Hard talkchess 2016 set in PGN

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

Moderators: hgm, Rebel, chrisw

Vinvin
Posts: 5228
Joined: Thu Mar 09, 2006 9:40 am
Full name: Vincent Lejeune

Hard talkchess 2016 set in PGN

Post by Vinvin »

After months of work, here's the PGN file with most obvious variations, 158 positions/games: https://www.dropbox.com/s/4a879s438wv09 ... 2.pgn?dl=0

Thanks to everyone who helped to create this set !

To do :
Add score (some are draw)
Add info about score constraint (sometimes moves is obvious but score has to be clear win or draw or distance to mate)
Add reference (players, composer, sources, ...)
Add themes (Fortress/king attack/zugzwang/long mate threat/perpetual check/...)
Add some lines
Chose an order to put the positions (a simple alphabetical sort ?)
Release the EPD file

Any comments are welcome !

Vincent
Top
User avatar
Rebel
Posts: 6991
Joined: Thu Aug 18, 2011 12:04 pm

Re: Hard talkchess 2016 set in PGN

Post by Rebel »

Vinvin wrote:Add reference (players, composer, sources, ...)

Any comments are welcome !
You can use ProDeo 1.85 for that purpose. Paste the second position results in:

[pgn]
[Event "Switzerland"]
[Site "Switzerland"]
[Date "1980.??.??"]
[Round "1"]
[White "Gusia"]
[Black "Barlov, D."]
[Result "1-0"]
[ECO "B49"]
[EventDate "1980.??.??"]
[PlyCount "43"]

1.e4 c5 2.Nf3 e6 3.d4 cxd4 4.Nxd4 Nc6 5.Nc3 Qc7 6.Be2 a6 7.O-O Nf6 8.Be3
Bb4 9.Na4 Bd6 10.Nxc6 bxc6 11.Nb6 Rb8 12.Nxc8 Rxc8 13.f4 e5 14.Kh1 O-O 15.
f5 Nxe4 16.f6 g6 17.Qe1 Bc5 18.Bxc5 Nxc5 19.Qh4 e4 20.Qh6 Ne6 21.Bg4 d5
22.Rf5 1-0
[/pgn]
The forth position:

[pgn][Event "Dubai olm"]
[Site "Dubai olm"]
[Date "1986.11.??"]
[Round "1"]
[White "Georgiev, K."]
[Black "Ionescu, C."]
[Result "1-0"]
[WhiteElo "2445"]
[BlackElo "2495"]
[ECO "B50"]
[EventDate "1986.??.??"]
[PlyCount "105"]

1.e4 c5 2.Nf3 e6 3.b3 Nc6 4.Bb2 d6 5.d4 cxd4 6.Nxd4 a6 7.Be2 Bd7 8.c4 Nf6
9.Nc3 Be7 10.O-O O-O 11.Nc2 Rc8 12.f4 Qb6+ 13.Kh1 Rfd8 14.Bd3 Be8 15.Qe2
Qc7 16.Rae1 Bf8 17.Ne3 Ne7 18.Bb1 Qb8 19.e5 dxe5 20.fxe5 Nd7 21.Bxh7+ Kxh7
22.Ne4 Kg8 23.Ng5 g6 24.Ng4 Nf5 25.Rxf5 exf5 26.Qf2 Be7 27.Nf6+ Nxf6 28.
exf6 Bxf6 29.Qh4 Rd1 30.Qh7+ Kf8 31.Ba3+ Rd6 32.Bxd6+ Qxd6 33.Rxe8+ Kxe8
34.Qxf7+ Kd8 35.Ne6+ Qxe6 36.Qxe6 Rc6 37.Qg8+ Kc7 38.Qxg6 Rd6 39.g3 Rd1+
40.Kg2 Rd2+ 41.Kh3 Be7 42.a4 Rf2 43.Qf7 Kd7 44.Qd5+ Kc7 45.c5 Bd8 46.b4
Kc8 47.b5 axb5 48.axb5 Bc7 49.Qe6+ Kd8 50.c6 bxc6 51.bxc6 Rd2 52.Qg8+ Ke7
53.Qg5+ 1-0[/pgn]
Etc.
Top
Dann Corbit
Posts: 12540
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Hard talkchess 2016 set in PGN

Post by Dann Corbit »

These look to be the hardest for Stockfish:

4kr2/5p1K/3p1Q2/1p4P1/4P3/1PP5/7b/8 w - -
4q1kr/p6p/1prQPppB/4n3/4P3/2P5/PP2B2P/R5K1 w - -
8/3P3k/n2K3p/2p3n1/1b4N1/2p1p1P1/8/3B4 w - -
8/8/8/2pK3p/8/5N2/1P4pq/5Rbk w - -
2bq1b2/1p3k2/1r1p1p1p/pNpP1PpP/P1P3P1/6B1/1P5Q/2K1R3 w - -
8/p4pk1/r1p1p1p1/P1PpRnPp/1P3P2/2P5/4RK2/8 w - -
1rb2r1k/3p1pp1/p3p2p/1p2P3/4q3/1N2B1R1/PPP1N1P1/2KR1B2 w - -
1r6/4k3/r2p2p1/2pR1p1p/2P1pP1P/pPK1P1P1/P7/1B6 b - -
r2qnrk1/p4pp1/1p2b2p/2p1Pp2/2P5/PnPB1QN1/6PP/R1B2RK1 w - -
4bbk1/3nq2r/pr2p3/Np1p1p1p/1PpP1Pp1/2P1P1P1/PQ1NB1PR/2K4R b - -
8/1p2b3/6k1/5p2/p2NbP1p/P1B1P1p1/1P5P/6K1 w - -
7q/b1p5/1p1Npkb1/pPP2ppP/P1P5/3B2P1/5P1R/K3R3 w - -
2q1r1k1/b1r2pp1/p4np1/2np4/1Q6/4PP1P/P1B1N1P1/1RBR1K2 b - -
r4r1k/1p4qp/b1pNRbp1/2P5/3p1P1P/PQ4P1/3R2BK/8 w - -
q1r5/k3ppp1/2R5/PP1KP2r/2p4Q/2p1p3/5pP1/3R4 w - -
1nrqr3/1b4bk/pp3pp1/3pPp1p/P1pP1P1P/2P1BN2/1PB2QP1/R4RK1 w - -
2r1k2r/4qpp1/p3p2p/1pn1P3/3N2P1/P7/1PP3QP/1K1R2R1 w k -
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
Top
Paloma
Posts: 1167
Joined: Thu Dec 25, 2008 9:07 pm
Full name: Herbert L

Re: Hard talkchess 2016 set in PGN

Post by Paloma »

Thanks Vincent for present this suite.
.
Top
User avatar
AdminX
Posts: 6340
Joined: Mon Mar 13, 2006 2:34 pm
Location: Acworth, GA

Re: Hard talkchess 2016 set in PGN

Post by AdminX »

Dann Corbit wrote:These look to be the hardest for Stockfish:
4kr2/5p1K/3p1Q2/1p4P1/4P3/1PP5/7b/8 w - -
[d]4kr2/5p1K/3p1Q2/1p4P1/4P3/1PP5/7b/8 w - -
Image
"Good decisions come from experience, and experience comes from bad decisions."
__________________________________________________________________
Ted Summers
Top
User avatar
Eelco de Groot
Posts: 4565
Joined: Sun Mar 12, 2006 2:40 am
Full name:   

Re: Hard talkchess 2016 set in PGN

Post by Eelco de Groot »

AdminX wrote:
Dann Corbit wrote:These look to be the hardest for Stockfish:
4kr2/5p1K/3p1Q2/1p4P1/4P3/1PP5/7b/8 w - -
[d]4kr2/5p1K/3p1Q2/1p4P1/4P3/1PP5/7b/8 w - -
Image
This is by far the best result I had with a Rainbow Serpent:

[D]4kr2/5p1K/3p1Q2/1p4P1/4P3/1PP5/7b/8 w - -

Engine: Rainbow Serpent 20162803_035 MP (i7 3600, 4 threads, 512 MB)
Latest change: No movecount pruning in (no pieces for side to move) pawn endgames.
(Hack, but effect on bench is within two percent increase. Might help a little here)
by T. Romstad, M. Costalba, J. Kiiski, G. Linscott

28/41 0:01 0.00 1.g6 Be5 2.Qxe5+ dxe5 3.g7 Ke7
4.gxf8B+ Kxf8 5.c4 bxc4 6.bxc4 f5
7.exf5 e4 8.c5 e3 9.c6 Ke7 10.Kg6 e2
11.f6+ Ke6 12.c7 Kd7 13.f7 e1Q
14.f8Q (12.312.596) 6619

29/41 0:02 0.00 1.g6 Be5 2.Qxe5+ dxe5 3.g7 Ke7
4.gxf8B+ Kxf8 5.c4 bxc4 6.bxc4 f5
7.exf5 e4 8.c5 e3 9.c6 Ke7 10.Kg6 e2
11.f6+ Ke6 12.c7 Kd7 13.f7 e1Q
14.f8Q (13.679.048) 6624
.
.
.
38/52 0:43 0.00 1.g6 Be5 2.Qxe5+ dxe5 3.g7 Ke7
4.gxf8B+ Kxf8 5.c4 bxc4 6.bxc4 f5
7.exf5 e4 8.c5 e3 9.c6 Ke7 10.c7 Kd7
11.f6 e2 12.f7 e1Q 13.c8Q+ Kxc8
14.f8Q+ (266.543.814) 6142

39/53 1:08 0.00 1.g6 Be5 2.Qxe5+ dxe5 3.g7 Ke7
4.gxf8B+ Kxf8 5.c4 bxc4 6.bxc4 f5
7.exf5 e4 8.c5 e3 9.c6 Ke7 10.c7 Kd7
11.f6 e2 12.f7 e1Q 13.c8Q+ Kxc8
14.f8Q+ (416.800.099) 6111

40/53 1:50 +0.03++ 1.Kh6 (643.835.736) 5817

40/53 1:52 +0.06++ 1.Kh6 (651.463.625) 5808

40/53 1:53 +0.12++ 1.Kh6 (657.291.454) 5806

40/53 1:55 +0.24++ 1.Kh6 (667.369.591) 5802

40/53 1:55 +0.49++ 1.Kh6 (671.474.902) 5800

40/53 2:02 +0.99++ 1.Kh6 (706.101.130) 5769

40/53 2:31 +1.98++ 1.Kh6 (855.228.843) 5661

40/53 19:17 +3.96++ 1.Kh6 (5.843.981.299) 5048

Just luck possibly. Amazingly, if 1.Kh6 Be5 2. Kg7 it seems to be already a Mate in 25 (possibly less?) I think this was a different Rainbow version and after searching the Rootposition. Not entirely sure about the _016.

[D]4kr2/5pK1/3p1Q2/1p2b1P1/4P3/1PP5/8/8 b - -

Engine: Rainbow Serpent 20162803_016 MP (512 MB)
by T. Romstad, M. Costalba, J. Kiiski, G. Linscott

25/35 0:02 +M32 2...Bg3 3.c4 b4 4.c5 dxc5 5.Qc6+ Ke7
6.Qb7+ Ke6 7.Kxf8 Ke5 8.Kxf7 c4 9.g6 Kf4
10.g7 c3 11.g8Q Be1 12.Qd5 Bh4
13.Qf5+ Ke3 14.Qh3+ Kf4 15.Qxh4+ Ke3 (13.482.554) 6408
.
.
.
43/51 8:56 +M24 2...Bg3 3.c4 bxc4 4.e5 Bxe5 5.bxc4 Bg3
6.c5 Be5 7.cxd6 Bxf6+ 8.gxf6 Rh8
9.Kxh8 Kd7 10.Kg7 Ke8 11.d7+ Kxd7
12.Kxf7 Kd6 13.Ke8 Ke5 14.f7 Kf4
15.f8Q+ Ke4 (3.528.747.257) 6571

46/57 14:28 +M25 2...Bh2 3.c4 bxc4 4.e5 Bxe5 5.bxc4 Bg3
6.c5 Be5 7.cxd6 Bxf6+ 8.gxf6 Rh8
9.d7+ Kd8 10.Kxh8 Kxd7 11.Kg8 Ke6
12.Kg7 Ke5 13.Kxf7 Kd6 14.Ke8 Ke5
15.f7 Ke4 (5.755.120.716) 6628

best move: Be5-h2 time: 14:28.281 min n/s: 6.628.185 nodes: 5.755.120.716
Debugging is twice as hard as writing the code in the first
place. Therefore, if you write the code as cleverly as possible, you
are, by definition, not smart enough to debug it.
-- Brian W. Kernighan
Top
kgburcham
Posts: 2016
Joined: Sun Feb 17, 2008 4:19 pm

