Cranky Walters

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

Moderators: hgm, Rebel, chrisw

Dann Corbit
Posts: 12540
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Cranky Walters

Post by Dann Corbit »

Ferdy wrote: I am surprised with CDrill's perf, since this engine has no transposition table, no null move pruning, no LMR rated around 1800 on CCRL 40/4.
The lack of transposition table is clearly a hindrance, but I strongly suspect that lack of null move pruning and LMR is a huge benefit for many of the problems in this test suite.

Both null move and LMR lop off stupid looking moves like tossing a rook for a bishop or a bishop for a pawn. Test sets are full of surprise moves, where the key move often involves a serious material loss (or harder yet, an interior move involves a serious material loss).

|I guess that if you turn off LMR and NULL MOVE pruning on strong engines, it will help them solve a large fraction of the problems in the set (though some of the problems may be depth driven also)
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
Ferdy
Posts: 4833
Joined: Sun Aug 10, 2008 3:15 pm
Location: Philippines

Re: Cranky Walters

Post by Ferdy »

Dann Corbit wrote:Yes, I have the eret suite. The only one unsolved for me is:
[d]4r1k1/1r1np3/1pqp1ppB/p7/2b1P1PQ/2P2P2/P3B2R/3R2K1 w - - acd 34; acs 1214; bm Be3; c3 "Bg7"; ce 38; id "ERET 084 - Koenigsangriff"; pm Be3; pv Be3 Nf8 Bxc4+;

For all the others, I have a mechanical solution that agrees with the proposed solution.

I have 383 positions from various test suites that he wrote and I have a total of 15 where I did not come up with the same answer as the proposed answer which also gives the best score.
Thanks good info.
Top
Ferdy
Posts: 4833
Joined: Sun Aug 10, 2008 3:15 pm
Location: Philippines

Re: Cranky Walters

Post by Ferdy »

Dann Corbit wrote:
Ferdy wrote: I am surprised with CDrill's perf, since this engine has no transposition table, no null move pruning, no LMR rated around 1800 on CCRL 40/4.
The lack of transposition table is clearly a hindrance, but I strongly suspect that lack of null move pruning and LMR is a huge benefit for many of the problems in this test suite.

Both null move and LMR lop off stupid looking moves like tossing a rook for a bishop or a bishop for a pawn. Test sets are full of surprise moves, where the key move often involves a serious material loss (or harder yet, an interior move involves a serious material loss).

|I guess that if you turn off LMR and NULL MOVE pruning on strong engines, it will help them solve a large fraction of the problems in the set (though some of the problems may be depth driven also)
Agreed those reductions and prunings are indeed some of the reasons of failing to solve the problem. We hope our clever programmers will be able to solve this issue. There seems to be a lot still to be discovered on the proper use of null move pruning :).
Top
peter
Posts: 3186
Joined: Sat Feb 16, 2008 7:38 am
Full name: Peter Martan

Re: Cranky Walters

Post by peter »

Hi Dann
Dann Corbit wrote:Yes, I have the eret suite. The only one unsolved for me is:
[d]4r1k1/1r1np3/1pqp1ppB/p7/2b1P1PQ/2P2P2/P3B2R/3R2K1 w - - acd 34; acs 1214; bm Be3; c3 "Bg7"; ce 38; id "ERET 084 - Koenigsangriff"; pm Be3; pv Be3 Nf8 Bxc4+;
84: ERET 084 - Koenigsangriff, Persson-Teichmeister, CorrGame 2009
[d]4r1k1/1r1np3/1pqp1ppB/p7/2b1P1PQ/2P2P2/P3B2R/3R2K1 w - - 0 1

Analysis by Brainfish 190317 64 POPCNT:
...
28.Le3 Sf8
+/= (0.59 --) Tiefe: 27/55 00:00:30 613MN
...
28.Lxc4+ Dxc4
= (0.20 --) Tiefe: 32/58 00:02:40 3591MN
28.Lg7
+/= (0.32 ++) Tiefe: 32/66 00:03:26 4644MN
...
28.Lg7 Kf7 29.Dh7 Se5 30.f4 Dxe4 31.Lxf6+ Ke6 32.Lxe5 Lxe2 33.f5+ Kd7 34.Txd6+ Kc8 35.Dxg6 Lb5 36.De6+ Ld7 37.Txd7 Txd7 38.Tb2 b5 39.a4 Dxg4+ 40.Kf2 De4 41.Td2 Ted8 42.axb5 Dxe5 43.Dxe5 Txd2+ 44.Kg3 T2d3+ 45.Kf4 T8d7 46.Kg5 a4 47.De4 Kb8 48.Df4+ T7d6 49.De5 a3 50.c4 a2 51.c5 T3d5 52.Dh8+ Td8 53.Dh2+ Kc8 54.Dxa2 Txc5 55.b6
+/- (1.36) Tiefe: 32/66 00:04:49 6513MN
...
28.Lg7 Kf7 29.Dh7 Dc5+ 30.Td4 Se5 31.Lxc4+ Sxc4 32.f4 Ke6 33.Lxf6 Kd7 34.Df7 Se3 35.Le5 Sc4 36.Lg7 Tc7 37.e5 Td8 38.Lf6 Kc8 39.Lxe7 Kb7 40.Dxc4 Dxc4 41.Txc4 Txe7 42.Td2 Te6 43.Kf2 g5 44.Tcd4 Kc6 45.exd6 gxf4 46.d7 Te7 47.Tc4+ Kb7 48.Kf3 Tdxd7 49.Txd7+ Txd7 50.g5 Tg7 51.Kxf4 Tg8 52.Kf5
+- (1.72) Tiefe: 35/77 00:08:34 11679MN


