Mate in 13

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

Moderator: Ras

peter
Posts: 3477
Joined: Sat Feb 16, 2008 7:38 am
Full name: Peter Martan

Re: Mate in 13

Post by peter »

F.Huber wrote: ↑Mon Dec 15, 2025 7:53 pm Much faster (9 sec) with XPK order
Great, even from orginal starting position of the problem with AutoTurbo + SearchOrder XPK:

Code: Select all

Engine: ChestUCI Ver.5.2 (0 MB)
von Franz Huber
FEN: 5B2/5R2/2p3K1/8/2P1P1PP/1p1BpPP1/prp1p3/bk1bR3 w - -   (11+10)
Stellungs-Analyse:  C0/R0/K2/P2/X2   W:10/36
AutoTurbo-Suche nach Spezial-Matt [C1/R0/K2/P0/X0] in 15 ...  (Hash=2044MB)
AutoTurbo-Suche nach Spezial-Matt [C0/R0/K2/P0/X1] in 15 ...  (Hash=2044MB)
AutoTurbo-Suche nach Spezial-Matt [C0/R0/K2/P0/X2] in 15 ...  (Hash=2044MB)
AutoTurbo-Suche nach Spezial-Matt [C0/R0/K2/P0/X3] in 15 ...  (Hash=2044MB)
AutoTurbo-Suche nach Spezial-Matt [C0/R0/K2/P0/X4] in 15 ...  (Hash=2044MB)
AutoTurbo-Suche nach Spezial-Matt [C0/R0/K2/P1/X4] in 15 ...  (Hash=2044MB)
Matt in 15 gefunden !  (1 Lösung in 00:00)
 15/15	 0:00 	 +M15 	1.Lc5 Kc1 2.Lxe3+ Kb1 3.Lh6 c5 4.Kg5 Kc1 5.Kh5+ Kb1 6.g5 Kc1 7.g6+ Kb1 8.Tf4 Kc1 9.Tg4+ Kb1 10.f4 Kc1 11.f5+ Kb1 12.Tf4 Kc1 13.Tff1+ Kb1 14.Txd1+ 4062
Bester Zug: Lf8-c5 Zeit: 0:00.813 min  K/s: 4.062.600  Knoten: 243.756
:)
But the "secret" here seems to be AutoTurbo, which I hadn't tried before, in this mode with other one kinds of SeachOrder: still under 1 minute at KPX and again almost at once with KXP.

Edit: of course there do exist best settings for Gustav too to make it solve as well as at once:

Code: Select all

*****  Gustav 4.2 h   64 Bit  *****    Berechnung beendet am  15.12.2025  21:21

Matt in 15 Zügen, Weiß beginnt, ,   

Lösung gefunden bei folgenden Parametern: 
Fluchtfelder: 1, max. Zugzahl Schwarz: 1, Ausnahmen: 1, Gültig bis: 14
alle Lösungen suchen / keine Duale in Kurzvarianten anzeigen
Zugzahl unbekannt / Hashtabelle: 1024 MB

Lösezeit:    00,46 s        Zaehler: 328409

1.Lf8-c5 Kb1-c1 2.Lc5xe3 Kc1-b1 3.Le3-h6 Bc6-c5 4.Kg6-g5 Kb1-c1 5.Kg5-h5 Kc1-b1 6.Bg4-g5 Kb1-c1 7.Bg5-g6 Kc1-b1 8.Tf7-f4 Kb1-c1 9.Tf4-g4 Kc1-b1 10.Bf3-f4 Kb1-c1 11.Bf4-f5 Kc1-b1 12.Tg4-f4 Kb1-c1 13.Tf4-f1 Kc1-b1 14.Te1xd1 e2xd1D 15.Tf1xd1#
Peter.
Top
User avatar
Steve Maughan
Posts: 1307
Joined: Wed Mar 08, 2006 8:28 pm
Location: Florida, USA

Re: Mate in 13

Post by Steve Maughan »

The development version of Juggernaut, running on my AMD laptop with eight cores, finds a mate in 19 after 33 secs. Not too bad.

