I re-run some test suites with "study" flag and then it improved in all including mates, BTW just this was impressive:
[D]8/8/2b1N3/5p2/3B4/2K5/6kp/R7 w - - 0 1
Analysis by Stockfish MateFinder 64 POPCNT:
1.Rh1
= (0.27 ++) Depth: 30/36 00:00:06 52352kN, tb=517742
.
.
+- (122.44 ++) Depth: 40/58 00:00:58 562mN, tb=5808998
1.Rh1 Bf3 2.Kd3 Bb7 3.Nf4+ Kg3 4.Bf2+ Kxf4 5.Rxh2 Ke5 6.Bg3+ Kd5 7.Rh5 Ke6 8.Kd4 Bf3 9.Rh6+ Kd7 10.Rd6+ Ke7 11.Ke5 Be4 12.Bf4 Kf7 13.Rd1 Ke7 14.Bg5+ Kf7 15.Be3 Ke7 16.Rc1 Kf7 17.Rg1 Ke7 18.Rg7+ Kf8 19.Bh6 Ke8 20.Bf4 Kd8 21.Ke6 Kc8 22.Rd7 Bb1 23.Rc7+ Kb8 24.Rc1+ Kb7
New MateFinder Release
Moderators: hgm, Rebel, chrisw
-
Dann Corbit
- Posts: 12542
- Joined: Wed Mar 08, 2006 8:57 pm
- Location: Redmond, WA USA
Re: New MateFinder Release
The study flag is used to reduce LMR pruning in a special circumstance:Jouni wrote:I re-run some test suites with "study" flag and then it improved in all including mates, BTW just this was impressive:
[D]8/8/2b1N3/5p2/3B4/2K5/6kp/R7 w - - 0 1
Analysis by Stockfish MateFinder 64 POPCNT:
1.Rh1
= (0.27 ++) Depth: 30/36 00:00:06 52352kN, tb=517742
.
.
+- (122.44 ++) Depth: 40/58 00:00:58 562mN, tb=5808998
1.Rh1 Bf3 2.Kd3 Bb7 3.Nf4+ Kg3 4.Bf2+ Kxf4 5.Rxh2 Ke5 6.Bg3+ Kd5 7.Rh5 Ke6 8.Kd4 Bf3 9.Rh6+ Kd7 10.Rd6+ Ke7 11.Ke5 Be4 12.Bf4 Kf7 13.Rd1 Ke7 14.Bg5+ Kf7 15.Be3 Ke7 16.Rc1 Kf7 17.Rg1 Ke7 18.Rg7+ Kf8 19.Bh6 Ke8 20.Bf4 Kd8 21.Ke6 Kc8 22.Rd7 Bb1 23.Rc7+ Kb8 24.Rc1+ Kb7
Code: Select all
// Step 15. Reduced depth search (LMR). If the move fails high it will be
// re-searched at full depth.
if ( depth >= 3 * ONE_PLY
&& moveCount > 1
&& !captureOrPromotion
&& thisThread->maxPly > depth
&& !(depth >= 16 * ONE_PLY && ss->ply <= 3 * ONE_PLY)
&& move != ss->killers[0]
&& move != ss->killers[1])
{
ss->reduction = reduction<PvNode>(improving, depth, moveCount);
// Increase reduction for cut nodes and moves with a bad history
if ( (!PvNode && cutNode)
|| ( thisThread->history[pos.piece_on(to_sq(move))][to_sq(move)] < VALUE_ZERO
&& cmh[pos.piece_on(to_sq(move))][to_sq(move)] <= VALUE_ZERO))
ss->reduction += ONE_PLY;
// Decrease reduction for moves with a good history
if ( thisThread->history[pos.piece_on(to_sq(move))][to_sq(move)] > VALUE_ZERO
&& cmh[pos.piece_on(to_sq(move))][to_sq(move)] > VALUE_ZERO)
ss->reduction = std::max(DEPTH_ZERO, ss->reduction - ONE_PLY);
// Decrease reduction for moves that escape a capture
if ( ss->reduction
&& type_of(move) == NORMAL
&& type_of(pos.piece_on(to_sq(move))) != PAWN
&& pos.see(make_move(to_sq(move), from_sq(move))) < VALUE_ZERO)
ss->reduction = std::max(DEPTH_ZERO, ss->reduction - ONE_PLY);
if (study && ss->ply < depth / 2 - ONE_PLY)
ss->reduction = DEPTH_ZERO;
Depth d = std::max(newDepth - ss->reduction, ONE_PLY);
value = -search<NonPV>(pos, ss+1, -(alpha+1), -alpha, d, true);
doFullDepthSearch = (value > alpha && ss->reduction != DEPTH_ZERO);
ss->reduction = DEPTH_ZERO;
}
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.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
-
zullil
- Posts: 6442
- Joined: Tue Jan 09, 2007 12:31 am
- Location: PA USA
- Full name: Louis Zulli
Re: New MateFinder Release
Well, with 6-man syzygy endgame tables, plain old Stockfish-master solves this in seconds:Jouni wrote:I re-run some test suites with "study" flag and then it improved in all including mates, BTW just this was impressive:
[D]8/8/2b1N3/5p2/3B4/2K5/6kp/R7 w - - 0 1
Analysis by Stockfish MateFinder 64 POPCNT:
1.Rh1
= (0.27 ++) Depth: 30/36 00:00:06 52352kN, tb=517742
.
