Not logged inRybka Chess Community Forum
Up Topic The Rybka Lounge / Test Positions / Speaking of checkmate verifications...
- - By user923005 (****) Date 2017-07-13 17:29
I have a small set where the solutions found are better than the proposed solutions:

4R3/8/5p2/2P2N2/5P2/B2pNkPp/1K1prn1p/3BbRnr w - - acd 21; acs 470; bm Rd8; c2 "#11;
00:00 @ C2/R-6/K2/P1/X1;"; ce 32748; dm 10; id "ChestDB.4434"; pm Rd8; pv Rd8 Ke4 c6 Rxe3 Rxf2 Bxf2 Nd6+ Kd5 c7 Re6 c8Q Bd4+ Kb3 Rxd6 Qb7+ Ke6 Rxd6+ Kf5 Qh7+;
1r2brk1/pp3p1p/3p1Q2/2pBp3/P2nP3/2NP2P1/1qP2R1P/5RK1 w - - acd 21; acs 468; bm Nb5; c2 "#12;
00:13 @ C0/R-3/K3/P9/X38;"; ce 32748; dm 10; id "ChestDB.4930"; pm Nb5; pv Nb5 Rc8 Kh1 Qa1 Rxa1 Bxb5 Bxf7+ Rxf7 Qxf7+ Kh8 Qf6+ Kg8 Qg5+ Kh8 Rf7 Ne6 Qh6 Kg8 Qxh7+;

8/1p3N1R/1p6/bP2p1p1/rkPpP1K1/p2P2P1/P2P4/8 w - - acd 23; acs 29; bm Nxe5; c2 "#12;
00:00 @ C2/R0/K2/P1/X1;"; ce 32746; dm 11; id "yacpdb id 192778"; pm Nxe5; pv Nxe5 Kc5 Nd7+ Kd6 Kf5 Rxc4 dxc4 g4 e5+ Kc7 Nf6+ Kb8 Rh8+ Ka7 Nd5 Bb4 Nc7 Bf8 Rxf8 d3 Ra8+;

r3k2r/1q1nbp1p/p2pp1p1/1p2P3/1P1NN3/P6Q/2P3PP/4RR1K w - - acd 25; acs 569; bm Rxf7; c2 "#14;
"; ce 32744; dm 12; id "ChestDB.6534"; pm Rxf7; pv Rxf7 Kd8 Rxe7 Ra7 Nxe6+ Kc8 Qxh7 Rd8 Nxd6+ Kb8 Nxd8 Qc8 Nxc8 Kxc8 Ne6 Rc7 Rd1 Kb8 Rexd7 Rxc2 Ra7 Kc8 Qb7+;

4KB2/1PpB4/2P3N1/p1pk2pp/3b4/2pPpN2/b1P1P2P/qr5n w - - acd 27; acs 520; bm h4; c2 "#15;
--:-- @ C0/R-4/K3/P9/X35;"; ce 32742; dm 13; id "ChestDB.6087"; pm h4; pv h4 Bf6 hxg5 Rb4 gxf6 Qg1 Ne7+ Kd6 f7 Bxf7+ Kxf7 Qg5 Nxg5 Rf4+ Kg7 a4 b8Q Ng3 Ng6+ Rxf8 Qxf8+ Kd5 Qf7+ Kd6 Qe6+;

3R1nqb/B2n1prb/4pPp1/1p1kN1P1/1P2rP1K/2p5/2P1P3/7B w - - acd 27; acs 248; bm e3; c2 "#23;
00:00 @ C2/R0/K2/P1/X1;"; ce 32742; dm 13; id "yacpdb id 67687"; pm e3; pv e3 Kd6 Bb8+ Kd5 Nd3 Kc6 Bxe4+ Kb6 Nc5 e5 fxe5 Nxb8 Rxb8+ Kc7 Na6+ Kd7 Rb7+ Kc8 Rc7+ Kd8 Bb7 Ne6 Rc8+ Kd7 Nb8+;

8/1p5p/1pp3p1/1k6/1p5p/1P5P/6pP/Q5Kb w - - acd 31; acs 60; bm Qg7; c2 "#19;
--:-- @ C2/R0/K4/P4/X7;"; ce 32738; dm 15; id "yacpdb id 344422"; pm Qg7; pv Qg7 Ka6 Qxh7 g5 Qg7 Ka7 Qxg5 c5 Qxh4 b5 Qf2 Kb6 h4 Kc6 h5 c4 h6 cxb3 h7 b2 Qxb2 Kd5 h8=Q Kc6 Qbf6+ Kd5 Qd4+ Ke6 Qhf6+;

1n1k4/3pN1N1/p1pP4/P1P1p3/KBP1p3/P1P1P3/8/8 w - - acd 35; acs 23; bm Ng6; c2 "#18;
00:15 @ C2/R0/K3/P1/X2;"; ce 32734; dm 17; id "yacpdb id 188637"; pm Ng6; pv Ng6 Kc8 Nh5 Kb7 Nxe5 Kc8 Nf6 Kd8 Nfxd7 Nxd7 Nxc6+ Ke8 Ne7 Nxc5+ Bxc5 Kd7 Ng6 Ke6 Kb4 Kd7 Nf8+ Ke8 d7+ Kd8 Bb6+ Ke7 d8=Q+ Kf7 Qd7+ Kg8 Bc5 Kh8 Qh7+;