Modern high class corr.- games would be best source for really difficult test positions, especially as for practical relevance compared to e.g. studies.

Main problem is, engines don't get the key moves if they are to be found interactively only, so automatic screening of databases with short TC doesn't find them neither
Peter.
Top
Dann Corbit
Posts: 12540
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Cranky Walters

Post by Dann Corbit »

After some weekend analysis, the following are the only unsolved positions from the list:
r4rk1/ppnqp2p/3p2p1/2pP4/P1P2B1P/6P1/1P3PK1/R2Q1R2 b - - bm Rxf4;
r3rnk1/1bq1bpp1/pp2p3/3pP2Q/P4p2/1NNB1R2/1PP4P/2KR4 w - - bm Rg1;
r1br2k1/7p/p2q1ppB/1p1P4/2P5/5N2/PQ3PPP/4R1K1 w - - bm Ng5;
r1b2rk1/pp3pb1/3q2pp/2nP4/6P1/2P2N2/P1Q1BPP1/2BR1RK1 w - - bm Rd4;
5rk1/ppqb2pp/2n1p3/3pP3/2pP4/P1P1BN1Q/2P2PP1/R3K3 b Q - bm Rxf3;
4r1k1/pp5p/2p1bR2/2P5/2B1n3/2B2K2/PP5P/8 w - - bm Rxe6;
3rrnk1/p1n1qpp1/1p5p/2pRP3/2P4P/1P2Q1PB/PB5K/4R3 w - - bm Red1;
3r2k1/p1R2ppp/1p6/P1b1PP2/3p4/3R2B1/5PKP/1r6 w - - bm f6;
2r1k1r1/3q4/p1nBpp1p/8/B5p1/4RbK1/PP1Q1P1P/2R5 w - - bm h4;

My last analysis run was posted at :
http://rybkaforum.net/cgi-bin/rybkaforu ... pid=572642
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: To Vincent --> Cranky Walters

Post by Paloma »

I read in the Rybka_Fora your Post with the Solutiontable from the Cranky Walters Testset by Dann Corbit.
http://rybkaforum.net/cgi-bin/rybkaforu ... ?tid=32055 (Last Post)

Why is the order of the positions so mixed?
E.g. There are 2 pos with 1.h4 in the Dann's epd (19 and 24)
In your solution table are the numbers 9 and 20
How do I know which position in your solution table is assigned to which h4-epd-position?

The same with 1.Ng6 >> 1 + 23 in Dann's epd, 6 + 12 in your table
Your solution table starts with__: Ke1, _Kf6, Rd4, Ng5, b7, _ Ng6, g4 etc
Original epd by Dann starts with: Ng6+, Bg7, Ke1, Kf6, Bxd8, b7, _ f6 etc
You are giving your set at the end, but I have Dann's epd already in march to my *.cbh integrated.

Things are now mixed. I must now fumble things in proper order.

Nevertheless many thanks for your solution table.
Top
Vinvin
Posts: 5228
Joined: Thu Mar 09, 2006 9:40 am
Full name: Vincent Lejeune

Re: To Vincent --> Cranky Walters

Post by Vinvin »

Paloma wrote:I read in the Rybka_Fora your Post with the Solutiontable from the Cranky Walters Testset by Dann Corbit.
http://rybkaforum.net/cgi-bin/rybkaforu ... ?tid=32055 (Last Post)

Why is the order of the positions so mixed?
E.g. There are 2 pos with 1.h4 in the Dann's epd (19 and 24)
In your solution table are the numbers 9 and 20
How do I know which position in your solution table is assigned to which h4-epd-position?

The same with 1.Ng6 >> 1 + 23 in Dann's epd, 6 + 12 in your table
Your solution table starts with__: Ke1, _Kf6, Rd4, Ng5, b7, _ Ng6, g4 etc
Original epd by Dann starts with: Ng6+, Bg7, Ke1, Kf6, Bxd8, b7, _ f6 etc
You are giving your set at the end, but I have Dann's epd already in march to my *.cbh integrated.

Things are now mixed. I must now fumble things in proper order.

Nevertheless many thanks for your solution table.
I can't remember how I got this order.
Sorry for the inconvenience.
Top
Vinvin
Posts: 5228
Joined: Thu Mar 09, 2006 9:40 am
Full name: Vincent Lejeune

Re: Cranky Walters

Post by Vinvin »

Dann Corbit wrote:After some weekend analysis, the following are the only unsolved positions from the list:
r4rk1/ppnqp2p/3p2p1/2pP4/P1P2B1P/6P1/1P3PK1/R2Q1R2 b - - bm Rxf4;
r3rnk1/1bq1bpp1/pp2p3/3pP2Q/P4p2/1NNB1R2/1PP4P/2KR4 w - - bm Rg1;
r1br2k1/7p/p2q1ppB/1p1P4/2P5/5N2/PQ3PPP/4R1K1 w - - bm Ng5;
r1b2rk1/pp3pb1/3q2pp/2nP4/6P1/2P2N2/P1Q1BPP1/2BR1RK1 w - - bm Rd4;
5rk1/ppqb2pp/2n1p3/3pP3/2pP4/P1P1BN1Q/2P2PP1/R3K3 b Q - bm Rxf3;
4r1k1/pp5p/2p1bR2/2P5/2B1n3/2B2K2/PP5P/8 w - - bm Rxe6;
3rrnk1/p1n1qpp1/1p5p/2pRP3/2P4P/1P2Q1PB/PB5K/4R3 w - - bm Red1;
3r2k1/p1R2ppp/1p6/P1b1PP2/3p4/3R2B1/5PKP/1r6 w - - bm f6;
2r1k1r1/3q4/p1nBpp1p/8/B5p1/4RbK1/PP1Q1P1P/2R5 w - - bm h4;

My last analysis run was posted at :
http://rybkaforum.net/cgi-bin/rybkaforu ... pid=572642
1) r4rk1/ppnqp2p/3p2p1/2pP4/P1P2B1P/6P1/1P3PK1/R2Q1R2 b - - bm Rxf4;
After 1. ... Rxf4 2. gxf4, SF sees no advantage for black with 2 best moves :

Code: Select all

Stockfish_170417_x64_modern:
 36/68	05:16	 3 408 884k	10 779k	+1,58	2. ... Qf5 3.Re1 e6 4.Qf3 Rf8 5.Rad1 exd5 6.Re7 Ne6 7.Qxd5 Qxd5+ 8.cxd5 Nxf4+ 9.Kf1 Rf7 10.Re8+ Rf8 11.Rde1 Nxd5 12.R8e6 Rf7 13.Rxd6 Nf6 14.Kg2 Kg7 15.a5 Rc7 16.a6 Nh5 17.Red1 Nf6 18.axb7 Rxb7 19.Ra1 Nh5 20.Ra5 Nf4+ 21.Kf3 Rf7 22.Rda6 Nh3+ 23.Ke3 Nxf2 24.Rxa7 Ng4+ 25.Ke4
 37/68	05:16	 3 408 884k	10 779k	+0,23	2. ... Rf8 3.Qf3 e5 4.Qh3 Qxh3+ 5.Kxh3 Rxf4 6.Rfc1 Na6 7.Kg3 Nb4 8.b3 Kg7 9.f3 Kf6 10.a5 Kf5 11.Rab1 Rd4 12.Rd1 Nd3 13.Rh1 h5 14.Rbd1 Kf6 15.Rh2 Kf5 16.Rb1 Nb4 17.Rg2 Nd3 18.Rd2 g5 19.hxg5 Kxg5 20.Rg1 h4+ 21.Kh2+ Kf4 22.Rg6 Kxf3 23.Rxd6 Nf4
2) r3rnk1/1bq1bpp1/pp2p3/3pP2Q/P4p2/1NNB1R2/1PP4P/2KR4 w - - bm Rg1;
After a bit of learning, with 2 best moves, Rg1 is clearly better than Rh3 :

Code: Select all

Stockfish_170417_x64_modern:
 37/68	03:45	 2 282 368k	10 132k	+1,70	1.Rh3 f5 2.Re1 f3 3.Nd1 d4 4.Nxd4 Bb4 5.Rg1 Qxe5 6.Qh8+ Kf7 7.Rhg3 Ng6 8.Qh7 Qxd4 9.Qxg6+ Kf8 10.Qh7 Qf4+ 11.Kb1 Qh6 12.Qxh6 gxh6 13.Rg8+ Ke7 14.R8g7+ Kf6 15.Rxb7 Re7 16.Rxe7 Bxe7 17.Rf1 Rg8 18.Bxa6 Bd6 19.Rxf3 Bxh2 20.Rh3 Bf4 21.Nc3 Rg3 22.Rh5 Rg2 23.Bd3 Bg5 24.Bb5 Rg4 25.Ka2 Rh4 26.Rxh4 Bxh4 27.Be2 Ke5 28.Kb3
 37/68	03:45	 2 282 368k	10 132k	+2,59	1.Rg1 f5 2.Rxg7+ Kxg7 3.Rh3 Ng6 4.Bxf5 exf5 5.Qh7+ Kf8 6.e6 Bd6 7.Qxg6 Be5 8.Rh5 Bg7 9.a5 Re7 10.Qh7 Rxe6 11.Rxf5+ Qf7 12.Rxf7+ Kxf7 13.Nd4 Rh6 14.Qf5+ Kg8 15.Qg4 Rf8 16.Nf5 Rxf5 17.Qxf5 Rf6 18.Qd7 Rf7 19.Qd8+ Kh7 20.axb6 Bxc3 21.bxc3 f3 22.Qh4+ Kg7 23.Qg3+ Kf8 24.Qf2 Ke8 25.Kb2 Kd7 26.h4 Kc6 27.h5 a5 28.Qg3 Kb5 29.Qg6 Rf8 30.Qg7
3) r1br2k1/7p/p2q1ppB/1p1P4/2P5/5N2/PQ3PPP/4R1K1 w - - bm Ng5;
SF clearly prefers Nd2 over Ng5, 3 best moves :