.
+- (122.44 ++) Depth: 40/58 00:00:58 562mN, tb=5808998
1.Rh1 Bf3 2.Kd3 Bb7 3.Nf4+ Kg3 4.Bf2+ Kxf4 5.Rxh2 Ke5 6.Bg3+ Kd5 7.Rh5 Ke6 8.Kd4 Bf3 9.Rh6+ Kd7 10.Rd6+ Ke7 11.Ke5 Be4 12.Bf4 Kf7 13.Rd1 Ke7 14.Bg5+ Kf7 15.Be3 Ke7 16.Rc1 Kf7 17.Rg1 Ke7 18.Rg7+ Kf8 19.Bh6 Ke8 20.Bf4 Kd8 21.Ke6 Kc8 22.Rd7 Bb1 23.Rc7+ Kb8 24.Rc1+ Kb7
Code: Select all
info depth 31 seldepth 36 multipv 1 score cp 12343 nodes 52346225 nps 19760749 hashfull 8 tbhits 725820 time 2649 pv a1h1 c6e4 d4f2 e4f3 c3d2 f3g4 d2e1 g2h1 e6f4 g4f3 f2a7 f3c6 e1f1 c6f3 f1f2 f3d1 a7e3 d1f3 f4e6 f3a8 e6g7 a8e4 g7h5 f5f4 h5f4-
Dann Corbit
- Posts: 12542
- Joined: Wed Mar 08, 2006 8:57 pm
- Location: Redmond, WA USA
Re: New MateFinder Release
How does it do on this one:zullil wrote:Well, with 6-man syzygy endgame tables, plain old Stockfish-master solves this in seconds:Jouni wrote:I re-run some test suites with "study" flag and then it improved in all including mates, BTW just this was impressive:
[D]8/8/2b1N3/5p2/3B4/2K5/6kp/R7 w - - 0 1
Analysis by Stockfish MateFinder 64 POPCNT:
1.Rh1
= (0.27 ++) Depth: 30/36 00:00:06 52352kN, tb=517742
.
.
+- (122.44 ++) Depth: 40/58 00:00:58 562mN, tb=5808998
1.Rh1 Bf3 2.Kd3 Bb7 3.Nf4+ Kg3 4.Bf2+ Kxf4 5.Rxh2 Ke5 6.Bg3+ Kd5 7.Rh5 Ke6 8.Kd4 Bf3 9.Rh6+ Kd7 10.Rd6+ Ke7 11.Ke5 Be4 12.Bf4 Kf7 13.Rd1 Ke7 14.Bg5+ Kf7 15.Be3 Ke7 16.Rc1 Kf7 17.Rg1 Ke7 18.Rg7+ Kf8 19.Bh6 Ke8 20.Bf4 Kd8 21.Ke6 Kc8 22.Rd7 Bb1 23.Rc7+ Kb8 24.Rc1+ Kb7Code: Select all
info depth 31 seldepth 36 multipv 1 score cp 12343 nodes 52346225 nps 19760749 hashfull 8 tbhits 725820 time 2649 pv a1h1 c6e4 d4f2 e4f3 c3d2 f3g4 d2e1 g2h1 e6f4 g4f3 f2a7 f3c6 e1f1 c6f3 f1f2 f3d1 a7e3 d1f3 f4e6 f3a8 e6g7 a8e4 g7h5 f5f4 h5f4
[d]r1bqrbk1/pp3ppp/1n1p4/3P1PP1/2p1N2Q/2PnN3/P2B2BP/4RRK1 w - - bm Re2;
Ellis version solves it in reasonable time.
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.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
-
MikeB
- Posts: 4889
- Joined: Thu Mar 09, 2006 6:34 am
- Location: Pen Argyl, Pennsylvania
Re: New MateFinder Release
Thank you for the SEG 11 fix.jhellis3 wrote:Repository Link: https://github.com/jhellis3/Stockfish/tree/mate_finder
-
Jouni
- Posts: 3293
- Joined: Wed Mar 08, 2006 8:15 pm
Re: New MateFinder Release
How about this?
[D]8/4p2p/2P1k3/4P1p1/1pp5/3p4/1P1P1PP1/1bN2K2 w - - 0 1
Analysis by Stockfish MateFinder 64 POPCNT:
1.Nb3 cxb3 2.g4 Bc2 3.Ke1 h6 4.f3 Bb1 5.Kf2 Bc2 6.Ke1
= (0.00) Depth: 17/32 00:00:00 851kN
1.Nb3 c3 2.bxc3 bxc3 3.Nd4+ Kd5 4.dxc3 d2 5.Ke2 Kxe5 6.c7 Bd3+ 7.Kxd2 Ba6 8.Nf3+ Kd6 9.Nxg5 h5 10.Ke3 Kxc7 11.Ne4 Kc6 12.Nd2 Kd6 13.g3 Bb7
+/- (0.96) Depth: 18/32 00:00:00 1622kN, tb=21
1.Nb3 c3 2.bxc3 bxc3 3.Nd4+ Kxe5 4.c7 cxd2 5.Nf3+ Kd6 6.Nxd2 Kxc7 7.Nxb1 Kc6 8.Ke1 Kc5 9.Kd2 Kd4 10.Nc3 h5 11.Nb5+ Kc4 12.Nc7 e5 13.Ne6 g4
+/- (1.09) Depth: 19/32 00:00:00 2367kN, tb=89
1.Nb3 cxb3 2.g4 Bc2 3.Ke1 h6 4.f3 Bb1 5.Kd1 Kxe5 6.c7 Kf4 7.c8Q Kxf3 8.Qf5+ Kg3 9.Kc1 Bc2 10.Qe6 Kf3 11.Qxh6 Kxg4
+- (5.19) Depth: 20/35 00:00:01 10247kN, tb=1792
1.Nb3 cxb3 2.g4 Bc2 3.Ke1 h6 4.f3 Bb1 5.Kd1 Bc2+ 6.Kc1 Kxe5 7.c7 Kf4 8.c8Q Kxf3 9.Qf5+ Kg3 10.Qe6 Kf3 11.Qxh6 Kxg4 12.Qe6+ Kf4 13.Qxe7 g4 14.Qxb4+ Kf3 15.Qb7+ Kf4 16.Qc7+ Kf3 17.Qc6+ Kf2 18.Qc5+ Kg3 19.Qf5 Kh3 20.Qe6 Kh4 21.Qe4 Kg3 22.Qe5+ Kh3 23.Qh5+ Kg3
+- (5.44) Depth: 21/44 00:00:01 12135kN, tb=1829
1.Nb3 cxb3 2.g4 Bc2 3.Ke1 h6 4.f3 Bb1 5.Kd1 Bc2+ 6.Kc1 Kxe5 7.c7 Kf4 8.c8Q Kxf3 9.Qf5+ Kg3 10.Qe6 Kf3 11.Qxh6 Kxg4 12.Qe6+ Kf4 13.Qxe7 g4 14.Qxb4+ Kf3 15.Qb7+ Kf4 16.Qc7+ Kf3 17.Qc6+ Kf2 18.Qc5+ Kg3 19.Qf5 Kh3 20.Qe6 Kh4 21.Qe4
+- (5.44) Depth: 22/44 00:00:01 12822kN, tb=1829
[D]8/4p2p/2P1k3/4P1p1/1pp5/3p4/1P1P1PP1/1bN2K2 w - - 0 1
Analysis by Stockfish MateFinder 64 POPCNT:
1.Nb3 cxb3 2.g4 Bc2 3.Ke1 h6 4.f3 Bb1 5.Kf2 Bc2 6.Ke1
= (0.00) Depth: 17/32 00:00:00 851kN
1.Nb3 c3 2.bxc3 bxc3 3.Nd4+ Kd5 4.dxc3 d2 5.Ke2 Kxe5 6.c7 Bd3+ 7.Kxd2 Ba6 8.Nf3+ Kd6 9.Nxg5 h5 10.Ke3 Kxc7 11.Ne4 Kc6 12.Nd2 Kd6 13.g3 Bb7
+/- (0.96) Depth: 18/32 00:00:00 1622kN, tb=21
1.Nb3 c3 2.bxc3 bxc3 3.Nd4+ Kxe5 4.c7 cxd2 5.Nf3+ Kd6 6.Nxd2 Kxc7 7.Nxb1 Kc6 8.Ke1 Kc5 9.Kd2 Kd4 10.Nc3 h5 11.Nb5+ Kc4 12.Nc7 e5 13.Ne6 g4
+/- (1.09) Depth: 19/32 00:00:00 2367kN, tb=89
1.Nb3 cxb3 2.g4 Bc2 3.Ke1 h6 4.f3 Bb1 5.Kd1 Kxe5 6.c7 Kf4 7.c8Q Kxf3 8.Qf5+ Kg3 9.Kc1 Bc2 10.Qe6 Kf3 11.Qxh6 Kxg4
+- (5.19) Depth: 20/35 00:00:01 10247kN, tb=1792
1.Nb3 cxb3 2.g4 Bc2 3.Ke1 h6 4.f3 Bb1 5.Kd1 Bc2+ 6.Kc1 Kxe5 7.c7 Kf4 8.c8Q Kxf3 9.Qf5+ Kg3 10.Qe6 Kf3 11.Qxh6 Kxg4 12.Qe6+ Kf4 13.Qxe7 g4 14.Qxb4+ Kf3 15.Qb7+ Kf4 16.Qc7+ Kf3 17.Qc6+ Kf2 18.Qc5+ Kg3 19.Qf5 Kh3 20.Qe6 Kh4 21.Qe4 Kg3 22.Qe5+ Kh3 23.Qh5+ Kg3
+- (5.44) Depth: 21/44 00:00:01 12135kN, tb=1829
1.Nb3 cxb3 2.g4 Bc2 3.Ke1 h6 4.f3 Bb1 5.Kd1 Bc2+ 6.Kc1 Kxe5 7.c7 Kf4 8.c8Q Kxf3 9.Qf5+ Kg3 10.Qe6 Kf3 11.Qxh6 Kxg4 12.Qe6+ Kf4 13.Qxe7 g4 14.Qxb4+ Kf3 15.Qb7+ Kf4 16.Qc7+ Kf3 17.Qc6+ Kf2 18.Qc5+ Kg3 19.Qf5 Kh3 20.Qe6 Kh4 21.Qe4
+- (5.44) Depth: 22/44 00:00:01 12822kN, tb=1829
Jouni
-
Damir
- Posts: 2803
- Joined: Mon Feb 11, 2008 3:53 pm
- Location: Denmark
- Full name: Damir Desevac
Re: New MateFinder Release
Hi Jouni,
Can you and the rest of SF Team use Stockfish Matefinder ideas to improve Stockfish ?
Can you and the rest of SF Team use Stockfish Matefinder ideas to improve Stockfish ?
-
Dann Corbit
- Posts: 12542
- Joined: Wed Mar 08, 2006 8:57 pm
- Location: Redmond, WA USA
Re: New MateFinder Release
The current version takes about twice as long to solve it as a previous version, but still solves it in a reasonable time on a mediocre machine:Dann Corbit wrote:How does it do on this one:zullil wrote:Well, with 6-man syzygy endgame tables, plain old Stockfish-master solves this in seconds:Jouni wrote:I re-run some test suites with "study" flag and then it improved in all including mates, BTW just this was impressive:
[D]8/8/2b1N3/5p2/3B4/2K5/6kp/R7 w - - 0 1
Analysis by Stockfish MateFinder 64 POPCNT:
1.Rh1
= (0.27 ++) Depth: 30/36 00:00:06 52352kN, tb=517742
.
