Mate in 4...

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

Moderator: Ras

zullil
Posts: 6442
Joined: Tue Jan 09, 2007 12:31 am
Location: PA USA
Full name: Louis Zulli

Re: Mate in 4...

Post by zullil »

Crafty-23.0 sees it instantly:

Code: Select all

Crafty v23.0 (8 cpus)

White(1): book off
book file disabled.
White(1): ponder off
pondering disabled.
White(1): setboard 8/8/2B3N1/5p2/6p1/6pk/4K2b/7r w - - 0 3
White(1): st 300
search time set to 300.00.
White(1): display 

       +---+---+---+---+---+---+---+---+
    8  |   | . |   | . |   | . |   | . |
       +---+---+---+---+---+---+---+---+
    7  | . |   | . |   | . |   | . |   |
       +---+---+---+---+---+---+---+---+
    6  |   | . |-B-| . |   | . |-N-| . |
       +---+---+---+---+---+---+---+---+
    5  | . |   | . |   | . |<P>| . |   |
       +---+---+---+---+---+---+---+---+
    4  |   | . |   | . |   | . |<P>| . |
       +---+---+---+---+---+---+---+---+
    3  | . |   | . |   | . |   |<P>|<K>|
       +---+---+---+---+---+---+---+---+
    2  |   | . |   | . |-K-| . |   |<B>|
       +---+---+---+---+---+---+---+---+
    1  | . |   | . |   | . |   | . |<R>|
       +---+---+---+---+---+---+---+---+
         a   b   c   d   e   f   g   h

White(1): go
              time limit 5:00 (+0.00) (5:00) [easy move]
              depth   time  score   variation (1)
starting thread 1
starting thread 2
starting thread 3
starting thread 4
starting thread 5
starting thread 6
starting thread 7
              time=0.01  mat=-4  n=80921  fh=89%  nps=1.0M
              ext-> check=2K qcheck=3K reduce=40K/9K
              predicted=0  evals=48K  50move=0  EGTBprobes=0  hits=0
              SMP->  splits=34  aborts=3  data=9/512  elap=0.01
terminating SMP processes.

mate in 4 moves.


White(1): Bg2+ 
              time used:   0.01
Top
User avatar
Roman Hartmann
Posts: 295
Joined: Wed Mar 08, 2006 8:29 pm

Re: Mate in 4...

Post by Roman Hartmann »

My own engine sees the mate in 4 almost instantly but I just noted that Ruffian 1.0.1/2.0.2/2.1 doesn't see the mate at all.

best regards
Roman

Output of Ruffian 2.1:

Code: Select all

New game
8/8/2B3N1/5p2/6p1/6pk/4K2b/7r w - - 0 1

Analysis by Ruffian 2.1.0:

1.Bxh1 
  ³  (-0.47)   Depth: 1/1   00:00:00
1.Nf4+ Kh4 2.Bxh1 
  =  (0.18)   Depth: 1/5   00:00:00
1.Nf4+ Kh4 2.Bxh1 g2 3.Nxg2+ Kg5 
  ³  (-0.64)   Depth: 2/8   00:00:00
1.Nf4+ Kh4 2.Bxh1 g2 3.Nxg2+ Kg5 4.Ne3 
  ³  (-0.34)   Depth: 3/10   00:00:00
1.Nf4+ Kh4 2.Bxh1 g2 3.Nxg2+ Kg5 4.Ne3 Be5 
  ³  (-0.69)   Depth: 4/11   00:00:00
1.Nf4+ Kh4 2.Bxh1 g2 3.Nxg2+ Kg5 4.Ne3 f4 5.Bd5 fxe3 6.Kxe3 
  ³  (-0.45)   Depth: 5/12   00:00:00
1.Nf4+ Kh4 2.Bxh1 g2 3.Nxg2+ Kg5 4.Ne1 Bd6 5.Nf3+ Kf4 6.Bg2 gxf3+ 7.Bxf3 
  ³  (-0.51)   Depth: 6/14   00:00:00  12kN
1.Nf4+ Kh4 2.Bxh1 Kg5 3.Ne6+ Kf6 4.Nf4 g2 5.Nd5+ Kg5 6.Bxg2 Be5 
  ³  (-0.46)   Depth: 7/16   00:00:00  29kN
1.Nf4+ Kh4 2.Bxh1 Kg5 3.Ne6+ Kf6 4.Nf4 g2 5.Nd5+ Kg5 6.Bxg2 Be5 7.Ke3 
  ³  (-0.36)   Depth: 8/18   00:00:00  47kN
1.Nf4+ Kh4 2.Bxh1 Kg5 3.Ne6+ Kf6 4.Nf4 g2 5.Nd5+ Kg5 6.Bxg2 Be5 7.Bf3 f4 
  ³  (-0.50)   Depth: 9/18   00:00:00  74kN
1.Nf4+ Kh4 2.Bxh1 Kg5 3.Ne6+ Kf6 4.Nf4 g2 5.Nd5+ Kg5 6.Bxg2 Be5 7.Bf3 f4 8.Ne7 
  ³  (-0.43)   Depth: 10/20   00:00:00  123kN
1.Nf4+ Kh4 2.Bxh1 Kg5 3.Ne6+ Kf6 4.Nf4 g2 5.Nd5+ Kg5 6.Bxg2 Bd6 7.Bf3 Bc5 8.Nc7 f4 
  ³  (-0.43)   Depth: 11/23   00:00:00  205kN
1.Nf4+ Kh4 2.Bxh1 Kg5 3.Ne6+ Kf6 4.Nf4 g2 5.Nd5+ Kg5 6.Bxg2 Be5 7.Ne3 Bd4 8.Kd3 Be5 9.Bd5 
  ³  (-0.34)   Depth: 12/24   00:00:00  530kN
1.Nf4+ Kh4 2.Bxh1 Kg5 3.Ne6+ Kf6 4.Nf4 g2 5.Nxg2 Be5 6.Ne3 g3 7.Kf3 f4 8.Nd5+ Kg5 9.Bg2 
  ³  (-0.33)   Depth: 13/27   00:00:01  1156kN
1.Nf4+ Kh4 2.Bxh1 Kg5 3.Ne6+ Kf6 4.Nf4 g2 5.Nxg2 Kg5 6.Ne1 Bd6 7.Bc6 g3 8.Kf3 f4 9.Ng2 Be5 
  ³  (-0.29)   Depth: 14/28   00:00:02  2479kN
1.Nf4+ Kh4 2.Bxh1 Kg5 3.Ne6+ Kf6 4.Nf4 g2 5.Nxg2 Kg5 6.Ne3 Be5 7.Kf1 g3 8.Nxf5 Kxf5 9.Kg2 Kf4 10.Kg1 
  =  (0.00)   Depth: 15/28   00:00:03  4548kN
Top
Chris Tatham
Posts: 142
Joined: Wed Mar 08, 2006 8:15 pm

Re: Mate in 4...

Post by Chris Tatham »

Blimey, even Numpty can find this one! :shock:

Code: Select all

58.327<--1:Depth  1  Best Move =  35 - 44 best score = -5.75
58.327<--1:Depth  1  Best Move =  35 - 45 best score = -5.45
58.327<--1:Depth  1  Best Move =  73 - 28 best score = -0.95
58.327<--1:think =  43200plye-plys =  0
58.327<--1:moved in (s) =  0
58.327<--1:Depth  2  Best Move =  73 - 28 best score = -1.25
58.327<--1:think =  43200plye-plys =  0
58.327<--1:moved in (s) =  0
58.327<--1:Depth  3  Best Move =  73 - 28 best score = -0.95
58.327<--1:Depth  3  Best Move =  77 - 56 best score = -0.8
58.327<--1:think =  43200plye-plys =  0
58.327<--1:moved in (s) =  0
58.327<--1:Depth  4  Best Move =  77 - 56 best score = -1.1
58.360<--1:think =  43200plye-plys =  0
58.360<--1:moved in (s) =  0
58.360<--1:Depth  5  Best Move =  77 - 56 best score = -0.7
58.610<--1:think =  43200plye-plys =  0
58.610<--1:moved in (s) =  0
58.610<--1:Depth  6  Best Move =  77 - 56 best score = -1
60.984<--1:think =  43200plye-plys =  300
60.984<--1:moved in (s) =  300
61.062<--1:Depth  7  Best Move =  77 - 56 best score = -0.7
67.719<--1:Depth  7  Best Move =  73 - 37 best score =  994
71.125<--1:think =  43200plye-plys =  1300
Top
Cubeman
Posts: 644
Joined: Fri Feb 02, 2007 3:11 am
Location: New Zealand

Re: Mate in 4...

Post by Cubeman »

I tested some of my PocketPC versions and got some strange results.
PF3 (Hiarcs 12.1) does not find the mate.
PF3 (Crafty 22.0b) finds it.
PF3 (Toga II 1.4.1SE) finds it.
Pocket Shredder finds it.
Chess Genius on iPhone finds it.
Glaurung on iPhone finds it.
The strange one was PocketGM (Naum 4) after a few seconds it showed the mate move 1.Bg2+ along with the mate in 4, then less than a second later it resorts back to 1.Nf4+ and stays with it?
Top
Allard Siemelink
Posts: 297
Joined: Fri Jun 30, 2006 9:30 pm
Location: Netherlands

Re: Mate in 4...

Post by Allard Siemelink »

Spark doesn't see it :oops:
(There appears to be a problem with its null move verification search)
Michael Diosi wrote:So which engine can solve it and which can't ?
Top
Tommy

Re: Mate in 4...

Post by Tommy »

It took my Tandy 1850 with it's "Ultra fast" 8Mhz processor 1:36s to find the move. :D

Image

Cheers,
Tom.
Top
rightrook
Posts: 1452
Joined: Wed Mar 08, 2006 8:45 pm

Re: Mate in 4...

Post by rightrook »

Thanks...good problem..

My Shredder 6 and Fritz 8 both find it instantly and announce the mate...:-)

regards

Robert
Top
lmader
Posts: 154
Joined: Fri Mar 10, 2006 1:20 am
Location: Sonora, Mexico

Re: Mate in 4...

Post by lmader »

It seems that engines like Rybka 3, Naum 4, and Fritz 11, that are apparently utilizing some very fancy techniques to enhance the search, are the ones that have trouble with this type of problem. The ones that are searching the tree in more direct ways find it more easily.
"The foundation of morality is to have done, once for all, with lying; to give up pretending to believe that for which there is no evidence, and repeating unintelligible propositions about things beyond the possibilities of knowledge." - T. H. Huxley
Top
jesper_nielsen

Re: Mate in 4...

Post by jesper_nielsen »

lmader wrote:It seems that engines like Rybka 3, Naum 4, and Fritz 11, that are apparently utilizing some very fancy techniques to enhance the search, are the ones that have trouble with this type of problem. The ones that are searching the tree in more direct ways find it more easily.
Well... my mid-level engine, Pupsi, also does not see the mate.

So i guess a fanzy technique is null move pruning without verification search. :D

My conditions for null move:
The side to move is not in check.
It is not a PV node.
The previous move was not a null move.
and
The side to move has at least one slider.
The evaluation is above beta.

The last condition is probably what causes the problems in this position, since black captures a bishop, going above beta in evaluation, and still has two sliders left! Thus null move allows black to stand pat when there is only one legal move, and that move leads to mate.

Kind regards,
Jesper
Top

Return to “Computer Chess Club: General Topics”