Code: Select all

 42/80	11:44	 7 676 083k	10 899k	+0,39	1.Ng5 Bf5 2.g4 Re8 3.Rxe8+ Rxe8 4.gxf5 Qe5 5.Qxe5 Rxe5 6.Nf3 Rxf5 7.Nd4 Rh5 8.d6 Kf7 9.d7 Ke7 10.cxb5 axb5 11.Bg7 Kxd7 12.Bxf6 b4 13.Nc2 Ke6 14.Bd8 Rd5 15.Bb6 Rd6 16.Be3 Rd1+ 17.Kg2 Rb1 18.Kf3 Rb2 19.Ne1 Rxa2 20.Nd3 Rc2 21.Nxb4 Rc4 22.Nd3 Kf5 23.Nf4 Ra4 24.Ne2 Rh4 25.Kg2 Rb4 26.Nd4+ Ke4 27.Nf3 Kf5 28.Kg3 Rg4+ 29.Kh3 Re4 30.Nd4+ Ke5 31.Ne2 Rc4 32.Kg3 Kf5 33.Nd4+ Ke5 34.Nf3+ Kf5
 42/80	11:44	 7 676 083k	10 899k	+0,64	1.c5 Qxd5 2.Qxf6 Ra7 3.Ng5 Qd1 4.Qe5 Rad7 5.Qe6+ Kh8 6.Nf3 Qd5 7.Qf6+ Kg8 8.c6 Rf7 9.Qh4 Bf5 10.h3 Rc8 11.Ne5 Re8 12.Re3 Rfe7 13.Qf6 Qe6 14.Nd7 Qxf6 15.Nxf6+ Kf7 16.Nxe8 Rxe3 17.Bxe3 Kxe8 18.a3 a5 19.Bd2 a4 20.f3 Bc8 21.Kf2 Ke7 22.Bb4+ Ke6 23.Ke3 Kd5 24.Kf4 h6 25.c7 Kc6 26.Ba5 Kd6 27.Ke4 b4 28.Bxb4+ Kxc7 29.Bf8 h5 30.Ke5 Ba6 31.Kf6 Bf1
 42/80	11:44	 7 676 083k	10 899k	+1,65	1.Nd2 Ra7 2.Ne4 Re7 3.f3 Rxe4 4.fxe4 bxc4 5.Qd4 Rd7 6.Rf1 Rf7 7.Qxc4 Bd7 8.Qd4 Be8 9.Rf3 Qe5 10.Qxe5 fxe5 11.Rc3 Rb7 12.Kf2 Ba4 13.Rc8+ Kf7 14.Kf3 Bd1+ 15.Ke3 Ba4 16.Rf8+ Ke7 17.Rh8 Kf6 18.Ra8 Bb5 19.Rf8+ Rf7 20.Bg5+ Kg7 21.Rc8 Rf1 22.Rc7+ Kf8 23.Be7+ Ke8 24.Bd6 Re1+ 25.Kf3 Rf1+ 26.Kg3 Re1 27.Re7+ Kd8 28.Rxe5 Re3+ 29.Kf4 Re2 30.Re7 h5
4) r1b2rk1/pp3pb1/3q2pp/2nP4/6P1/2P2N2/P1Q1BPP1/2BR1RK1 w - - bm Rd4;
What's the point of 1.Rd4 ? SF doesn't understand and simply play 1...Re8 :

Code: Select all

Stockfish_170417_x64_modern:
 32/59	05:57	 3 224 672k	9 008k	+0,97	1. ... Qf6 2.g5 hxg5 3.Bxg5 Qd6 4.Bf4 Qb6 5.Rb4 Bf5 6.Qd1 Qd8 7.Be5 Bxe5 8.Nxe5 Re8 9.Ng4 Bxg4 10.Bxg4 Qd6 11.Bf3 Rad8 12.Rh4 Qf6 13.Qd4 Qxd4 14.Rxd4 Rd6 15.g3 Kg7 16.Rc1 f5 17.Kg2 Kf6 18.g4
 32/59	05:57	 3 224 672k	9 008k	+0,74	1. ... Bxd4 2.cxd4 Nd7 3.Bxh6 Re8 4.Qd2 Nf6 5.Ne5 Nxd5 6.Bc4 Be6 7.f4 Nf6 8.f5 Ne4 9.Qd3 Bxc4 10.Qxc4 Rxe5 11.dxe5 Qb6+ 12.Kh2 gxf5 13.Bc1 Kg7 14.g3 Qc5 15.Qxc5 Nxc5 16.gxf5 Nd3 17.Bf4 Re8 18.Rb1 Nxe5 19.Rxb7 Nd3 20.Rxa7 Kf6 21.Kg2 Nxf4+ 22.gxf4 Kxf5
 33/59	05:57	 3 224 672k	9 008k	+0,16	1. ... Re8 2.Bf4 Qd8 3.Be3 Bd7 4.Rc4 b6 5.Nd4 Rxe3 6.fxe3 Qg5 7.Nb3 Be5 8.Rcf4 Bxf4 9.exf4 Qe7 10.Qd2 Ne4 11.Qe3 Re8 12.Bf3 Nf6 13.Qxe7 Rxe7 14.d6 Re3 15.c4 Rd3 16.Rd1 Rxd1+ 17.Bxd1 Ne4 18.Bf3 Nxd6 19.c5 Ne8 20.Kf2 bxc5 21.Nxc5
5) 5rk1/ppqb2pp/2n1p3/3pP3/2pP4/P1P1BN1Q/2P2PP1/R3K3 b Q - bm Rxf3;
Rf3, what's the point ? SF simply recaptures with the Q with white advantage :

Code: Select all

Stockfish_170417_x64_modern:
 36/49	03:15	 2 017 802k	10 327k	+1,60	2. ... Qd8 3.Qh3 Qe8 4.Kd2 Qg6 5.g4 h6 6.f4 Ne7 7.Rf1 b6 8.Bf2 Ba4 9.Rc1 Qh7 10.Bh4 Ng6 11.f5 exf5 12.Qf3 Kh8 13.Qxd5 Qg8 14.Qxg8+ Kxg8 15.Bg3 f4 16.Bh2 Bd7 17.Rg1 Be6 18.g5 hxg5 19.Rxg5 Kf7 20.Rg1 Bd5 21.Rf1 f3 22.Ke3 Ne7
 37/49	03:15	 2 017 802k	10 327k	+1,27	2. ... Qc8 3.a4 b6 4.Qd1 Qf8 5.Kd2 Be8 6.f3 Bg6 7.Qc1 Qf5 8.Qb2 h6 9.Rf1 Qh5 10.Rg1 Kh7 11.Qa3 Qf5 12.Rc1 Be8 13.Rh1 Kg8 14.Qd6 Bg6 15.Rc1 Na5 16.Ke1 Qf7 17.Qa3 Nb7 18.Kf2 Bf5 19.Qb2 Bg6 20.Qb5 Na5 21.Kf1 Kh7 22.Kg1 Kg8
 37/49	03:15	 2 017 802k	10 327k	+1,27	2. ... Be8 3.Qd1 Qe7 4.Kd2 b6 5.Qb1 h6 6.f3 Qd7 7.a4 Qf7 8.Qh1 Qe7 9.Qc1 Qf7 10.Qa3 Qf5 11.Rc1 Bg6 12.Qd6 Na5 13.Ke1 Qf7 14.Qa3 Nb7 15.Kf2 Na5 16.Qd6 Kh7 17.Kg1 Nb7 18.Qb4 Kg8 19.Kf2 Qc7 20.Ke1 Qf7 21.Qb5
6) 4r1k1/pp5p/2p1bR2/2P5/2B1n3/2B2K2/PP5P/8 w - - bm Rxe6;
SF evals Bxe6+ as completely winning too :

Code: Select all

Stockfish_170417_x64_modern:
 39/74	01:57	 1 704 160k	14 517k	+4,48	1.Rxe6 Ng5+ 2.Kf4 Nxe6+ 3.Kf5 Kf7 4.Bf6 b5 5.cxb6 axb6 6.a4 b5 7.axb5 cxb5 8.Bd5 b4 9.h4 h6 10.Bc4 Kg8 11.Bxe6+ Kh7 12.b3 Rf8 13.Bc4 Re8 14.Bd3 Re1 15.Be7 Re3 16.Be4 Kg7 17.Bf6+ Kf8 18.Bd5 Rd3 19.Bc4 Rd2 20.Be5 Rg2 21.Bd6+ Ke8 22.Bf4 h5 23.Bg5 Ra2 24.Kg6 Kf8 25.Be3 Rg2+ 26.Kh6 Ke7 27.Kxh5 Kf6 28.Bg5+ Kg7 29.Be7 Rg6 30.Bf8+ Kf6 31.Bxb4 Rg2 32.Bc5
 39/74	01:57	 1 704 160k	14 517k	+4,69	1.Bxe6+ Rxe6 2.Rxe6 Ng5+ 3.Kf4 Nxe6+ 4.Ke5 Nxc5 5.Kd6 Nd3 6.Kc7 b5 7.Kxc6 b4 8.Bd4 a6 9.Kb6 Kf7 10.Kxa6 Ke6 11.Kb5 Kd5 12.Bf6 Kd6 13.Kc4 Nf4 14.Bd4 Nd5 15.Kb5 Kd7 16.Bc5 b3 17.a4 h5 18.Kc4 Kc6 19.Bd4 Ne7 20.Kxb3 Nf5 21.Kc3 Kd5 22.Bf2 Nd6 23.Kb4 Kc6 24.Be3 h4 25.a5 Nf5 26.Bc5 Nh6 27.a6 h3 28.Kc4 Ng4 29.b4 Ne5+ 30.Kd4 Nf7
7) 3rrnk1/p1n1qpp1/1p5p/2pRP3/2P4P/1P2Q1PB/PB5K/4R3 w - - bm Red1;
After : 1. Red1 Nxd5 2. cxd5 Ng6 3. e6 Nf8 4. d6 Rxd6 5. Qe5 Rd4 6. Bxd4 cxd4 7.exf7+ Qxf7 8. Qxd4, SF evals the positions as about =

8) 3r2k1/p1R2ppp/1p6/P1b1PP2/3p4/3R2B1/5PKP/1r6 w - - bm f6;
SF prefers Bh4 big time ! 3 best moves :

Code: Select all

Stockfish_170417_x64_modern:
 31/60	02:46	 1 721 338k	10 343k	+1,97	1.f6 Rf8 2.e6 Re1 3.e7 Re8 4.Bh4 Re4 5.Rd7 Bxe7 6.Rxe7 R4xe7 7.fxe7 bxa5 8.Rxd4 f6 9.Rd7 Kf7 10.Rxa7 Rxe7 11.Rxa5 g5 12.Bg3 Re6 13.Rd5 Rc6 14.Rd7+ Kg6 15.Rd6 Rc5 16.Rb6 Rf5 17.Bd6 g4 18.Ba3 h5 19.Bc1 Rf3 20.Be3 Kf7 21.Rb4 Ke6 22.Rc4 Rf5 23.Rc6+ Kf7 24.Bd4 h4
 31/60	02:46	 1 721 338k	10 343k	+2,77	1.a6 h6 2.Bh4 Ra8 3.Rg3 d3 4.Rxd3 Rb4 5.Bg3 Rd4 6.Rxd4 Bxd4 7.Rd7 Bc5 8.Kf3 Kf8 9.e6 fxe6 10.fxe6 g6 11.Be5 Ke8 12.Rh7 Bf8 13.Ke4 b5 14.Bd4 h5 15.Bxa7 b4 16.Kd5 b3 17.e7 b2 18.exf8Q+ Kxf8 19.Bc5+ Ke8
 32/60	02:46	 1 721 338k	10 343k	+3,42	1.Bh4 Re8 2.a6 Kf8 3.e6 fxe6 4.f6 gxf6 5.Bxf6 e5 6.Rf3 Re6 7.Be7+ Kg8 8.Rf8+ Kg7 9.Bxc5+ Kg6 10.f4 Rf6 11.fxe5 Rxf8 12.Bxf8 d3 13.Rd7 Kf5 14.Rxd3 Ra1 15.Bd6 Rxa6 16.Rf3+ Ke6 17.Rf6+ Kd5 18.Bb8 Ra1 19.Rd6+ Kc5 20.Kf2 Ra2+ 21.Kf3 a5 22.e6
9) 2r1k1r1/3q4/p1nBpp1p/8/B5p1/4RbK1/PP1Q1P1P/2R5 w - - bm h4;
SF thinks 1.Bb3 (then 3.h4) is very close to 1.h4. 2 best lines :

Code: Select all

Stockfish_170417_x64_modern:
 40/73	26:23	 16 515 812k	10 428k	+2,16	1.Bb3 Nd8 2.Rxc8 Qxc8 3.h4 Bc6 4.Bf4 Rg7 5.Qc3 Rf7 6.Qd3 Bd7 7.Qg6 Qc6 8.Kh2 Qb6 9.Qxg4 e5 10.Qg3 Ke7 11.Bxh6 Ne6 12.Rd3 a5 13.a3 e4 14.Be3 Qb5 15.Qd6+ Ke8 16.Rc3 Qe5+ 17.Qxe5 fxe5 18.Bxe6 Bxe6 19.Rc5 Kd7 20.Rxa5 Rf3 21.Kg2 Bh3+ 22.Kg1 Bg4 23.b4 Kd6 24.Ra7 Rh3 25.Kg2 Bf5 26.Ra6+ Kd5 27.Rh6 Kc4 28.Rh5 Bd7 29.b5 Kxb5 30.Rxe5+ Kc6 31.Rxe4 Kb5 32.a4+ Kc6 33.Rd4 Be6
 40/73	26:23	 16 515 812k	10 428k	+2,29	1.h4 Kf7 2.Rxe6 Qxe6 3.Bb3 Nd4 4.Bxe6+ Nxe6 5.Bc7 Ng7 6.Qd7+ Kg6 7.Qd3+ Kf7 8.Qxa6 Rce8 9.Qd3 Re4 10.Kh2 Rge8 11.Ba5 h5 12.Rc7+ Kg6 13.Qd7 R4e7 14.Qxe7 Rxe7 15.Rxe7 Bd5 16.b4 Nf5 17.Rd7 Bc6 18.Rc7 Be8 19.Rc8 Bd7 20.Rd8 Bc6 21.Rg8+ Kf7 22.Rc8 Be8 23.Rc5 Ke6 24.b5 Nd6 25.b6 Nb7 26.Rc8 Bd7 27.Ra8
Top
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: Cranky Walters

