchess 25.4 needs the same time like Stockfish dev.
Shashchess 25.4 GoldDigger is fast.

To the Stockfish developers: Please fix this! Thanks!
DrEinstein
Posts: 75
Joined: Wed Sep 15, 2021 8:50 pm
Full name: Albert Einstein

Re: EN-Test 2022 - new testsuite

Post by DrEinstein »

Eduard, what is the mean developing time of an engine made by you? Three to five days? I would do this calculation by myself, but I have no more idea how many engines you have developed since the beginning.
To be honest, I'm starting to think about if I should just copy your developing career.
Note, no quotation marks and no irony or sarcasm :)
But I was thinking about to write .... copy and paste your dev...
I guess sooner or later I will compile one of your (older) engines and try it out :D
DrEinstein
Posts: 75
Joined: Wed Sep 15, 2021 8:50 pm
Full name: Albert Einstein

Re: EN-Test 2022 - new testsuite

Post by DrEinstein »

Eduard wrote: Sun Nov 20, 2022 6:18 am ...
Shashchess 25.4 needs the same time like Stockfish dev.
Shashchess 25.4 GoldDigger is fast.

To the Stockfish developers: Please fix this! Thanks!
Eduard you assume that some SFdevs are reading here. Maybe they do.
However, I am sure that they are aware of the 'GoldDigger code' and probably even have written it in the past and fishtested it already. It probably was an ELO looser and thus wasn't applied to SFmaster and won't be applied even if you ask them.
Eduard
Posts: 1439
Joined: Sat Oct 27, 2018 12:58 am
Location: Germany
Full name: N.N.

Re: EN-Test 2022 - new testsuite

Post by Eduard »

If you're a chess connoisseur, which I'm assuming you are, then you'll have to admit that Stockfish does pretty bad math here. You like to talk about Elos. Which Elos? Bullet Elos or what?

An engine that takes 1:20 minutes in the position shown to see that Qb5 is a fatal error is unusable for my purposes. Defending such an engine what you do is a joke, sorry.

As for the many engine names, it's actually all the same engine. The Attack Engine was first, there was also a download including Sorce Code, and it was downloaded almost 200 times. I didn't like some of what the users had done, so I ended the Attack version. I do everything privately now and Deutronius only have 3 people who play with me on the server. These are good friends who also send me all the games. We often test day and night on the server. Sometimes I actually spend many hours tweaking. Yesterday I actually sat at the PC from morning to evening. I have a lot of fun. :-)
gordonr
Posts: 237
Joined: Thu Aug 06, 2009 8:04 pm
Location: UK

Re: EN-Test 2022 - new testsuite

Post by gordonr »

Eduard wrote: Sun Nov 20, 2022 6:08 am Stockfish dev and a new Stockfish-clone are BLIND!

Live from PlayChess.com:

[fen]r3r2k/p1pq3p/R3bppN/2P5/1p1pP1P1/3P4/2P2P1P/3Q2RK b - - 0 1[/fen]


Analysis by Stockfish dev-20221119-41c6a74d:

30...Qb5 31.Qf3 Kg7 32.Rxe6 Rxe6 33.Nf5+ Kf7 34.Qh3 Kg8 35.Nxd4 Qe8 36.Nxe6 Qxe6 37.Qe3 a5 38.Qd4 Rb8 39.Qd5 Qxd5 40.exd5 a4 41.Kg2 Rd8 42.Rb1 Rxd5 43.Rxb4 Rxc5 44.Rxa4 Rxc2 45.Kf3 Kf7 46.h4 Rc1 47.Ra5 Ke7 48.Ra7 Kd6 49.Ra6+ Ke7 50.Kf4 Rh1 51.Rc6 Rxh4 52.Rxc7+ Kd6 53.Ra7 h5 54.f3 Ke6 55.Ra6+ Ke7 56.Ra4
= (0.00) Depth: 44/51 00:00:20 393MN, tb=26327
The position is equal

Stockfish is blind, after 20s on 20 Threads -->0.00??

Analysis by Deutronius 201122:

30...Qe7 31.Qd2 Qxc5 32.h4 Re7 33.Rga1 Rf8 34.Rxa7 Kg7 35.Kg1 Qc3 36.Qf4 Bc8 37.Kg2 Qc5 38.Qd2 Qc3 39.Qxc3 dxc3 40.g5 fxg5 41.hxg5 b3 42.cxb3 Rd8 43.Ra8 Red7 44.d4 Rxd4 45.R8a7 Be6 46.R1a6 Rxe4 47.Rxc7+ Rd7 48.Rxc3 Re5 49.f4 Re4 50.Rb6
+/= (0.54 --) Depth: 32/40 00:00:10 181MN, tb=1294

My Engine Deutronius sees the loss already after 3s, and need to play very soon Qe7. White is better but is not won.
I think there is some luck in a small amount of runs. Here's one with a lucky 5s...

FEN: r3r2k/p1pq3p/R3bppN/2P5/1p1pP1P1/3P4/2P2P1P/3Q2RK b - - 0 1

Stockfish_19_Nov_2022:
Found 510 tablebases
NNUE evaluation using nn-ad9b42354671.nnue enabled

28/30 00:00 21,210k 35,291k 0.00 Qb5 Qf3 Kg7 Rxe6 Rxe6 Nf5+ Kf7 Nxd4 Qxc5 Nxe6 Kxe6 d4 Qxd4 Rd1 Qc4 e5 Rf8 exf6 Rxf6 Qa8 Rf7 Rd3 Kf6 Qh8+ Ke6
29/28 00:00 22,909k 35,462k 0.00 Qb5 Qf3 Kg7 Rxe6 Rxe6 Nf5+ Kf7 Nxd4 Qxc5 Nxe6 Kxe6 d4 Qxd4 Rd1 Qc4 e5 Rf8 exf6 Rxf6 Qa8 Rf7 Rd3 Kf6 Qh8+ Ke6
30/34 00:00 24,608k 35,356k 0.00 Qb5 Qf3 Kg7 Rxe6 Rxe6 Nf5+ Kf7 Nxd4 Qxc5 Nxe6 Kxe6 d4 Qxd4 Rd1 Qc4 e5 Rf8 exf6 Rxf6 Qa8 Rf7 Rd3 Kf6 Qh8+ Ke6
31/28 00:00 28,037k 35,445k 0.00 Qb5 Qf3 Kg7 Rxe6 Rxe6 Nf5+ Kf7 Nxd4 Qxc5 Nxe6 Kxe6 d4 Qxd4 Rd1 Qc4 e5 Rf8 exf6 Rxf6 Qa8 Rf7 Rd3 Kf6 Qh8+ Ke6
32/28 00:00 31,738k 35,581k 0.00 Qb5 Qf3 Kg7 Rxe6 Rxe6 Nf5+ Kf7 Nxd4 Qxc5 Nxe6 Kxe6 d4 Qxd4 Rd1 Qc4 e5 Rf8 exf6 Rxf6 Qa8 Rf7 Rd3 Kf6 Qh8+ Ke6 Qc8+ Kf6
33/32 00:01 37,093k 35,260k 0.00 Qb5 Qf3 Kg7 Rxe6 Rxe6 Nf5+ Kf7 Nxd4 Qxc5 Nxe6 Kxe6 d4 Qxd4 Rd1 Qc4 e5 Rf8 exf6 Rxf6 Qa8 Rf7 Rd3 Kf6 Qh8+ Ke6
34/39 00:01 40,245k 35,179k 0.00 Qb5 Qf3 Kg7 Rxe6 Rxe6 Nf5+ Kf7 Nxd4 Qxc5 Nxe6 Kxe6 d4 Qxd4 Rd1 Qc4 e5 Rf8 exf6 Rxf6 Qa8 Rf7 h4 Qe2 Qd5+ Ke7 Qd8+ Ke6
35/29 00:01 49,185k 35,410k 0.00 Qb5 Qf3 Kg7 Rxe6 Rxe6 Nf5+ Kf7 Nxd4 Qxc5 Nxe6 Kxe6 d4 Qxd4 Rd1 Qc4 e5 Rf8 exf6 Rxf6 Qa8 Rf7 h4 Qe2 Qd5+ Ke7 Qd8+ Ke6 Qd5+
36/28+ 00:03 106,197k 31,920k +0.44 Qb5 Qf3
36/28+ 00:04 148,349k 30,506k +0.60 Qb5 Qf3
36/55- 00:05 163,091k 30,168k +0.47 Qe7
36/55 00:25 726,514k 28,089k +0.47 Qe7
38/57 00:25 726,514k 28,089k +0.31 Qe7 Qd2 Qxc5 h4 c6 Rga1 Qd6 Kg1 Rf8 Rxa7 Rxa7 Rxa7 Qb8 Re7 Qd6 Rb7 c5 Ra7 Qb8 Ra6 Qc8 Ra1 Qb8 Kg2 Bd7 g5 f5 exf5 Bxf5 Nxf5 Rxf5 Qe2 Qb7+ f3 Rf7 Qe4 Qxe4 fxe4 c4 Kg3 cxd3 cxd3 Kg7 Rb1 Rb7 Kf4 Rb5 e5 Kf7 Ke4 Ke6 Kxd4 Rd5+ Ke3 Rxe5+ Kd4 Rb5


Another run that wasn't so lucky.... 19s this time...

42/30+ 00:05 196,808k 33,055k +0.06 Qb5 Qf3
42/30+ 00:09 299,083k 31,255k +0.44 Qb5 Qf3
42/61- 00:19 552,406k 29,062k +0.34 Qe7
42/78 00:44 1,233,543k 27,652k +0.22 Qe7 Qf3 Rf8 Qf4 Bd7 Rga1 b3 cxb3 Bb5 Rxa7 Bxd3 f3 Rxa7 Rxa7 Qxc5 Qxc7 Qxc7 Rxc7 Bb5 Nf7+ Kg8 Nd6 Rb8 Kg2 d3 Nxb5 Rxb5 Rc3 d2 Rd3 Rxb3 Rxd2 Kf7 h4 h6 Kg3 g5 h5 Kg8 Rd6 Kg7 Rd7+ Kg8 Kf2 Ra3 Ke2
43/52+ 00:45 1,248,648k 27,638k +0.31 Qe7 Qd2
43/52+ 00:45 1,267,997k 27,617k +0.33 Qe7 Qd2
43/52 00:46 1,280,572k 27,601k +0.33 Qe7 Qd2
41/63 00:46 1,280,572k 27,601k +0.48 Qe7 Qd2 Bc8 Ra5 Rb8 Rxa7 Rb5 c6 Rb6 Rb1 Rf8 Kg1 Kg7 Ra4 Qc5 Rbxb4 Rxb4 Rxb4 Qc3 Nf5+ gxf5 Qxc3 dxc3 gxf5 Kf7 Kg2 Ba6 Rb3 Ke7 Rxc3 Bb5 Kf3 Kd6 d4 Ra8 Kf4 Ra4 Rh3 Rxd4 Rh6 Ke7 f3 Rd6 Rxh7+ Kd8 h4 Rxc6