3q3r/4kp1n/1P2P1pp/Q2Pp3/K3B3/1PP4P/1P2r3/4B3 w - - acd 47; acs 375; bm Qa7+; c2 "#28;
"; ce 32722; dm 23; id "ChestDB.5846"; pm Qa7+; pv Qa7+ Kd6 c4 Kc5 b7+ Qb6 Bb4+ Kd4 Qxb6+ Kxe4 exf7 Rg2 b8Q Rxb8 Qxb8 Rf2 Qg8 Rxf7 Qxf7 Ng5 Qxg6+ Kf3 Qxh6 Ne4 Qh5+ Kf2 Qf5+ Ke3 Qg4 Nf6 Qg5+ Ke2 Qxf6 Ke3 Qf1 Ke4 Qf2 Kd3 d6 Ke4 d7 Kd3 d8Q+ Ke4 Qd5+;

I have 6381 positions where the proposed solution matches my analysis (at least as far as depth to solution).

And I have 137 that need work (the best solution that I have found is not as good as the proposed solution).
Attachment: mate-verifications.7z - Verified checkmates (276k)
Attachment: mates-that-need-work.epd - Mates that need work (solution found by computer is not as good as the proposed solution). (28k)
Parent - - By Akbarfan (***) Date 2017-07-13 21:01
In your first Pos on mates-that-need-work.epd
8/8/B7/7N/8/p7/Pp1p4/1K1k2B1 w - -

your last five moves in pv are: Bd3 Ng2, Nh3 Ne3, Bf2+;
Correct is (like Chest) : Bd3 Ng2, Nxg2#

So mate in 13 is correct.

Chest is usually correct in its analysis.
Parent - - By user923005 (****) Date 2017-07-13 22:38
If you can post your log files, it would help me to store the results in my database.

I find that ChestUCI is difficult to use properly.  When I set it to 'use database' it does no analysis.  When I set it to 'Auto-turbo' with database hints, it seems to ignore the database setting.  What I would like is a setting where it uses the parameters that were successful and also actually performs the search.  I want the analysis and the pv.

It there a simple setting for an entry that is already in the database?  I find the ChestUCI documentation almost incomprehensible, though I suspect that the German version is much better.

I use the command line version of Chest all the time (Chest319) but it is not very useful for long mates like a mate in 13.
For direct mates in 7 or less, though it is fabulous.
Parent - By Akbarfan (***) Date 2017-07-14 00:03 Edited 2017-07-14 00:10
I use ChestUCI 5.2 as Uci-Engine in Shredder GUI

Copy this from ChestUCI Database: "8/8/B7/7N/8/p7/Pp1p4/1K1k2B1 w - - bm #13; 00:00 @ C0/R0/K2/P2/X8;"
(without " "  of course)   and paste it to ShrGUI.

As result i have this:

Engine: ChestUCI Ver.5.2 (0 MB)
von Franz Huber
8/8/B7/7N/8/p7/Pp1p4/1K1k2B1 w - -
Stellungs-Analyse:  C0/R0/K2/P1/X1   W:3/18
Parameter im Clipboard gefunden:  #13; 00:00 @ C0/R0/K2/P2/X8;
Suche nach Spezial-Matt [C0/R0/K2/P2/X8] in 13 ...  (Hash=1024MB)
d=1  t=0.00s [1.00]  n=0kN [1.00]   Hash: spd=1.00  in=1  out=0
d=2  t=0.00s [1.00]  n=0kN [1.00]   Hash: spd=1.00  in=1  out=0
d=3  t=0.00s [1.00]  n=0kN [45.00]   Hash: spd=0.99  in=15  out=0
d=4  t=0.00s [1.00]  n=1kN [8.68]   Hash: spd=1.28  in=47  out=0
d=5  t=0.00s [1.00]  n=2kN [1.97]   Hash: spd=2.76  in=62  out=0
d=6  t=0.00s [1.00]  n=2kN [1.47]   Hash: spd=3.29  in=78  out=0
d=7  t=0.00s [1.00]  n=3kN [1.17]   Hash: spd=4.46  in=103  out=0
d=8  t=0.00s [1.00]  n=4kN [1.53]   Hash: spd=8.14  in=150  out=0
d=9  t=0.00s [1.00]  n=5kN [1.26]   Hash: spd=10.75  in=177  out=0
d=10  t=0.00s [1.00]  n=6kN [1.22]   Hash: spd=15.53  in=223  out=0
d=11  t=0.00s [1.00]  n=7kN [1.16]   Hash: spd=19.36  in=266  out=0
d=12  t=0.00s [1.00]  n=9kN [1.30]   Hash: spd=22.87  in=438  out=0
d=13  t=0.00s [1.00]  n=10kN [1.03]   Hash: spd=23.02  in=461  out=0
Suche abgeschlossen ...  (Zeit=0.00s)
Matt in 13 gefunden !  (1 Lösung in 00:00)
13/13   0:00    +M13   1.Sf4 Ke1 2.Sg2+ Kd1 3.Se3+ Ke1 4.Sc2+ Kd1 5.Sxa3 Ke1 6.Sc2+ Kd1 7.Se3+ Ke1 8.Sg2+ Kd1 9.Sf4 Ke1 10.Kxb2 d1S+ 11.Kc1 Se3 12.Ld3 Sc4 13.Sg2+
Bester Zug: Sh5-f4 Zeit: 0:00.125 minKnoten: 9.752
Up Topic The Rybka Lounge / Test Positions / Speaking of checkmate verifications...

Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill