Not logged inRybka Chess Community Forum
Up Topic The Rybka Lounge / Test Positions / An interesting ending very difficult for engines
- - By MrKris (***) Date 2020-11-12 01:50
From:
http://talkchess.com/forum3/viewtopic.php?f=2&t=74893&start=40#p872068

I suppose it could be called a problem but it looks somewhat like it might have come from a game, say after a big attack by white with some 'remnant' advanced pawns still around or something.


[Event "R7 2700X | RTX 2060"]
[Site "talkchess.com"]
[Date "2020.11.11"]
[Round ""]
[White "various engines"]
[Black ""]
[Result "1-0"]
[BlackElo "2400"]
[Time "16:32:27"]
[WhiteElo "2400"]
[TimeControl "300+3"]
[SetUp "1"]
[FEN "2b1rk2/5p2/p1P5/2p2P2/2p5/7B/P7/2KR4 w - - 0 1"]
[Termination "unterminated"]
[PlyCount "0"]
[WhiteType "human"]
[BlackType "human"]

{Lc0-.23.2_192x16-t60-J64-210 , depth 17/43, time 02:07 , 5,250kn,43knps, +1.57} 1.f6
( {Cfish-201024 - PV=8 , depth 54/63, time 12:01 , 16,215,930kn, 22,481knps, +28.44 >>>>>>} 1. f6 Kg8 2.Rd7 Re1+ 3.Kc2 {+- etc.} )
1... Bxh3 2.c7 Bf5 3.Rd8 c3 4.Kd1 {zugzwang} c2+ 5.Kc1 a5 6.a4 c4 7.Kd2 c1R 8.Kxc1 c3 9.Kd1 c2+ 10.Kc1 Bc8 11.Kxc2 Bf5+ 12.Kc3 Be6 13.Kd4 Bc8 14.Kc5 Bf5 15.Kb6 Bc8 {etc. +-}
1-0


     Times (mm:ss) & depth to find 1.f6:  (R7 2700X 14th | RTX 2060)

     00:02 10/23 Lc0.24.1-SVt40-1541  repeatable: closed & restarted Arena to verify
     02:06 14/31 Lc0 26.3 J92-330
     02:07 17/43 Lc0-.23.2_192x16-t60-J64-210
     02:15 15/35 Lc0-.23.2-net63618
     04:52 18/43 Lc0.24.1-Ls14.3
     07:15 70/30 Cfish-201024-Pure
     12:01 54/63 Cfish-201024 PV=8
     48:47 19/41 Lc0.24.1-SVlgt60-3010


Parent - - By The Wizard (***) Date 2020-11-20 22:39
2b1rk2/5p2/p1P5/2p2P2/2p5/7B/P7/2KR4 w - - 0 1


Analysis by ShashChess 15.1:

1.c7 f6 2.Bg2 Ke7 3.Bc6 Rg8 4.Rd5 Rg1+ 5.Kb2 Rg2+ 6.Kc3 Rxa2 7.Kxc4 Rc2+ 8.Kb3 Rc1 9.Rd8 Bxf5 10.Be4 Be6+ 11.Bd5 Bf5 12.c8R Bxc8 13.Rxc8 Kd6 14.Kb2 Rg1 15.Bb7 f5 16.Kc3 Ke5 17.Rxc5+ Kf4 18.Rc4+ Kg3 19.Bxa6
  White has an edge: = (0.25)  Depth: 1/1   00:00:01  1555N
  The position is equal: = (0.06)  Depth: 25/42   00:00:01  27536kN, tb=66349
1.Rd5 f6 2.Kd2 Re4 3.Rd8+ Re8 4.c7 Ke7 5.Kc3 Rxd8 6.cxd8Q+ Kxd8 7.Kxc4 Kc7 8.Kxc5 Bd7 9.Bg4 Bc8 10.a3 Bd7 11.Kd5 Bc8 12.Bh3 Bd7 13.Bg4
  The position is equal: = (0.00)  Depth: 26/42   00:00:01  42283kN, tb=105138
  The position is equal: = (0.00)  Depth: 32/20   00:00:01  101MN, tb=327991
1.c7 f6 2.Bf1 Ke7 3.Bxc4 Rh8 4.Rd5 Bb7 5.Rd2 Bc8
  The position is equal: = (0.00)  Depth: 33/26   00:00:01  104MN, tb=335737
  The position is equal: = (0.00)  Depth: 40/12   00:00:02  198MN, tb=745977
1.f6 Bxh3 2.c7 Kg8 3.Rd8 Rf8 4.Kc2 Bf5+ 5.Kc3 Bc8 6.Kxc4 Be6+ 7.Kc3 c4 8.Kd4 c3 9.Kxc3 Bh3 10.Kc4 Be6+ 11.Kb4 Bc8 12.a3 Be6 13.a4 Rxd8 14.cxd8Q+ Kh7 15.Kc3 Kg6 16.Qd3+ Bf5 17.Qxa6 Be6 18.a5 Bf5 19.Qb5 Bc8 20.Qe5 Be6 21.Qg3+ Kf5 22.Qd3+ Kf4 23.Qd4+ Kg5
  White is clearly winning: +- (6.37 ++)  Depth: 41/52   00:00:04  386MN, tb=1651638
  White is clearly winning: +- (10.24)  Depth: 41/52   00:00:10  1481MN, tb=4184301