Post by MikeB »

Vinvin wrote: ...
2) r3rnk1/1bq1bpp1/pp2p3/3pP2Q/P4p2/1NNB1R2/1PP4P/2KR4 w - - bm Rg1;
After a bit of learning, with 2 best moves, Rg1 is clearly better than Rh3 :

Code: Select all

Stockfish_170417_x64_modern:
 37/68	03:45	 2 282 368k	10 132k	+1,70	1.Rh3 f5 2.Re1 f3 3.Nd1 d4 4.Nxd4 Bb4 5.Rg1 Qxe5 6.Qh8+ Kf7 7.Rhg3 Ng6 8.Qh7 Qxd4 9.Qxg6+ Kf8 10.Qh7 Qf4+ 11.Kb1 Qh6 12.Qxh6 gxh6 13.Rg8+ Ke7 14.R8g7+ Kf6 15.Rxb7 Re7 16.Rxe7 Bxe7 17.Rf1 Rg8 18.Bxa6 Bd6 19.Rxf3 Bxh2 20.Rh3 Bf4 21.Nc3 Rg3 22.Rh5 Rg2 23.Bd3 Bg5 24.Bb5 Rg4 25.Ka2 Rh4 26.Rxh4 Bxh4 27.Be2 Ke5 28.Kb3
 37/68	03:45	 2 282 368k	10 132k	+2,59	1.Rg1 f5 2.Rxg7+ Kxg7 3.Rh3 Ng6 4.Bxf5 exf5 5.Qh7+ Kf8 6.e6 Bd6 7.Qxg6 Be5 8.Rh5 Bg7 9.a5 Re7 10.Qh7 Rxe6 11.Rxf5+ Qf7 12.Rxf7+ Kxf7 13.Nd4 Rh6 14.Qf5+ Kg8 15.Qg4 Rf8 16.Nf5 Rxf5 17.Qxf5 Rf6 18.Qd7 Rf7 19.Qd8+ Kh7 20.axb6 Bxc3 21.bxc3 f3 22.Qh4+ Kg7 23.Qg3+ Kf8 24.Qf2 Ke8 25.Kb2 Kd7 26.h4 Kc6 27.h5 a5 28.Qg3 Kb5 29.Qg6 Rf8 30.Qg7
not sure if I agree...
[d]r3rnk1/1bq1bpp1/pp2p3/3pP2Q/P4p2/1NNB1R2/1PP4P/2KR4 w - -

Code: Select all

info depth 47 seldepth 86 multipv 1 score cp 261 nodes 31639743402 nps 10247676 hashfull 999 tbhits 1451945 time 3087504 pv f3h3

info depth 51 seldepth 93 multipv 1 score cp 273 nodes 34511142588 nps 11224044 hashfull 999 tbhits 24725083 time 3074751 pv d1g1 f7f5 g1g7 g8g7 f3h3 f8g6 d3f5 e6f5 h5h7 g7f8 e5e6 e7d6 h7g6 d6e5 h3h5 e5g7 a4a5 e8e7 g6h7 e7e6 h5f5 c7f7 f5f7 f8f7 b3d4 e6f6 h7h5 f7g8 a5b6 f4f3 c3d5 f6b6 h5f3 a8e8 d4f5 b7d5 f3d5 g8h7 f5g7 h7g7 d5d4 b6f6 c2c4 g7g6 c4c5 g6f5 d4f2 f5g6 f2g1 g6f5 g1f1 f5g6 f1c4 e8e6 c4d5 e6c6 c1c2 c6e6 h2h4 g6g7 c2c3 g7f8 h4h5 f8e7 d5g5 e7d7 c3c4 d7e8 c4d5 e8f8 d5d4 f8e8 d4c4 e8f7 c4b3 f7f8 b3c3 f6f3 c3b4 f3f6 g5d5 f8e7 d5g5 e7d7
Top
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: Cranky Walters

Post by MikeB »