.
+- (122.44 ++) Depth: 40/58 00:00:58 562mN, tb=5808998
1.Rh1 Bf3 2.Kd3 Bb7 3.Nf4+ Kg3 4.Bf2+ Kxf4 5.Rxh2 Ke5 6.Bg3+ Kd5 7.Rh5 Ke6 8.Kd4 Bf3 9.Rh6+ Kd7 10.Rd6+ Ke7 11.Ke5 Be4 12.Bf4 Kf7 13.Rd1 Ke7 14.Bg5+ Kf7 15.Be3 Ke7 16.Rc1 Kf7 17.Rg1 Ke7 18.Rg7+ Kf8 19.Bh6 Ke8 20.Bf4 Kd8 21.Ke6 Kc8 22.Rd7 Bb1 23.Rc7+ Kb8 24.Rc1+ Kb7Code: Select all
info depth 31 seldepth 36 multipv 1 score cp 12343 nodes 52346225 nps 19760749 hashfull 8 tbhits 725820 time 2649 pv a1h1 c6e4 d4f2 e4f3 c3d2 f3g4 d2e1 g2h1 e6f4 g4f3 f2a7 f3c6 e1f1 c6f3 f1f2 f3d1 a7e3 d1f3 f4e6 f3a8 e6g7 a8e4 g7h5 f5f4 h5f4
[d]r1bqrbk1/pp3ppp/1n1p4/3P1PP1/2p1N2Q/2PnN3/P2B2BP/4RRK1 w - - bm Re2;
Ellis version solves it in reasonable time.
Code: Select all
Searching: r1bqrbk1/pp3ppp/1n1p4/3P1PP1/2p1N2Q/2PnN3/P2B2BP/4RRK1 w - - 0 1
infinite: 0 ponder: 0 time: 0 increment: 0 moves to go: 0
1 +0.15 00:00 49 Nf6+ gxf6
2 +0.34 00:00 215 Rb1 Kh8
3 +0.45 00:00 397 Rb1 Kh8 f6
4 +0.34 00:00 1017 Re2 Kh8 f6 g6
5 +0.45 00:00 1405 Rb1 Kh8 f6 g6 Nf2
6 +0.58 00:00 1816 Rb1 Kh8 f6 g6 Nf2 Qc7 Be4
7 +0.45 00:00 2323 Rb1 Kh8 f6 g6 Nf2 Qc7 Nxd3
8 +0.32 00:00 5469 Rb1 Kh8 f6 g6 Nf2 Qc7 Nxd3 cxd3
9 +0.44 00:00 10295 Rb1 Kh8 Qg3 Qc7 g6 h6 gxf7 Qxf7 f6
10 +0.60 00:00 15761 Rb1 Kh8 Qh5 Bd7 g6 fxg6 fxg6 h6 Nf5 Re5 Nexd6
11 +0.83 00:00 62401 Rb1 Bd7 Ng4 Nxd5 Rxb7 Ne5 Nxe5 Rxe5 f6 g6 Rfb1 a5
12 +0.63 00:00 131370 Rb1 Bd7 Ng4 Nxd5 Rxb7 Ne5 Nxe5 Rxe5 f6 g6 Rfb1 Qc8 Bf3 Nb6
13 +0.73 00:00 147727 Rb1 Bd7 Ng4 Nxd5 Rxb7 Ne5 Nxe5 Rxe5 Ng3 Nb6 Rxd7 Nxd7 Bxa8 Qxa8
14 +0.74 00:00 237065 Rb1 Ne5 Be1 Kh8 Bg3 Nd3 Rf3 Rb8 f6 g6 Nxc4 Nxc4 Rxd3 Qa5
15 +0.74 00:01 781544 Rb1 Ne5 Qh5 Na4 a3 Kh8 g6 fxg6 fxg6 h6 Nf5 Bxf5 Qxf5 Rb8 Be3 Qa5
16 +0.69 00:01 1102K Rb1 Ne5 Be1 Nd3 Qh5 Nxe1 Rbxe1 Re5 Rf4 h6 Bf3 Rxe4 Rxe4 Qxg5+ Qxg5 hxg5 Re8
17 +0.76 00:08 5642K Rb1 Ne5 Qg3 Na4 h4 Nc5 Nxc5 dxc5 f6 Qd6 fxg7 Bxg7 Be4 Qa6 h5 Qxa2 Qg2 Bf8 g6
18 +0.93 00:11 7466K Rb1 Kh8 Qh5 Re5 Ng3 g6 Qh4 Bxf5 Nexf5 gxf5 Nxf5 Qd7 Be4 Nxd5 Nxd6 Rxe4 Nxe4 b5 Nf6 Nxf6 gxf6
19 +0.99 00:13 8920K Rb1 Re5 Ng4 Nxd5 Rb5 Rxe4 Bxe4 Nc7 f6 g6 Rbb1 Nc5 Nh6+ Bxh6 Qxh6 Qf8 Qxf8+ Kxf8 Bg2 Bf5 Rb2 Bd3 Re1 N7e6 h4 Rb8 Be3
20 +0.79 00:16 10497K Rb1 Nd7 Ng4 N3c5 Nxc5 Nxc5 Qh5 Nd3 g6 fxg6 fxg6 h6 Bxh6 Bxg4 Qxg4 Qc8 Qxc8 Raxc8 Bg5 Nc5 Rf3 Re5 Bh4 Rh5 Bf2 Rg5 Rg3 Rxg3 Bxg3 Re8 Bf3
21 +0.79 00:16 10724K Rb1 Nd7 Ng4 N3c5 Nxc5 Nxc5 Qh5 Nd3 g6 fxg6 fxg6 h6 Bxh6 Bxg4 Qxg4 Qc8 Qxc8 Raxc8 Bg5 Re5 Bh4 Nc5 Rf3 Rh5 Bf2 Rg5 Rg3 Rxg3
22 +0.72 00:25 16797K Rb1 Nd7 Nxc4 N7c5 Be3 b6 Ncd2 Re5 f6 g6 Bd4 Rf5 c4 Rxf1+ Bxf1 Qd7 h3 Nxe4 Nxe4 Ne5 Qg3 Qa4 Bxe5 dxe5 Qxe5 Bd7
23 +0.65 02:04 81769K Rd1 Ne5 h3 Nd3 Rb1 Re5 Ng3 Rb8 Nxc4 Nxc4 Qxc4 Nc5 f6 gxf6 gxf6 Qd7 Bf4 Re8 Qd4 b6 Be3 Re5 c4 Bb7 Bf4
24 +0.60 03:16 127356K Rb1 Nd7 Nxc4 N7c5 Na3 Re5 Ng3 Bd7 Nc4 Re8 Qh5 b5 g6 fxg6 fxg6 h6 Ne3 Qg5 Qxg5 hxg5 Nef5 Ne5 Nd4 a6 Ne6 Rac8 Nxc5 Rxc5 Ne4 Rc4 Bxg5 Nxg6
25 +0.64 03:29 135947K Rb1 Nd7 Nxc4 N7c5 Na3 Bd7 Nb5 Nxe4 Bxe4 Nc5 Bc2 Re2 Rf2 Rxf2 Qxf2 Bxb5 Rxb5 Qe8 Rb4 Qe5 f6 g6 Rd4 Re8 h4 Qe2 Qxe2 Rxe2 Kf1
26 +0.71 05:32 214578K Rb1 Nd7 Ng4 N3e5 h3 b6 Qh5 Bb7 Rbd1 Nxg4 hxg4 Re5 Qh1 Nc5 Bf4 Qe8 Nxc5 bxc5 Rb1 Ba6 Bxe5 Qxe5 Rf2 Qe3 Qh5 h6 g6 f6
27 +0.84 06:12 237366K Rb1 Nd7 Ng4 N3e5 h3 b6 Qh5 Bb7 Bf4 Nxg4 hxg4 Qc7 Rbd1 Rac8 Be3 Re5 Qh4 Nc5 Bxc5 dxc5 g6 h6 g5 Rxd5 Rxd5 Bxd5 gxh6 Bxe4 Bxe4 gxh6 Bd5