[d]8/5R2/2p3K1/8/2P1P1PP/1p1BBPP1/prp1p3/bk1bR3 w - -

Code: Select all

Engine: Juggernaut 1.34 [8-Threads] (4096 MB)
by Steve Maughan

21/47  0:01   +20.77   1.Rd7 c5 2.Bxe2 c1Q 3.Rdxd1 Rc2 
                       4.Bxc1 Rxc1 5.Rxc1+ Kb2 6.Rxa1 Kc2 
                       7.h5 b2 8.Rxa2 Kb3 9.Ra8 Kc2 10.h6 Kd2 
                       11.Ra2 Kxe1 12.Rxb2 Kf2 13.h7 Kxg3 
                       14.h8Q (53.454.728) 29954 

22/50  0:02   +26.08   1.Rd7 c5 2.Bxe2 c1Q 3.Rdxd1 Rc2 
                       4.Bxc1 Rxc1 5.Rxc1+ Kb2 6.Rxa1 Kc2 
                       7.e5 b2 8.Rxa2 Kb3 9.e6 Kxa2 10.e7 b1Q+ 
                       11.Rxb1 Kxb1 12.e8Q Kc1 13.h5 Kd2 
                       14.h6 (88.976.672) 35647 

23/50  0:03   +26.26   1.Rd7 c5 2.Bxe2 c1Q 3.Rdxd1 Rc2 
                       4.Bxc1 Rxc1 5.Rxc1+ Kb2 6.e5 Ka3 
                       7.Rxa1 b2 8.e6 bxa1Q 9.Rxa1 Kb2 
                       10.Rxa2+ Kxa2 11.e7 Kb2 12.e8Q Kc1 
                       13.h5 Kd2 14.h6 (137.149.717) 39594 

24/52  0:05   +27.92   1.Rd7 c5 2.Bxe2 c1Q 3.Rdxd1 Rc2 
                       4.Bxc1 Bb2 5.Be3+ Bc1 6.Bxc1 Rxc1 
                       7.Rxc1+ Kb2 8.h5 Ka3 9.Ra1 Kb4 10.h6 Ka5 
                       11.h7 Kb6 12.h8Q Kb7 13.e5 Kc7 14.e6 (261.988.894) 44195 

25/58  0:08   +33.77   1.Rd7 c5 2.Bxe2 c1Q 3.Rdxd1 Rc2 
                       4.Bxc1 Bb2 5.Be3+ Bc1 6.Bxc1 Rxc1 
                       7.Rxc1+ Kb2 8.h5 Ka3 9.Ra1 Kb4 10.h6 Ka5 
                       11.h7 Kb6 12.h8Q Kb7 13.e5 Kc7 14.f4 (397.675.933) 46983 

26/60  0:15   +34.25   1.Rd7 c5 2.Bxe2 c1Q 3.Rdxd1 Rc2 
                       4.Bxc1 Bb2 5.Be3+ Bc1 6.Bxc1 Rxc1 
                       7.Rxc1+ Kb2 8.h5 Ka3 9.Ra1 Kb4 10.h6 Ka5 
                       11.h7 Kb6 12.h8Q Kb7 13.Bd1 Kc7 
                       14.Bxb3 (750.945.503) 48244 

27/60  0:33   +M19     1.Rd7 c5 2.Bxe2 c1Q 3.Rdxd1 Rc2 
                       4.Bxc1 Bb2 5.Be3+ Bc1 6.Bxc1 Rxc1 
                       7.Rxc1+ Kb2 8.h5 Ka3 9.Ra1 Kb4 10.h6 Ka5 
                       11.h7 Kb6 12.h8Q Kc7 13.Bd1 b2 
                       14.Rxa2 (1.576.429.823) 47665
http://www.chessprogramming.net - Juggernaut & Maverick Chess Engine
Top
User avatar
towforce
Posts: 12693
Joined: Thu Mar 09, 2006 12:57 am
Location: Birmingham UK
Full name: Graham Laight

Re: Mate in 13

Post by towforce »

Steve Maughan wrote: ↑Tue Dec 16, 2025 9:51 am The development version of Juggernaut, running on my AMD laptop with eight cores, finds a mate in 19 after 33 secs. Not too bad.