(,  20.11.2020)
Parent - By MrKris (***) Date 2020-11-24 21:57
Thanks! Same thing here, ShashChess 15.1 5 seconds, latest Cfish over 11 1/2 minutes!
2b1rk2/5p2/p1P5/2p2P2/2p5/7B/P7/2KR4 w - - 0 1
R7 2700X, 8 GBh, Sy.6, 7-threads each engine simultaneously:

ShashChess15.1-avx2:
  1/1  00:00           199  100k  +0.61  1.Rd5
... ... ...
29/33  00:05       15,179k  3,128k  +2.88  1.f6 Bxh3 2.c7 Bf5 3.Rd8 a5 4.Kb2 a4 5.Kc3 Be6 6.a3 Bc8 7.Kxc4 Bf5 8.Kxc5 Bh3 9.Kc4 Bc8 10.Kb5 Bh3 11.Kc5 Bf5 12.Kb6 Bc8 13.Kc6
...eval. gets higher

Cfish-201115-avx2:
  1/1  00:00           168  56k  +0.54  1.Rd5
,,, ,,, ,,,
69/32  10:51    5,270,468k  8,097k   -0.00  1.c7...
  70/9+  11:29    5,619,398k  8,164k  +0.08  1.f6
70/13+  11:34    5,665,783k  8,175k  +0.16  1.f6
70/13+  11:35    5,679,551k  8,178k  +0.28  1.f6
70/13+  11:36    5,689,612k  8,180k  +0.46  1.f6
70/13+  11:37    5,694,642k  8,181k  +0.71  1.f6
70/13+  11:38    5,707,279k  8,183k  +1.03  1.f6
70/13+  11:39    5,721,222k  8,187k  +1.47  1.f6
70/23+  11:42    5,746,079k  8,193k  +2.03  1.f6
70/23+  11:50    5,834,926k  8,224k  +2.75  1.f6
70/23+  12:01    5,952,802k  8,258k  +3.68  1.f6
...eval. slowly gets higher
Parent - By user923005 (****) Date 2020-11-20 23:48 Upvotes 1
2b1rk2/5p2/p1P5/2p2P2/2p5/7B/P7/2KR4 w - - bm f6; ce 32706; dm 31; c0 "
study by Smyslov"; c3 "f6"; c5 "2010.02.23;"; pv f6 Bxh3 c7 Bf5 Rd8 c3 Kd1 a5 a4 c4 Kc1 c2 Kd2 c1=R Kxc1 c3 Kd1 c2+ Kc1 Be6 Kxc2 Bg4 Kd3 Bh3 Kd4 Be6 Kc5 Bf5 Kb5 Kg8 c8=Q Bxc8 Rxe8+ Kh7 Rxc8 Kg6 Rc6 Kh5 Kxa5 Kg4 Kb5 Kg5 a5 Kg4 a6 Kf4 Rc5 Kf3 a7 Ke3 a8=Q Kf4 Rc4+ Ke3 Qe4+ Kf2 Rc2+ Kg3 Rg2+ Kh3 Qg4+;
Up Topic The Rybka Lounge / Test Positions / An interesting ending very difficult for engines

Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill