Best Move Bxd6

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

Moderator: Ras

Howard E
Posts: 261
Joined: Wed Mar 08, 2006 8:49 pm

Best Move Bxd6

Post by Howard E »

[d] 3rr1k1/1bq2p2/p2b2pQ/1pnp1pP1/3N4/P1N2RB1/1PP4P/2KR4 w - - 0 1 bm Bxd6;
Top
zullil
Posts: 6442
Joined: Tue Jan 09, 2007 12:31 am
Location: PA USA
Full name: Louis Zulli

Re: Best Move Bxd6

Post by zullil »

Crafty-22.9 agrees:

Code: Select all

White(1): display

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

White(1): go
              time limit 5:00 (+0.00) (5:00)
              depth   time  score   variation (1)
starting thread 1
                8->   0.09  -0.35   Bxd6 Qxd6 Rh3 Qf4+ Kb1 Qe5 Ka1 Ne4
                                    <HT> (s=8)
                9     0.10  -0.29   Bxd6 Qxd6 Rh3 Qf4+ Kb1 Qe5 Nf3 Qg7
                                    Qxg7+ Kxg7 Nd4 (s=7)
                9     0.18     +1   Ndxb5!!
                9     0.20   0.47   Ndxb5 axb5 Nxb5 Bxg3 Nxc7 Bxc7 Rh3
                                    Be5 Rh4
                9->   0.21   0.47   Ndxb5 axb5 Nxb5 Bxg3 Nxc7 Bxc7 Rh3
                                    Be5 Rh4 (s=3)
               10     0.24   0.18   Ndxb5 axb5 Nxb5 Bxg3 Nxc7 Bxc7 Rh3
                                    Be5 Rd2 Ne4 (s=2)
               10->   0.35   0.18   Ndxb5 axb5 Nxb5 Bxg3 Nxc7 Bxc7 Rh3
                                    Be5 Rd2 Ne4 (s=6)
               11     0.42  -0.09   Ndxb5 axb5 Nxb5 Bxg3 Nxc7 Bxc7 Rh3
                                    Be5 Rf1 Bg7 Qh4 (s=5)
               11->   0.78  -0.09   Ndxb5 axb5 Nxb5 Bxg3 Nxc7 Bxc7 Rh3
                                    Be5 Rf1 Bg7 Qh4 (s=7)
               12     0.92   0.03   Ndxb5 axb5 Nxb5 Bxg3 Nxc7 Bxc7 Rh3
                                    Be5 Rh4 Bg7 Qh7+ Kf8 Rf4 (s=6)
               12->   1.42   0.03   Ndxb5 axb5 Nxb5 Bxg3 Nxc7 Bxc7 Rh3
                                    Be5 Rh4 Bg7 Qh7+ Kf8 Rf4 (s=4)
               13     1.69  -0.25   Ndxb5 axb5 Nxb5 Bxg3 Nxc7 Bxc7 Rh3
                                    Be5 Kb1 Ne4 Ka1 Bg7 Qh7+ Kf8 (s=3)
               13->   3.68  -0.25   Ndxb5 axb5 Nxb5 Bxg3 Nxc7 Bxc7 Rh3
                                    Be5 Kb1 Ne4 Ka1 Bg7 Qh7+ Kf8 (s=7)
               14     4.30  -0.08   Ndxb5 axb5 Nxb5 Bxg3 Nxc7 Bxc7 Rh3
                                    Be5 Re1 Ne4 Kb1 Rc8 c3 Rc4 <HT> (s=6)
               14->   7.02  -0.08   Ndxb5 axb5 Nxb5 Bxg3 Nxc7 Bxc7 Rh3
                                    Be5 Re1 Ne4 Kb1 Rc8 c3 Rc4 <HT> (s=5)
               15     8.18  -0.21   Ndxb5 axb5 Nxb5 Bxg3 Nxc7 Bxc7 b4 Na4
                                    b5 d4 Rh3 Bf4+ Kb1 Nc3+ Ka1 Be5 Qh7+
                                    Kf8 (s=4)
               15->  24.45  -0.21   Ndxb5 axb5 Nxb5 Bxg3 Nxc7 Bxc7 b4 Na4
                                    b5 d4 Rh3 Bf4+ Kb1 Nc3+ Ka1 Be5 Qh7+
                                    Kf8 (s=9)
               16    27.49  -0.34   Ndxb5 axb5 Nxb5 Bxg3 Nxc7 Bxc7 Rh3
                                    Bf4+ Kb1 Be5 Re1 Ne4 Ka2 Rc8 c3 Bg7
                                    Qh7+ Kf8 (s=8)
               16    39.94     +1   Bxd6!!
               16    41.93     +3   Bxd6!!
               16    48.91     +M   Bxd6!!
              time=5:00  mat=-1  n=1379388244  fh=96%  nps=4.6M
terminating SMP processes.
White(1): Bxd6 
              time used:   5:00
Top
JVMerlino
Posts: 1404
Joined: Wed Mar 08, 2006 10:15 pm
Location: San Francisco, California

