Hard-Talkchess-2020 set, final release

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

Re: Hard-Talkchess-2020 set, final release

Post by Vinvin »

2 new results :

Code: Select all

Bluefish-XI(Tact=2,defens=no) : 181 pos (671 sec) New number 1 !!!!
houdidit603(default) :          125 pos (1692 sec)
Current top 10

Code: Select all

 1) Bluefish-XI(Tact=2,defens=no) : 181 pos (671 sec)
 2) Eman 4.30   :                   163 pos (1087 sec)
 3) Bluefish-XI :                   159 pos (1078 sec)
 4) CorChess v6                     156 pos (1203 sec)
 5) houdidit603-tact=1 :            155 pos (1123 sec)
 6) Stockfish 11 :                  141 pos (1459 sec)
 7) Honey-XI :                      139 pos (1440 sec)
 8) Stockfish 10 :                  137 pos (1574 sec)
 9) Crystal 2020-05-02 :            136 pos (1421 sec)
10) Black-Diamond-XI :              136 pos (1472 sec)
I reworked the results sheet, I removed the moves to hold only the time to solution : http://home.scarlet.be/vincentlejeune/c ... 2020v2.ods

Vinvin wrote: ↑Tue May 12, 2020 1:00 am Current Top 5 :

Code: Select all

1) Eman 4.30 :          163 pos (1087 sec)
2) Bluefish-XI :        159 pos (1078 sec)
3) CorChess v6 :        156 pos (1203 sec)
4) houdidit603(tact=1): 155 pos (1123 sec)
5) Stockfish 11 :       141 pos (1459 sec)
Table in a textfile (only results>100) : http://home.scarlet.be/vincentlejeune/c ... 2020-3.txt
Vinvin wrote: ↑Tue May 12, 2020 12:42 am 2 more engines : Eman 4.30 (empty learn file + read only) and Houdidit 6.03(tactical=1) with very good results :

Code: Select all

Engine :          number found ; avg time; avg time+penalty

Eman 4.30:             163 pos ; 316 sec ; 1087 sec
Houdidit 6.03(tact=1)  155 pos ; 196 sec ; 1123 sec
The sheet updated : http://home.scarlet.be/vincentlejeune/c ... s-2020.ods
Top
Vinvin
Posts: 5228
Joined: Thu Mar 09, 2006 9:40 am
Full name: Vincent Lejeune

Re: Hard-Talkchess-2020 set, final release

Post by Vinvin »

Dann Corbit wrote: ↑Thu May 14, 2020 7:20 pm I am giving the vvhard set of 11 problems 63 threads at 2 hours just to see if blue can do it.
Found by 9 engines ->"Hard-Talkchess-2020.015";bm h4;
Never Found ->"Hard-Talkchess-2020.023";bm Nh4;
Found by 4 engines ->"Hard-Talkchess-2020.026";bm Nf4;
Found by 2 engines ->"Hard-Talkchess-2020.033";bm Nd6+;
Found by 5 engines ->"Hard-Talkchess-2020.077";bm Na5;
Found by 10 engines ->"Hard-Talkchess-2020.090";bm Bf8+;
Found by Sting only ->"Hard-Talkchess-2020.102";bm hxg6;
Found by 2 engines ->"Hard-Talkchess-2020.131";bm Nc7;
Found by 12 engines ->"Hard-Talkchess-2020.135";bm Bc7;
Found by 7 engines (sometimes by luck ?) ->"Hard-Talkchess-2020.138";bm Qd4+;
Found by 10 engines ->"Hard-Talkchess-2020.176";bm d6;
Top
Dann Corbit
Posts: 12540
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Hard-Talkchess-2020 set, final release

Post by Dann Corbit »

I guess that for the positions where blue has problems it is depth, since it takes a long time to go deep.
For these last 11 only 3 were solved (c9 shows outcome of 2 hour search with 63 threads):
1r3rk1/3bb1pp/1qn1p3/3pP3/3P1N2/2Q2N2/2P3PP/R1BR3K w - - bm h4; id "Hard-Talkchess-2020.015"; c9 h4;
1rb1qrk1/2b2pp1/p3pBn1/3pP1Pp/1ppP4/2P1QN2/PP3P1P/R2BR1K1 w - - bm Nh4; id "Hard-Talkchess-2020.023"; c9 Nh4;
1rb2r1k/3p1pp1/p3p2p/1p2P3/4q3/1N2B1R1/PPP1N1P1/2KR1B2 w - - bm Nf4; id "Hard-Talkchess-2020.026"; c9 Nd2;
2b1r3/r2ppN2/8/1p1p1k2/pP1P4/2P3R1/PP3PP1/2K5 w - - bm Nd6+; id "Hard-Talkchess-2020.033"; c9 Nh6+;
4q1kn/2Bp1p2/1N1PpPp1/1P2P1P1/2N3b1/6p1/1P4Pb/7K w - - bm Na5; id "Hard-Talkchess-2020.077"; c9 b3;
6R1/8/2pB3k/2P4p/5p1q/5P2/4P1K1/8 w - - bm Bf8+; id "Hard-Talkchess-2020.090"; c9 Bf8+;
7q/b1p5/1p1Npkb1/pPP2ppP/P1P5/3B2P1/5P1R/K3R3 w - - bm hxg6; id "Hard-Talkchess-2020.102"; c9 Nb7;
Nq3n1k/8/P1p2Npp/2Kp3R/pp6/n3p3/8/4Q3 w - - bm Nc7; id "Hard-Talkchess-2020.131"; c9 Rxh6+;
n1QBq1k1/5p1p/5KP1/p7/8/8/8/8 w - - bm Bc7; id "Hard-Talkchess-2020.135"; c9 gxh7+;
q1r5/k3ppp1/2R5/PP1KP2r/2p4Q/2p1p3/5pP1/3R4 w - - bm Qd4+; id "Hard-Talkchess-2020.138"; c9 Qxe7+;
r2qr1k1/pp1bbp2/n5p1/2pPp2p/8/P2PP1PP/1P2N1BK/R1BQ1R2 w - - bm d6; id "Hard-Talkchess-2020.176"; c9 Ng1;

To see if the issue is depth, I am running the problems at one hour using bleeding edge SF to see if depth solves the trouble makers.
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
Dann Corbit
Posts: 12540
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Hard-Talkchess-2020 set, final release

Post by Dann Corbit »

Here is some of the log from the never found one:

Code: Select all

rb1qrk1/2b2pp1/p3pBn1/3pP1Pp/1ppP4/2P1QN2/PP3P1P/R2BR1K1 w - - acd 35; ce 101; acs 3167; acn 216428000; pv h4 bxc3 bxc3 Kh7 Nh2 Rh8 Bxh5 Kg8 Bxg6 fxg6 Kg2 Qf7 Nf3 Bd7 Qc1 Rb5 h5 Rxh5 Rh1 Rb8 Qd2 Ba4 Rag1 Kf8 Qc1 Ke8 Rxh5 gxh5 Nh4 Kd7 g6 Qf8 Bg5 Bd8 Bxd8 Kxd8 f4 Kc7 Kf3 Bd7 Rg5 a5 Ng2 Bc6 f5 exf5 ; bm h4;
1rb1qrk1/2b2pp1/p3pBn1/3pP1Pp/1ppP4/2P1QN2/PP3P1P/R2BR1K1 w - - acd 35; ce 101; acs 3458; acn 236194000; pv h4 bxc3 bxc3 Kh7 Nh2 Rh8 Bxh5 Kg8 Bxg6 fxg6 Kg2 Qf7 Nf3 Bd7 Qc1 Rb5 h5 Rxh5 Rh1 Rb8 Qd2 Ba4 Rag1 Kf8 Qc1 Ke8 Rxh5 gxh5 Nh4 Kd7 g6 Qf8 Bg5 Bd8 Bxd8 Kxd8 f4 Kc7 Kf3 Bd7 Rg5 a5 Ng2 Bc6 f5 exf5 ; bm h4;
1rb1qrk1/2b2pp1/p3pBn1/3pP1Pp/1ppP4/2P1QN2/PP3P1P/R2BR1K1 w - - acd 36; ce 163; acs 3960; acn 273136000; pv Nh4 Kh7 Bc2 Rh8 cxb4 Rxb4 Qd2 Rb7 Re3 Bd7 Qe2 a5 b3 cxb3 axb3 Bb5 Qd1 Bb6 Nxg6 fxg6 Qxh5+ Kg8 Qg4 gxf6 gxf6 Rh6 Rg3 Rbh7 Rc1 Kf8 Bxg6 Rxg6 Qxg6 Qxg6 Rxg6 Bd7 Rg4 Rh8 Rc3 Rg8 Rxg8+ Kxg8 Rg3+ Kf8 Rg4 ; bm Nh4;
1rb1qrk1/2b2pp1/p3pBn1/3pP1Pp/1ppP4/2P1QN2/PP3P1P/R2BR1K1 w - - acd 36; ce 163; acs 4020; acn 277595000; pv Nh4 Kh7 Bc2 Rh8 cxb4 Rxb4 Qd2 Rb7 Re3 Bd7 Qe2 a5 b3 cxb3 axb3 Bb5 Qd1 Bb6 Nxg6 fxg6 Qxh5+ Kg8 Qg4 gxf6 gxf6 Rh6 Rg3 Rbh7 Rc1 Kf8 Bxg6 Rxg6 Qxg6 Qxg6 Rxg6 Bd7 Rg4 Rh8 Rc3 Rg8 Rxg8+ Kxg8 Rg3+ Kf8 Rg4 ; bm Nh4;
1rb1qrk1/2b2pp1/p3pBn1/3pP1Pp/1ppP4/2P1QN2/PP3P1P/R2BR1K1 w - - acd 36; ce 163; acs 4080; acn 279354000; pv Nh4 Kh7 Bc2 Rh8 cxb4 Rxb4 Qd2 Rb7 Re3 Bd7 Qe2 a5 b3 cxb3 axb3 Bb5 Qd1 Bb6 Nxg6 fxg6 Qxh5+ Kg8 Qg4 gxf6 gxf6 Rh6 Rg3 Rbh7 Rc1 Kf8 Bxg6 Rxg6 Qxg6 Qxg6 Rxg6 Bd7 Rg4 Rh8 Rc3 Rg8 Rxg8+ Kxg8 Rg3+ Kf8 Rg4 ; bm Nh4;
1rb1qrk1/2b2pp1/p3pBn1/3pP1Pp/1ppP4/2P1QN2/PP3P1P/R2BR1K1 w - - acd 36; ce 163; acs 4140; acn 283481000; pv Nh4 Kh7 Bc2 Rh8 cxb4 Rxb4 Qd2 Rb7 Re3 Bd7 Qe2 a5 b3 cxb3 axb3 Bb5 Qd1 Bb6 Nxg6 fxg6 Qxh5+ Kg8 Qg4 gxf6 gxf6 Rh6 Rg3 Rbh7 Rc1 Kf8 Bxg6 Rxg6 Qxg6 Qxg6 Rxg6 Bd7 Rg4 Rh8 Rc3 Rg8 Rxg8+ Kxg8 Rg3+ Kf8 Rg4 ; bm Nh4;
1rb1qrk1/2b2pp1/p3pBn1/3pP1Pp/1ppP4/2P1QN2/PP3P1P/R2BR1K1 w - - acd 37; ce 148; acs 4500; acn 307895000; pv Nh4 Kh7 Bc2 Rh8 cxb4 Rxb4 Qd2 a5 Re3 Bd7 Rf3 Bb6 a3 Rb5 Rb1 a4 Nxg6 fxg6 Bxa4 Ra5 Bc2 Ra7 b3 Rxa3 Rg3 c3 Qd1 Kg8 h4 Bb5 Rxc3 Ba6 Qc1 Ra2 Qd2 Rh7 b4 Bb5 Rc1 Bc4 Rg3 Bb5 Kh2 Ba4 Bxa4 Rxa4 Rgc3 Ra8 Rc6 gxf6 Rxb6 fxe5 dxe5 ; bm Nh4;
1rb1qrk1/2b2pp1/p3pBn1/3pP1Pp/1ppP4/2P1QN2/PP3P1P/R2BR1K1 w - - acd 37; ce 148; acs 4800; acn 330660000; pv Nh4 Kh7 Bc2 Rh8 cxb4 Rxb4 Qd2 a5 Re3 Bd7 Rf3 Bb6 a3 Rb5 Rb1 a4 Nxg6 fxg6 Bxa4 Ra5 Bc2 Ra7 b3 Rxa3 Rg3 c3 Qd1 Kg8 h4 Bb5 Rxc3 Ba6 Qc1 Ra2 Qd2 Rh7 b4 Bb5 Rc1 Bc4 Rg3 Bb5 Kh2 Ba4 Bxa4 Rxa4 Rgc3 Ra8 Rc6 gxf6 Rxb6 fxe5 dxe5 ; bm Nh4;
1rb1qrk1/2b2pp1/p3pBn1/3pP1Pp/1ppP4/2P1QN2/PP3P1P/R2BR1K1 w - - acd 37; ce 148; acs 4920; acn 339174000; pv Nh4 Kh7 Bc2 Rh8 cxb4 Rxb4 Qd2 a5 Re3 Bd7 Rf3 Bb6 a3 Rb5 Rb1 a4 Nxg6 fxg6 Bxa4 Ra5 Bc2 Ra7 b3 Rxa3 Rg3 c3 Qd1 Kg8 h4 Bb5 Rxc3 Ba6 Qc1 Ra2 Qd2 Rh7 b4 Bb5 Rc1 Bc4 Rg3 Bb5 Kh2 Ba4 Bxa4 Rxa4 Rgc3 Ra8 Rc6 gxf6 Rxb6 fxe5 dxe5 ; bm Nh4;
1rb1qrk1/2b2pp1/p3pBn1/3pP1Pp/1ppP4/2P1QN2/PP3P1P/R2BR1K1 w - - acd 37; ce 148; acs 5040; acn 348077000; pv Nh4 Kh7 Bc2 Rh8 cxb4 Rxb4 Qd2 a5 Re3 Bd7 Rf3 Bb6 a3 Rb5 Rb1 a4 Nxg6 fxg6 Bxa4 Ra5 Bc2 Ra7 b3 Rxa3 Rg3 c3 Qd1 Kg8 h4 Bb5 Rxc3 Ba6 Qc1 Ra2 Qd2 Rh7 b4 Bb5 Rc1 Bc4 Rg3 Bb5 Kh2 Ba4 Bxa4 Rxa4 Rgc3 Ra8 Rc6 gxf6 Rxb6 fxe5 dxe5 ; bm Nh4;
1rb1qrk1/2b2pp1/p3pBn1/3pP1Pp/1ppP4/2P1QN2/PP3P1P/R2BR1K1 w - - acd 38; ce 142; acs 6000; acn 411410000; pv Nh4 Kh7 Bc2 Bd7 cxb4 a5 Qd2 axb4 Re3 Rh8 Rh3 Ra8 Qxb4 Kg8 Nxg6 fxg6 Qc5 Bb8 Ra3 Rxa3 Qxa3 Kf7 h4 gxf6 gxf6 Qf8 Qf3 Bc7 Kh2 Ke8 Rg1 Kd8 Rxg6 Kc8 Rg7 Qb4 f7 Rf8 Rg8 Bd8 Qxh5 Kc7 Rxf8 Qxf8 a3 ; bm Nh4;
1rb1qrk1/2b2pp1/p3pBn1/3pP1Pp/1ppP4/2P1QN2/PP3P1P/R2BR1K1 w - - acd 38; ce 142; acs 6300; acn 431930000; pv Nh4 Kh7 Bc2 Bd7 cxb4 a5 Qd2 axb4 Re3 Rh8 Rh3 Ra8 Qxb4 Kg8 Nxg6 fxg6 Qc5 Bb8 Ra3 Rxa3 Qxa3 Kf7 h4 gxf6 gxf6 Qf8 Qf3 Bc7 Kh2 Ke8 Rg1 Kd8 Rxg6 Kc8 Rg7 Qb4 f7 Rf8 Rg8 Bd8 Qxh5 Kc7 Rxf8 Qxf8 a3 ; bm Nh4;
1rb1qrk1/2b2pp1/p3pBn1/3pP1Pp/1ppP4/2P1QN2/PP3P1P/R2BR1K1 w - - acd 38; ce 142; acs 6660; acn 456027000; pv Nh4 Kh7 Bc2 Bd7 cxb4 a5 Qd2 axb4 Re3 Rh8 Rh3 Ra8 Qxb4 Kg8 Nxg6 fxg6 Qc5 Bb8 Ra3 Rxa3 Qxa3 Kf7 h4 gxf6 gxf6 Qf8 Qf3 Bc7 Kh2 Ke8 Rg1 Kd8 Rxg6 Kc8 Rg7 Qb4 f7 Rf8 Rg8 Bd8 Qxh5 Kc7 Rxf8 Qxf8 a3 ; bm Nh4;
1rb1qrk1/2b2pp1/p3pBn1/3pP1Pp/1ppP4/2P1QN2/PP3P1P/R2BR1K1 w - - acd 38; ce 142; acs 7200; acn 490800000; pv Nh4 Kh7 Bc2 Bd7 cxb4 a5 Qd2 axb4 Re3 Rh8 Rh3 Ra8 Qxb4 Kg8 Nxg6 fxg6 Qc5 Bb8 Ra3 Rxa3 Qxa3 Kf7 h4 gxf6 gxf6 Qf8 Qf3 Bc7 Kh2 Ke8 Rg1 Kd8 Rxg6 Kc8 Rg7 Qb4 f7 Rf8 Rg8 Bd8 Qxh5 Kc7 Rxf8 Qxf8 a3 ; bm Nh4;
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
Vinvin
Posts: 5228
Joined: Thu Mar 09, 2006 9:40 am
Full name: Vincent Lejeune

Re: Hard-Talkchess-2020 set, final release

Post by Vinvin »

Some comments about the 8 positions :

Only Sting-SF finds this 2 fortresses and trapped piece :
2b1r3/r2ppN2/8/1p1p1k2/pP1P4/2P3R1/PP3PP1/2K5 w - - bm Nd6+; id "Hard-Talkchess-2020.033"; c9 Nh6+;
7q/b1p5/1p1Npkb1/pPP2ppP/P1P5/3B2P1/5P1R/K3R3 w - - bm hxg6; id "Hard-Talkchess-2020.102"; c9 Nb7;

Pure tactics :
1rb2r1k/3p1pp1/p3p2p/1p2P3/4q3/1N2B1R1/PPP1N1P1/2KR1B2 w - - bm Nf4; id "Hard-Talkchess-2020.026"; c9 Nd2;
q1r5/k3ppp1/2R5/PP1KP2r/2p4Q/2p1p3/5pP1/3R4 w - - bm Qd4+; id "Hard-Talkchess-2020.138"; c9 Qxe7+;

Queen trapped :
4q1kn/2Bp1p2/1N1PpPp1/1P2P1P1/2N3b1/6p1/1P4Pb/7K w - - bm Na5; id "Hard-Talkchess-2020.077"; c9 b3;

Only CorChessV6 finds it in a reasonable time :
Nq3n1k/8/P1p2Npp/2Kp3R/pp6/n3p3/8/4Q3 w - - bm Nc7; id "Hard-Talkchess-2020.131"; c9 Rxh6+;

A lot of engines find this one (maybe a null move problem ?) :
n1QBq1k1/5p1p/5KP1/p7/8/8/8/8 w - - bm Bc7; id "Hard-Talkchess-2020.135"; c9 gxh7+;

Very tactical CorChessV6 like it fast. May be Bluefish-FD-XI is right and this move is over evaluated ? May be not ?
r2qr1k1/pp1bbp2/n5p1/2pPp2p/8/P2PP1PP/1P2N1BK/R1BQ1R2 w - - bm d6; id "Hard-Talkchess-2020.176"; c9 Ng1;
Top
zullil
Posts: 6442
Joined: Tue Jan 09, 2007 12:31 am
Location: PA USA
Full name: Louis Zulli