I am pleased that it picked Rd7 as the first move, because that's actually the first move I came up with looking at it visually! :)
Human chess is partly about tactics and strategy, but mostly about memory
Top
peter
Posts: 3477
Joined: Sat Feb 16, 2008 7:38 am
Full name: Peter Martan

Re: Mate in 13

Post by peter »

Joerg Oster wrote: ↑Mon Dec 15, 2025 7:48 pm Matefish's Proof-Number Search is really quick at this one :D

Code: Select all

setoption name Hash value 2048
setoption name ProofNumberSearch value true
position fen 8/5R2/2p3K1/8/2P1P1PP/1p1BBPP1/prp1p3/bk1bR3 w - - 0 1
go mate 13
info string Starting Proof-Number Search ...
info time 572 multipv 1 depth 8 seldepth 25 nodes 1586752 nps 2774041 hashfull 0 tbhits 0 score cp 0 pv e3h6
info time 772 multipv 1 depth 12 seldepth 25 nodes 3097305 nps 4012053 hashfull 0 tbhits 0 score cp 0 pv e3h6
info time 972 multipv 1 depth 12 seldepth 25 nodes 4673861 nps 4808498 hashfull 0 tbhits 0 score cp 0 pv e3h6
info time 1172 multipv 1 depth 12 seldepth 25 nodes 6246886 nps 5330107 hashfull 0 tbhits 0 score cp 0 pv e3h6
info time 1372 multipv 1 depth 12 seldepth 25 nodes 7833956 nps 5709880 hashfull 0 tbhits 0 score cp 0 pv e3h6
info time 1572 multipv 1 depth 12 seldepth 25 nodes 9425133 nps 5995631 hashfull 0 tbhits 0 score cp 0 pv e3h6
info time 1772 multipv 1 depth 12 seldepth 25 nodes 11016562 nps 6217021 hashfull 0 tbhits 0 score cp 0 pv e3h6
info time 1972 multipv 1 depth 13 seldepth 25 nodes 12550701 nps 6364452 hashfull 0 tbhits 0 score cp 0 pv e3h6
info string Nodes statistics:   Total 5876393   Solved 307802   Proven 66   Disproven 307736   Recycled 138993
info string Success! Mate in 13 found!
info time 2376 multipv 1 depth 13 seldepth 25 nodes 15609465 nps 6569640 hashfull 0 tbhits 0 score mate 13 pv e3h6 c6c5 g6g5 b1c1 g5h5 c1b1 g4g5 b1c1 g5g6 c1b1 f7f4 b1c1 f4g4 c1b1 f3f4 b1c1 f4f5 c1b1 g4f4 b1c1 f4f1 c1b1 e1d1 e2d1n f1d1
bestmove e3h6 ponder c6c5
Great performance.
Even with my last one version compiled by Jim Ablett way back then about 1.5 years ago, and at original starting position of the composed puzzle, if I set KingMoves to 1:

Code: Select all

5B2/5R2/2p3K1/8/2P1P1PP/1p1BpPP1/prp1p3/bk1bR3 w - -

Engine: Matefish 3.2 BMI2(1024 MB)
von the Stockfish developers and J. Oster

29/29  0:01   +M15     1.Lc5 Kc1 2.Lxe3+ Kb1 3.Lh6 c5 4.Kg5 Kc1 
                       5.Kh5+ Kb1 6.g5 Kc1 7.g6+ Kb1 8.Tf4 Kc1 
                       9.Tg4+ Kb1 10.f4 Kc1 11.f5+ Kb1 
                       12.Tf4 Kc1 13.Tff1+ Kb1 14.Txd1+ (3.912.545) 3375 

Bester Zug: Lf8-c5 Zeit: 0:01.156 min  K/s: 3.375.793  Knoten: 3.912.545
Yet without KingMoves restricted (0):

Code: Select all

Starting Proof-Number Search ...
Failure! No mate found!
 29/29	 0:00 	 0.00 	1.Txd1+ (75) 0
Bester Zug: Te1xd1 Zeit: 0:00.718 min  K/s: 137  Knoten: 75
Time for a new compile anyhow
:)
(at mine overall hash is to be set to 1024Mb max. yet still too), regards
Peter.
Top

Return to “Computer Chess Club: General Topics”