Re: Hard talkchess 2016 set in PGN

Post by kgburcham »

[D] 4kr2/5pK1/3p1Q2/1p2b1P1/4P3/1PP5/8/8 b - -


Engine: Komodo 10 64-bit (8192 MB)
by Don Dailey, Larry Kaufman, Mark Lefler
Threads now set to 8
39.00 0:02 +0.07-- 1...Bg3 2.c4 (146.580.681) 52116
39.01 0:06 +0.14-- 1...Bg3 2.c4 (352.887.384) 55626
39.01 0:09 +0.24-- 1...Bg3 2.c4 (577.630.147) 58400
39.01 0:12 +0.39-- 1...Bg3 2.c4 (769.768.141) 59859
39.01 0:14 +0.61-- 1...Bg3 2.c4 (896.959.968) 60362
39.01 0:17 +0.94-- 1...Bg3 2.c4 (1.107.270.029) 61567
39.01 0:21 +1.43-- 1...Bg3 2.c4 (1.349.476.056) 62044
39.01 0:22 +2.16-- 1...Bg3 2.c4 (1.419.234.288) 61999
39.01 0:23 +3.25-- 1...Bg3 2.c4 (1.462.277.464) 61326
39.01 0:25 +4.88-- 1...Bg3 2.c4 (1.536.353.496) 59375
39.01 1:10 +M25 1...Bg3 2.c4 bxc4 3.e5 Bxe5 4.bxc4 Bg3 (2.692.003.859) 38294
40.01 4:05 +M23 1...Bg3 2.c4 bxc4 3.e5 Bxe5 4.bxc4 Bg3 (7.244.141.851) 29537
41.01 4:21 +M23 1...Bg3 2.c4 bxc4 3.e5 Bxe5 4.bxc4 Bg3 (7.679.732.174) 29353
42.01 5:00 +M23 1...Bg3 2.c4 bxc4 3.e5 Bxe5 4.bxc4 Bg3 (8.732.814.872) 29066
43.01 6:24 +M21 1...Bg3 2.c4 bxc4 3.e5 Bxe5 4.bxc4 Bg3 5.c5 Be5 6.cxd6 Bxf6+ 7.gxf6 Rh8 8.Kxh8 Kd7 9.Kg7 Ke6 10.d7 Kxd7 11.Kxf7 Kd6 12.Kg7 Kd5 13.f7 Kc4 14.f8Q Kc3 (11.094.886.785) 28831
44.01 6:59 +M21 1...Bg3 2.c4 bxc4 3.e5 Bxe5 4.bxc4 Bg3 5.c5 Be5 6.cxd6 Bxf6+ 7.gxf6 Rh8 8.Kxh8 Kd7 9.Kg7 Ke6 10.d7 Kxd7 11.Kxf7 Kd6 12.Kg7 Kd5 13.f7 Kc4 14.f8Q Kc3 (12.077.633.775) 28815
45.01 9:11 +M21 1...Bg3 2.c4 bxc4 3.e5 Bxe5 4.bxc4 Bg3 5.c5 Be5 6.cxd6 Bxf6+ 7.gxf6 Rh8 8.Kxh8 Kd7 9.Kg7 Ke6 10.d7 Kxd7 11.Kxf7 Kd6 12.Kg7 Kd5 13.f7 Kc4 14.f8Q Kc3 (15.709.221.676) 28471
no chess program was born totally from one mind. all chess programs have many ideas from many minds.
Top
yanquis1972
Posts: 1766
Joined: Wed Jun 03, 2009 12:14 am

Re: Hard talkchess 2016 set in PGN

Post by yanquis1972 »

[d]4kr2/5p1K/3p1Q2/1p4P1/4P3/1PP5/7b/8 w - - 0 1

Analysis by Stockfish 210516 64 BMI2:

1.Kh6
+- (1.68 ++) Depth: 73/55 00:06:27 5860MN, tb=26870198
(, 23.05.2016)


lot of long hangs with this position, whether i use TBs or not. (not sure if a caveat for these positions is disabling TB-assistance..?)
Top
User avatar
Eelco de Groot
Posts: 4565
Joined: Sun Mar 12, 2006 2:40 am
Full name:   

Re: Hard talkchess 2016 set in PGN

Post by Eelco de Groot »

Thanks for checking this Kim!

Black presumably lasts a bit longer by the way if he plays Rg8. This was the end of my search in the morning:


[D]4kr2/5p1K/3p1Q2/1p4P1/4P3/1PP5/7b/8 w - -
.
.
.
40/53 19:17 +3.96++ 1.Kh6 (5.843.981.299) 5048

40/94 186:38 +5.04 1.Kh6 Rg8 2.Kh5 Be5 3.Qf1 Rh8+ 4.Kg4 Rh2
5.Qxb5+ Kf8 6.Qb6 Kg7 7.b4 Re2 8.Kf5 Bxc3
9.b5 Be5 10.Qa7 Re1 11.b6 Rf1+
12.Kg4 Rg1+ 13.Kf3 Bd4 14.Qc7 (57.901.637.059) 5170

41/94 237:39 +5.07++ 1.Kh6 (72.737.017.511) 5100

41/94 243:53 +5.10++ 1.Kh6 (74.528.084.266) 5092

41/94 254:42 +5.16++ 1.Kh6 (77.683.258.737) 5083

41/94 260:22 +5.29++ 1.Kh6 (79.330.211.172) 5077

41/94 279:23 +5.53++ 1.Kh6 (84.880.977.750) 5063

41/94 299:53 +5.66 1.Kh6 Rg8 2.Qf5 Rg6+ 3.Kh5 Be5
4.Qc8+ Ke7 5.Qb8 Rg7 6.Qxb5 Rh7+
7.Kg4 Rh2 8.Qb7+ Kf8 9.Qb6 Kg7 10.b4 Re2
11.Kf5 Bxc3 12.b5 Be5 13.g6 fxg6+
14.Ke6 (90.960.982.208) 5055

best move: Kh7-h6 time: 372:52.110 min n/s: 5.055.140 nodes: 112.753.256.412
Debugging is twice as hard as writing the code in the first
place. Therefore, if you write the code as cleverly as possible, you
are, by definition, not smart enough to debug it.
-- Brian W. Kernighan
Top
Vinvin
Posts: 5228
Joined: Thu Mar 09, 2006 9:40 am
Full name: Vincent Lejeune

Re: Hard talkchess 2016 set in PGN

Post by Vinvin »

Vinvin wrote:After months of work, here's the PGN file with most obvious variations, 158 positions/games: https://www.dropbox.com/s/4a879s438wv09 ... 2.pgn?dl=0

Thanks to everyone who helped to create this set !

To do :
Add score (some are draw)
Add info about score constraint (sometimes moves is obvious but score has to be clear win or draw or distance to mate)
Add reference (players, composer, sources, ...)
Add themes (Fortress/king attack/zugzwang/long mate threat/perpetual check/...)
Add some lines
Chose an order to put the positions (a simple alphabetical sort ?)
Release the EPD file

Any comments are welcome !

Vincent
Little update here : http://home.scarlet.be/vincentlejeune/h ... nbeta3.pgn

Correction fir this one :
[Event "?"]
[Site "?"]
[White "W"]
[Black "B"]
[Result "*"]
[FEN "b3r3/q2B2k1/3Q2p1/1p5p/3pP3/5P2/1p4PP/5RK1 b - - 0 1"]
[Comment "bm d3+; c1 "diff=6";"]

1. ... d3+ 2. Kh1 Qf2 3. Rd1 d2 4. h3 Qe1+ 5. Kh2 Rf8 6. Bf5 gxf5 *

Remove this one : 6Nk/pp2Np1p/2p2Pp1/2R2bP1/7K/P7/1q1n1Q2/5n2 w - - 0 1

Add this one : 1b6/2p5/1kB5/1PN5/8/3K4/8/q7 w - - bm Nb3;
Top

Return to “Computer Chess Club: General Topics”