MikeB wrote:
Vinvin wrote: ...
2) r3rnk1/1bq1bpp1/pp2p3/3pP2Q/P4p2/1NNB1R2/1PP4P/2KR4 w - - bm Rg1;
After a bit of learning, with 2 best moves, Rg1 is clearly better than Rh3 :

Code: Select all

Stockfish_170417_x64_modern:
 37/68	03:45	 2 282 368k	10 132k	+1,70	1.Rh3 f5 2.Re1 f3 3.Nd1 d4 4.Nxd4 Bb4 5.Rg1 Qxe5 6.Qh8+ Kf7 7.Rhg3 Ng6 8.Qh7 Qxd4 9.Qxg6+ Kf8 10.Qh7 Qf4+ 11.Kb1 Qh6 12.Qxh6 gxh6 13.Rg8+ Ke7 14.R8g7+ Kf6 15.Rxb7 Re7 16.Rxe7 Bxe7 17.Rf1 Rg8 18.Bxa6 Bd6 19.Rxf3 Bxh2 20.Rh3 Bf4 21.Nc3 Rg3 22.Rh5 Rg2 23.Bd3 Bg5 24.Bb5 Rg4 25.Ka2 Rh4 26.Rxh4 Bxh4 27.Be2 Ke5 28.Kb3
 37/68	03:45	 2 282 368k	10 132k	+2,59	1.Rg1 f5 2.Rxg7+ Kxg7 3.Rh3 Ng6 4.Bxf5 exf5 5.Qh7+ Kf8 6.e6 Bd6 7.Qxg6 Be5 8.Rh5 Bg7 9.a5 Re7 10.Qh7 Rxe6 11.Rxf5+ Qf7 12.Rxf7+ Kxf7 13.Nd4 Rh6 14.Qf5+ Kg8 15.Qg4 Rf8 16.Nf5 Rxf5 17.Qxf5 Rf6 18.Qd7 Rf7 19.Qd8+ Kh7 20.axb6 Bxc3 21.bxc3 f3 22.Qh4+ Kg7 23.Qg3+ Kf8 24.Qf2 Ke8 25.Kb2 Kd7 26.h4 Kc6 27.h5 a5 28.Qg3 Kb5 29.Qg6 Rf8 30.Qg7
not sure if I agree...
[d]r3rnk1/1bq1bpp1/pp2p3/3pP2Q/P4p2/1NNB1R2/1PP4P/2KR4 w - -

Code: Select all

info depth 47 seldepth 86 multipv 1 score cp 261 nodes 31639743402 nps 10247676 hashfull 999 tbhits 1451945 time 3087504 pv f3h3

info depth 51 seldepth 93 multipv 1 score cp 273 nodes 34511142588 nps 11224044 hashfull 999 tbhits 24725083 time 3074751 pv d1g1 f7f5 g1g7 g8g7 f3h3 f8g6 d3f5 e6f5 h5h7 g7f8 e5e6 e7d6 h7g6 d6e5 h3h5 e5g7 a4a5 e8e7 g6h7 e7e6 h5f5 c7f7 f5f7 f8f7 b3d4 e6f6 h7h5 f7g8 a5b6 f4f3 c3d5 f6b6 h5f3 a8e8 d4f5 b7d5 f3d5 g8h7 f5g7 h7g7 d5d4 b6f6 c2c4 g7g6 c4c5 g6f5 d4f2 f5g6 f2g1 g6f5 g1f1 f5g6 f1c4 e8e6 c4d5 e6c6 c1c2 c6e6 h2h4 g6g7 c2c3 g7f8 h4h5 f8e7 d5g5 e7d7 c3c4 d7e8 c4d5 e8f8 d5d4 f8e8 d4c4 e8f7 c4b3 f7f8 b3c3 f6f3 c3b4 f3f6 g5d5 f8e7 d5g5 e7d7
a better pv for f3h3

Code: Select all

nfo depth 52 seldepth 92 multipv 1 score cp 288 nodes 43710467383 nps 22076874 hashfull 993 tbhits 4372558 time 1979921 pv f3h3 f7f5 d1e1 f4f3 c3d1 d5d4 h5h8 g8f7 h3g3 g7g6 d3f5 e8d8 g3g6 e6f5 g6g7 f7e8 h8h5 e8d7 b3d4 d7c8 d4f5 c7d7 c1b1 d7d2 h5f7 d2d1 e1d1 d8d1 b1a2 b7d5 f7d5 d1d5 f5e7 c8d7 e7d5 d7c6 d5b4 c6c5 b4d3 c5d4 g7f7 d4e4 d3e1 e4e5 e1f3 e5d6 f7f6 d6c7 f6h6 f8d7 f3d4 c7b7 h2h4 a8f8 h4h5 f8f4 c2c3 d7f6 a2b3 f4f2 b3c4 b7c7 b2b4 c7d7 a4a5 b6a5 b4a5 d7c7 c4c5 f2f1 c3c4 f1f2 c5b4 c7d7
Top

Return to “Computer Chess Club: General Topics”