Re: Hard-Talkchess-2020 set, final release

Post by zullil »

Vinvin wrote: ↑Sat May 16, 2020 3:42 am Some comments about the 8 positions :

Very tactical CorChessV6 like it fast. May be Bluefish-FD-XI is right and this move is over evaluated ? May be not ?
r2qr1k1/pp1bbp2/n5p1/2pPp2p/8/P2PP1PP/1P2N1BK/R1BQ1R2 w - - bm d6; id "Hard-Talkchess-2020.176"; c9 Ng1;
Stockfish-dev seems to agree that d6 is "best":

info depth 55 seldepth 71 multipv 1 score cp 146 nodes 179584459114 nps 30865499 hashfull 1000 tbhits 0 time 5818291 pv d5d6 e7d6 g2b7 a6c7 b7a8 d8a8 e3e4 a8b7 e2c3 d7e6 d1e2 d6e7 c1e3 e8b8 a1b1 b7a6 b1d1 a6c6 d1d2 b8d8 e2f3 a7a5 g3g4 h5h4 f3d1 c6b7 d1e2 b7c6 f1f3 a5a4 e2e1 c7b5 c3b5 c6b5 e3f2 g6g5 e1e2 f7f6 d2c2 g8g7 f2g1 e6f7 c2c3 d8b8 f3f2 b8b7 e2c2 b5a5 h2g2 b7b5 f2d2 f7e6 g1e3 e6b3 c2c1 a5b6 c1g1 b6c6 g1f2 c6b6 g2g1 b6d6 g1h2 d6c6 f2f3 c6b6 d2f2 b3e6 f3e2 b6d8

info depth 55 seldepth 77 multipv 2 score cp 110 nodes 179584459114 nps 30865499 hashfull 1000 tbhits 0 time 5818291 pv b2b3 a6c7 a3a4 e8f8 e3e4 c7e8 e2g1 h5h4 g3g4 g6g5 g1f3 f7f6 c1e3 b7b6 f3d2 e8d6 d2c4 d7e8 d1e2 e8g6 a1a2 a8c8 e2f3 g8g7 e3d2 d8d7 f3d1 c8e8 f1f2 f8h8 d1e2 h8f8 d2e3 e8b8 f2f1 f8c8 e3d2 c8f8 e2f2 f8c8 f1c1 c8f8 c1a1 b8b7 f2e2 b7b8 d2c3 b8c8 c4e3 c8b8 c3b2 d7c7 a1f1 c7d7 f1c1

info depth 55 seldepth 83 multipv 3 score cp 85 nodes 179584459114 nps 30865499 hashfull 1000 tbhits 0 time 5818291 pv e3e4 e7g5 c1g5 d8g5 d1b3 b7b6 b3c4 a6c7 b2b4 c7b5 b4c5 b6c5 c4c5 h5h4 g3h4 g5h4 a3a4 e8c8 c5f2 h4f2 f1f2 b5d6 f2f6 d6e8 f6a6 c8c2 e2g1 f7f6 g1f3 d7c8 a6c6 c2b2 d3d4 c8d7 c6a6 e5d4 f3d4 b2b6 a6a5 a7a6 a5c5 b6b2 h2g3 a8c8 c5c8 d7c8 g2f3 g8g7 h3h4 c8d7 g3f4 b2h2 a1h1 h2h1 f3h1 e8d6 h1g2 g7f7
Top
Dann Corbit
Posts: 12540
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Hard-Talkchess-2020 set, final release

Post by Dann Corbit »

Here are some analysis records from my alternate evals table. Unfortunately, I do not record the engine name:

Code: Select all

r2qr1k1/pp1bbp2/n5p1/2pPp2p/8/P2PP1PP/1P2N1BK/R1BQ1R2 w - - acd 37; ce 68; bm b3; c3 "d6"; pm b3; pv b3 Bf6 Ra2 Rf8 Ng1 Nc7 a4 Ne8 Nf3 Nd6 e4 Qe7 Be3 b6 Raf2 Bg7 Nh4 Kh7 Bf3 Bf6 Ng2 Bg7 Qe2 f6 Nh4 Be8 Qd2 Bf7 Bg2 Rad8 Qd1 Rc8 Qe2 Rcd8 Qd2 Rc8 Qd1 Rcd8 Kg1 Be8 Qd2 Bd7 Qe2 Rc8; 
r2qr1k1/pp1bbp2/n5p1/2pPp2p/8/P2PP1PP/1P2N1BK/R1BQ1R2 w - - acd 36; ce 120; bm b3; c3 "d6"; pm b3; pv b3 Nc7 a4 Na6 Ng1 h4 g4 Nb4 e4 g5 Qd2 f6 Ne2 Na6 Nc3 Nc7 Nd1 b6 Ne3 Rf8 Nf5 Ne8 Ba3 Bxf5 gxf5 Nd6 Bf3 Kg7 Bh5 Rh8 Qe2 Nb7 Bg6 Rc8 Bc1 Na5 Qd1 Qc7 Kg2 Bd6 Bd2 Nb7; 
r2qr1k1/pp1bbp2/n5p1/2pPp2p/8/P2PP1PP/1P2N1BK/R1BQ1R2 w - - acd 36; ce 75; bm b3; c3 "d6"; pm b3; pv b3 Bf6 Ra2 Rf8 Ng1 Nc7 Raf2 Bg7 a4 Ne8 Nf3 Nd6 e4 Rc8 Bg5 f6 Be3 Kh7 Nh4 b6 Qf3 Be8 Qe2 Rb8 Nf3 Kg8 Qd1 Qe7 Nd2 Bd7 Qe2 Rbd8 Nc4 Kh7 Qd2 Rc8; 
r2qr1k1/pp1bbp2/n5p1/2pPp2p/8/P2PP1PP/1P2N1BK/R1BQ1R2 w - - acd 33; ce 107; bm b3; c3 "d6"; pm b3; pv b3 Nc7 e4 Rc8 Ra2 Kg7 Nc3 Rf8 Qd2 Rh8 a4 f6 Nd1 Ne8 Ne3 Nd6 Qe1 Qb6 Qc3 h4 g4 g5 Raf2 Be8 Kg1 Bg6 Bd2 a5 Nc4 Nxc4 bxc4 Qc7 Qxa5 Qxa5 Bxa5; 

r2qr1k1/pp1bbp2/n5p1/2pPp2p/8/P2PP1PP/1P2N1BK/R1BQ1R2 w - - acd 58; ce 107; bm d6; c3 "d6"; pm d6; pv d6 Bxd6; 
r2qr1k1/pp1bbp2/n5p1/2pPp2p/8/P2PP1PP/1P2N1BK/R1BQ1R2 w - - acd 57; ce 117; bm d6; c3 "d6"; pm d6; pv d6 Bxd6 Bxb7 Nc7 Bxa8 Qxa8 e4 Qb8 Nc3 Be6 Qc2 Rd8 Be3 Bf8 b4 cxb4 axb4 Qxb4 Rxa7 Nb5 Nxb5 Qxb5 Rd1 Bb3 Qf2 Rd7 Rxd7 Qxd7 Rd2 Bb4 Rb2 Qxd3 Bg5 Qc4 Qe2 Qxe2+ Rxe2 Bf8 Rb2 Bc4 Rb8 Kg7 Rb6 Kh7 Kg2 Bd3 Kf3 Bc2 Rc6 Bd1+ Kg2 Ba4 Rc7 Kg7 Be7 Bxe7 Rxe7 Kf6 Rb7 Bc2 Kf3 Bd3 Ra7 Bc4 Ra5 Ke6 Rc5 Bb3 Rb5 Bd1+ Kf2 Ba4 Ra5 Bc6 Ke3 Kd6; 
r2qr1k1/pp1bbp2/n5p1/2pPp2p/8/P2PP1PP/1P2N1BK/R1BQ1R2 w - - acd 56; ce 108; bm d6; c3 "d6"; pm d6; pv d6 Bxd6 Bxb7 Nc7 Bxa8 Qxa8 e4 Qb8 Nc3 Be6 Qc2 Rd8 Be3 Bf8 b4 cxb4 axb4 Qxb4 Rxa7 Nb5 Nxb5 Qxb5 Rd1 Bb3 Qf2 Rd7 Rxd7 Qxd7 Rd2 Bb4 Rb2 Qxd3 Bg5 Qc4 Bf6 Qc3 h4 Bf8 Qd2 Qxd2+ Rxd2 Bg7 Rd8+ Kh7 Bxg7 Kxg7 Kg2 Be6 Ra8 Kf6 Ra5 Bg4 Kf2 Bd7 Ke3 Ke6 Ra6+ Ke7 Kf2 Be6 Rb6 Kf6 Rb5 Bd7 Ra5 Be6 Ke3 Bd7 Ra7 Ke7 Ra6 Be6 Kd3 Bd7 Rb6 Be6 Rb8 Kd6 Ke3 Bd7 Rb7 Kc6 Ra7 Kd6 Kf3 Bg4+ Kf2 Bd7 Ra6+ Bc6 Kf3 f5; 
r2qr1k1/pp1bbp2/n5p1/2pPp2p/8/P2PP1PP/1P2N1BK/R1BQ1R2 w - - acd 55; ce 109; bm d6; c3 "d6"; pm d6; pv d6 Bxd6 Bxb7 Nc7 Bxa8 Qxa8 e4 Qb7 Nc3 Be6 Qe2 Rb8 Rb1 Be7 Be3 Qc6 Rbc1 Qd7 Qf2 Bf8 h4 Ne8 Qc2 Rb3 Rf2 a5 Qd2 Be7 Bg5 Bf8 Qe3 a4 Qe1 Rb8 Qd1 Bb3 Qf3 Be6 Re1 Bg4 Qe3 Be6 Ref1 Rb3 Qd2 Rb6 Qd1 Rb3 Rd2 Qc6 Be3 Qd7 Qc2 Qc6 Rdf2 Bd6 Bg5 Qd7 Qd2 Bf8 Rd1; 

r2qr1k1/pp1bbp2/n5p1/2pPp2p/8/P2PP1PP/1P2N1BK/R1BQ1R2 w - - acd 37; ce 62; bm e4; c3 "d6"; pm e4; pv e4 Nc7 Nc3 b6 Qf3 f6 Bh6 Nb5 Nxb5 Bxb5 Rf2 Qd6 Raf1 Rac8 b3 Bd7 Bd2 Kg7 Qe2 Rf8 Qe3 Rh8 Bc3 Bb5 Qf3 Rhf8 Qd1 Bd7 Qe2 Rh8 Qb2 Rhf8 Rc1 Rh8 Qe2 Qb8 a4 Qd6 Rcf1 Rhf8; 
r2qr1k1/pp1bbp2/n5p1/2pPp2p/8/P2PP1PP/1P2N1BK/R1BQ1R2 w - - acd 35; ce 86; bm e4; c3 "d6"; pm e4; pv e4 Nc7 Nc3 Kg7 Qd2 Rh8 a4 Ne8 b3 f6 Nb1 Nd6 Qf2 b6 Na3 h4 g4 g5 Nc2 Be8 Ba3 Bg6 Ne3 Rc8 Qe2 Nb7 Bb2 Nd6 Rac1 a6 Ba3 Qd7 Kg1 Nb7 Bb2 b5 Ba3 Na5 Qd1 Bd6; 
r2qr1k1/pp1bbp2/n5p1/2pPp2p/8/P2PP1PP/1P2N1BK/R1BQ1R2 w - - acd 34; ce 119; bm e4; c3 "d6"; pm e4; pv e4 Bg5 Bxg5 Qxg5 Qb3 b6 Qc4 b5 Qc3 Qe7 Qd2 Kg7 Ng1 f6 Nf3 Qd6 Qe3 Qb6 Nd2 Rac8 Rac1 Rf8 Qf2 Rce8 Qe2 Qd6 b3 Rc8 Qe3 Qb6 Qf2 Rh8 Rc2 Rhe8 Qe2 Qd6 Qe3 Qb6 Rfc1 Rcd8 Nf3 b4 axb4; 
r2qr1k1/pp1bbp2/n5p1/2pPp2p/8/P2PP1PP/1P2N1BK/R1BQ1R2 w - - acd 34; ce 94; bm e4; c3 "d6"; pm e4; pv e4 Nc7 a4 Na6 Rf3 b6 Bd2 Kg7 Qc1 f6 Qc4 Nc7 Qc3 Rf8 b3 Ne8 Raf1 Nd6 R3f2 Rc8 Ng1 Nf7 Nf3 Qc7 Qb2 f5 Bc3 Bf6 Rc1 Rh8 h4 Rhd8 Rff1 Qd6 Qe2; 

r2qr1k1/pp1bbp2/n5p1/2pPp2p/8/P2PP1PP/1P2N1BK/R1BQ1R2 w - - acd 37; ce 61; bm Nc3; c3 "d6"; pm Nc3; pv Nc3 Nc7 e4 b6 Qf3 f6 Bh6 Nb5 Nxb5 Bxb5 Rf2 Qd6 Raf1 Rac8 b3 Kh7 Bd2 Kg7 Qe3 Rh8 Kg1 Bd7 Qe2 Rhf8 Qe1 Rfe8 Qc1 Kg8 Bc3 Kg7 Qe3 Rf8 Rf3 b5 Bd2 Rh8 Kh2 Rce8 Qe1 f5; 
r2qr1k1/pp1bbp2/n5p1/2pPp2p/8/P2PP1PP/1P2N1BK/R1BQ1R2 w - - acd 36; ce 62; bm Nc3; c3 "d6"; pm Nc3; pv Nc3 Nc7 e4 b6 Qf3 f6 Bh6 Nb5 Nxb5 Bxb5 Rf2 Rc8 b3 Qd6 Raf1 Bd7 Bd2 Kg7 Qd1 Bb5 Bc1 Bd7 Qd2 Rh8 Qb2 Rhf8 Qe2 Rh8 Be3 Rhd8 Bd2 Rf8 Qe1 Bb5 Qe3 Rh8 Qe2 Rhf8 Qe1; 
r2qr1k1/pp1bbp2/n5p1/2pPp2p/8/P2PP1PP/1P2N1BK/R1BQ1R2 w - - acd 35; ce 116; bm Nc3; c3 "d6"; pm Nc3; pv Nc3 Nc7 a4 Rc8 Nb1 Rf8 Na3 Ne8 Nc4 f6 e4 Nd6 b3 Kg7 Ra2 Qc7 Qd2 Rh8 h4 Nxc4 bxc4 Bg4 Qc3 Rcf8 Raf2 Bd6 a5 Bd7 Qb3 a6 Qa2 Bg4 Be3 f5 Bg5; 
r2qr1k1/pp1bbp2/n5p1/2pPp2p/8/P2PP1PP/1P2N1BK/R1BQ1R2 w - - acd 35; ce 63; bm Nc3; c3 "d6"; pm Nc3; pv Nc3 Nc7 e4 b6 Qf3 f6 Bh6 Nb5 Nxb5 Bxb5 Rf2 Qd6 Raf1 Kh7 Bc1 Kg7 b3 Rh8 Qe2 Raf8 Bd2 Rc8 Kg1 Rcf8 a4 Bd7 Kh2 Rb8 Be3 Rbf8 Bc1 Ra8 Kg1 Raf8 Bd2 Rb8 Qd1 Rhf8; 

r2qr1k1/pp1bbp2/n5p1/2pPp2p/8/P2PP1PP/1P2N1BK/R1BQ1R2 w - - acd 37; ce 107; bm Ng1; c3 "d6"; pm Ng1; pv Ng1 Rc8 e4 Bg5 Rf2 Bxc1 Qxc1 Kg7 Qe3 f6 Raf1 Rf8 g4 hxg4 hxg4 Qe7 Bh3 Nc7 g5 f5 exf5 Bxf5 Bxf5 gxf5 Rxf5 Nxd5 Qe4 Qd6 Ne2 Rxf5 Rxf5 Rf8 Qxe5+ Qxe5+ Rxe5 Nf4 Nxf4 Rxf4 Rxc5 Kg6 Rd5 Rg4 Rd7 b5 Rxa7 Rd4 Ra5 Kxg5 Rxb5+ Kf4 Rb4 Ke3 Rxd4;
I guess that d6 is best, but the other moves would get some points on a varadic scale.
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
Dann Corbit
Posts: 12540
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Hard-Talkchess-2020 set, final release

Post by Dann Corbit »

zullil wrote: ↑Sat May 16, 2020 4:05 pm
Vinvin wrote: ↑Sat May 16, 2020 3:42 am Some comments about the 8 positions :

Very tactical CorChessV6 like it fast. May be Bluefish-FD-XI is right and this move is over evaluated ? May be not ?
r2qr1k1/pp1bbp2/n5p1/2pPp2p/8/P2PP1PP/1P2N1BK/R1BQ1R2 w - - bm d6; id "Hard-Talkchess-2020.176"; c9 Ng1;
Stockfish-dev seems to agree that d6 is "best":

info depth 55 seldepth 71 multipv 1 score cp 146 nodes 179584459114 nps 30865499 hashfull 1000 tbhits 0 time 5818291 pv d5d6 e7d6 g2b7 a6c7
Don't forget to enable tablebase files. I would have tens of billions of tablebase hits by ply 55 here
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
Vinvin
Posts: 5228
Joined: Thu Mar 09, 2006 9:40 am
Full name: Vincent Lejeune

Re: Hard-Talkchess-2020 set, final release

Post by Vinvin »

1 new result -> Eman 5.40 with option "Full Analysis Depth=12" (same as default in Eman 4.30) takes the 2nd place.
It finds 4 positions more than V4.30 : 167 pos (929 sec)

New top 10 :

Code: Select all

 1) Bluefish-XI(Tact=2,defens=no) : 181 pos (671 sec)
 2) Eman 5.40   :                   167 pos (929 sec)
 3) Eman 4.30   :                   163 pos (1087 sec)
 4) Bluefish-XI :                   159 pos (1078 sec)
 5) CorChess v6 :                   156 pos (1203 sec)
 6) houdidit603-tact=1 :            155 pos (1123 sec)
 7) Stockfish 11 :                  141 pos (1459 sec)
 8) Honey-XI :                      139 pos (1440 sec)
 9) Stockfish 10 :                  137 pos (1574 sec)
10) Crystal 2020-05-02 :            136 pos (1421 sec)
Results sheet updated : http://home.scarlet.be/vincentlejeune/c ... 2020v2.ods
Vinvin wrote: ↑Fri May 15, 2020 12:03 am 2 new results :

Code: Select all

Bluefish-XI(Tact=2,defens=no) : 181 pos (671 sec) New number 1 !!!!
houdidit603(default) :          125 pos (1692 sec)
Current top 10

Code: Select all

 1) Bluefish-XI(Tact=2,defens=no) : 181 pos (671 sec)
 2) Eman 4.30   :                   163 pos (1087 sec)
 3) Bluefish-XI :                   159 pos (1078 sec)
 4) CorChess v6                     156 pos (1203 sec)
 5) houdidit603-tact=1 :            155 pos (1123 sec)
 6) Stockfish 11 :                  141 pos (1459 sec)
 7) Honey-XI :                      139 pos (1440 sec)
 8) Stockfish 10 :                  137 pos (1574 sec)
 9) Crystal 2020-05-02 :            136 pos (1421 sec)
10) Black-Diamond-XI :              136 pos (1472 sec)
I reworked the results sheet, I removed the moves to hold only the time to solution : http://home.scarlet.be/vincentlejeune/c ... 2020v2.ods
Top
Vinvin
Posts: 5228
Joined: Thu Mar 09, 2006 9:40 am
Full name: Vincent Lejeune

Re: Hard-Talkchess-2020 set, final release

Post by Vinvin »

Results for Ethereal 12.25 added.
21 more positions found over Ethereal 11.75 !!

Code: Select all

Ethereal 12.25 : 98 solutions found (2126 seconds)
Ethereal 11.75 : 77 solutions found (2520 seconds)
Results sheet updated : http://home.scarlet.be/vincentlejeune/c ... 2020v2.ods
Top

Return to “Computer Chess Club: General Topics”