28 +0.90 08:37 331656K Rb1 Nd7 Ng4 N3e5 f6 Nxg4 fxg7 Bxg7 Qxg4 Ne5 Qh5 Bg4 Qh4 h5 Rxb7 Qc8 Nf6+ Bxf6 gxf6 Ng6 Qg5 Re2 Qh6 Qf8 Qxf8+ Rxf8 Rf2 Rxf2 Kxf2 Re8 Bf1 Re5 Bxc4 Rf5+ Ke3 Rf3+ Kd4 Ne5 Rxa7 Rf2 Ra8+ Kh7 Be3 Rxh2
29 +0.80 22:25 881100K Re2 Re5 Qg3 Bd7 g6 fxg6 fxg6 h6 Rf7 Qc8 Kh1 Qe8 Nxc4 Nxc4 Qxd3 Nxd2 Rxd2 Be7 Rdf2 Kh8 c4 b6 Rf1 Bg4 Kg1 Qc8 h3 Bd7 Qb3 Qg8 Qa3 b5 c5 dxc5 Nxc5
30 +1.12 36:47 1455M Re2 Ne5 Rf4 Nbd7 Qh5 g6 Qh4 b5 Nc2 Bb7 Nd4 Bxd5 Nxd6 Bxg2 Nxe8 Qxe8 Kxg2 Qe7 f6 Qc5 Rfe4 Qd5 Qg3 Bc5 Nxb5 Rb8 a4 Rd8 Nd4 Nd3 Kg1 Nf8 Kh1 Ne6 Nxe6 fxe6
31 +1.09 49:05 1942M Re2 Ne5 Rf4 Nbd7 Qh5 g6 Qh4 b5 Ref2 a6 Bh3 Kh8 Ng4 Nxg4 Rxg4 a5 Rgf4 Rxe4 Rxe4 Ne5 Be3 Bxf5 Bxf5 gxf5 Rxf5 Qd7 Qh3 Ng6 Bd4+ Kg8 Re3 Re8 Rxe8 Qxe8 Rf1 b4 Rf2 Qe1+ Qf1 Qe4 Rxf7
32 +1.44 1:06:29 2642M Re2 Ne5 Rf4 Nbd7 Qh5 g6 Qh4 a6 Ref2 b5 fxg6 fxg6 Bh3 Kh8 Be6 Ra7 Bf7 Re7 Nxd6 Bg7 Be6 Rxe6 dxe6 Nc5 Rd4 Nxe6 Nf7+ Rxf7 Rxd8+ Nxd8 Qg3 Rxf2 Kxf2 Ne6 Nd5 Nc5 Qh4 Bb7 Nf6 Bxf6 gxf6 Ne4+ Ke3
33 +1.32 1:26:38 3424M Re2 Ne5 Qh5 Nbd7 Rf4 g6 Qh4 Rb8 Ref2 b5 Ng4 Nxg4 Rxg4 Rxe4 Bxe4 Rb7 Rg3 Ne5 Rh3 h5 gxh6 Qxh4 Rxh4 Be7 h7+ Kh8 f6 Bf8 Kg2 a5 a3 Rb8 Rf1 Bg4 Kg3 Be2 Re1 Bd3 h3
34 +1.09 1:33:10 3684M Re2 Ne5 Qh5 Nbd7 Rf4 g6 Qh4 b5 Ref2 Rb8 Bh3 Kh8 Ng4 Nxg4 Rxg4 Rxe4 Rxe4 Ne5 Be3 Bxf5 Bxf5 gxf5 Rxf5 a5 Bd4 Bg7 Qh3 b4 Rh4 bxc3 Rxh7+ Kg8 Bxc3 Rb1+ Kg2 Qb6 Rf2 Qc5 Qf5 Qxd5+ Kh3 Qd3+ Qxd3 cxd3
35 +1.64 3:37:03 8587M Re2 Ne5 Bh1 Nbd7 Rg2 Kh8 Rg3 Nc5 Nxc5 dxc5 d6 Bxd6 Be4 Nd7 Rh3 Nf8 Nxc4 Bc7 Rd3 Bd7 f6 Rxe4 Qxe4 gxf6 Nd6 Bxd6 Rxd6 Qe8 Qxb7 Rb8 Qd5 Qe2 Qf3 Qxf3 Rxf3 Be6 Rxf6 Bxa2 Be3 Be6 Bxc5 a5 Bd4 Kg8 Rb6 Rxb6 Bxb6
36 +1.64 4:12:16 10022M Re2 Re5 Bh3 Rxe4 Qxe4 Ne5 Rg2 Nbd7 Nxc4 Nxc4 Qxc4 Ne5 Qb3 b6 Rg3 Ba6 Rf4 Qc7 Rd4 Bc4 Qc2 Qc5 Be3 Qa5 Rd2 Re8 Bg2 Bb5 Rd4 Rc8 Bf4 g6 f6 Bc4 h4 Qxa2 Qxa2 Bxa2 Rd2 Bc4 Bxe5 dxe5 d6
37 +1.55 4:48:09 11494M Re2 Re5 Bh3 Rxe4 Qxe4 Ne5 Rg2 Nbd7 Nxc4 Nxc4 Qxc4 Ne5 Qb3 b6 Rg3 Ba6 Rf4 Qc7 Rd4 Bc4 Qc2 g6 f6 Qb7 Be3 Qa6 Rd2 Bd3 Qd1 Bc4 Bf1 Qa5 Bxc4 Nxc4 Rf2 Re8 Bf4 Qc5 Kg2 h5 gxh6 Kh7 Rf1
38 +1.55 5:29:59 13216M Re2 Re5 Bh3 Rxe4 Qxe4 Ne5 Rg2 Nbd7 Nxc4 Nxc4 Qxc4 Ne5 Qb3 b6 Rg3 Ba6 Rf4 Rc8 Be3 f6 Qa4 Bd3 gxf6 Qxf6 Bd4 Rc7 Bf1 Bxf1 Rxf1 Be7 Qe8+ Qf8 Qxf8+ Kxf8 Rh3 Kg8 Kg2 Bf6 Re1 Rc4 Rg3 Ra4 Re2 Ra5
39 +1.48 8:15:42 20044M Re2 Re5 Bh3 Rxe4 Qxe4 Nd7 Nxc4 N7e5 Nxe5 Nxe5 Bf4 Bd7 Re3 Rc8 Rd1 Nc4 Rg3 b5 Rd4 Qb6 Qf3 Ne5 Qf2 Qa5 Qd2 Nc4 Qg2 Ne5 Qb2 Nc4 Qb1 Nb6 Rdd3 Qa4 Bd2 Qe4 Qe1 Qxe1+ Bxe1 Re8 Bf2 Re2 Bxb6 axb6 Rg2 Re1+ Kf2 Ra1
40 +1.6810:13:33 25011M Re2 Re5 Bh3 Rxe4 Qxe4 Nd7 Qxc4 N7e5 Qh4 Bd7 Rg2 g6 c4 h5 Bc3 Bg7 Rd2 Rc8 f6 Bf8 Bd4 Nc5 Rc2 Qc7 Bxd7 Qxd7 Qg3 Re8 Qg2 Qa4 Rc3 b5 Qc2 Qxc2 Rxc2 bxc4 Kg2 Ncd3 h4 Rc8 Rc3 Rc7 Ra3 Nb4 Rb1 Nbd3 Rxa7 Rxa7
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.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
-
jhellis3
- Posts: 546
- Joined: Sat Aug 17, 2013 12:36 am
-
MikeB
- Posts: 4889
- Joined: Thu Mar 09, 2006 6:34 am
- Location: Pen Argyl, Pennsylvania
Re: New MateFinder Release
Vey nice. What the best way/settings to use this and for what type of positions?jhellis3 wrote:Updates:
1) Sync with latest SF.
2) Move to full hash key.
3) Increase max depth to 250.
4) Improve mate detection.
https://dl.dropboxusercontent.com/u/114 ... eFinder.7z
The archive includes the source and two very basic windows compiles.
If anybody wants a Mac version, PM me and I will post a link. Let me know what type of CPU or Model you have too so I can make sure it will run on your Mac.