Re: Best Move Bxd6

Post by JVMerlino »

Howard E wrote:[d] 3rr1k1/1bq2p2/p2b2pQ/1pnp1pP1/3N4/P1N2RB1/1PP4P/2KR4 w - - 0 1 bm Bxd6;
CM9000 default personality finds this instantly, and sees the win in six seconds. This is on a P4-3.0:

Code: Select all

Time	Depth	Score	Positions	Moves
0:00	1/3	0.01	1777		1.Bxd6 Qxd6 2.Rh3 Qe5
0:00	1/4	0.36	7833		1.Bxd6 Qxd6 2.Rh3 Qe5 3.Kb1
0:00	1/5	0.07	34651		1.Bxd6 Qxd6 2.Rh3 Qe5 3.Kb1 Ne4
					4.Nce2
0:00	1/6	-0.08	93995		1.Bxd6 Qxd6 2.Rh3 Qe5 3.Kb1 Ne4
					4.Nce2 f4
0:01	1/7	-0.08	285395		1.Bxd6 Qxd6 2.Rh3 Qe5 3.Kb1 Ne4
					4.Nce2 f4
0:06	1/8	3.51	1253797		1.Bxd6 Qxd6 2.Rh3 Nd3+ 3.Rdxd3
					Qe5 4.Qh7+ Kf8 5.Rd2 Rc8 6.Qh8+
					Qxh8 7.Rxh8+ Kg7 8.Rxe8 Rxe8 9.Re2
0:10	1/9	3.94	2306651		1.Bxd6 Qxd6 2.Rh3 Nd3+ 3.Rdxd3
					Qe5 4.Qh7+ Kf8 5.Rde3 Qxe3+ 6.Rxe3
					Rxe3 7.Nxf5 gxf5 8.g6 Re1+ 9.Kd2
					fxg6 10.Kxe1 Re8+ 11.Ne2
0:19	1/10	3.74	4950305		1.Bxd6 Qxd6 2.Rh3 Nd3+ 3.Rdxd3
					Qe5 4.Qh7+ Kf8 5.Rde3 Qxe3+ 6.Rxe3
					Rxe3 7.Nxf5 gxf5 8.g6 fxg6 9.Qh6+
					Kf7 10.Qxe3 d4
0:50	2/11	3.80	12875263	1.Bxd6 Qxd6 2.Rh3 Nd3+ 3.Rdxd3
					Qe5 4.Rh4 Qg7 5.Rdh3 Qxh6 6.Rxh6
					Kg7 7.Rh7+ Kf8 8.Rh8+ Ke7 9.Rxe8+
					Kxe8 10.Rh7
jm
Top
Cubeman
Posts: 644
Joined: Fri Feb 02, 2007 3:11 am
Location: New Zealand

Re: Best Move Bxd6

Post by Cubeman »

Nice test position.Some engines get carried away with the 1.Ndxb5 etc.Hiarcs 12.1 on my PPC even found 1.Rxf5?! and gives it a +0.64.From looking at the previous posts it is hard to know why 1.Bxd6 Qxd6 2.Rh3 Qe5 loses.Only after testing the resulting position do I see that 3.Nxf5!! is the winner.It is good that some engines discover the 3.Nxf5 in their original PV even though it is not displayed and explains the strange looking 1.Bxd6 Qxd6 2.Rh3 Nd3+!?.Pocket Shredder and Naum 4 find the winning line and award 1.Bxd6 > 3.50.
Top
ernest
Posts: 2053
Joined: Wed Mar 08, 2006 8:30 pm

Re: Best Move Bxd6

Post by ernest »

Instantaneous with Rybka 3: 8-)

Code: Select all

Analysis by Rybka 3 32-bit:
1.Ncxb5 axb5 2.Nxb5 Bxg3 3.Nxc7 Bxc7 4.Rxf5 Ne6 5.Rf6 Bf4+ 6.Rxf4 Nxf4 7.Qh4 Ne2+ 8.Kb1 
  =  (-0.20)   Depth: 8   00:00:00  41kN
1.Ncxb5 axb5 2.Nxb5[] Bxg3 3.Nxc7 Bxc7 4.Rxf5 Ne6 5.Rf6 Bf4+ 6.Rxf4 Nxf4 7.Qh4 Ne2+ 8.Kb1 
  =  (-0.20)   Depth: 9   00:00:00  65kN
1.Bxd6 
  =  (0.00 !)   Depth: 9   00:00:01  74kN
1.Bxd6 
  =  (0.20 !)   Depth: 9   00:00:01  81kN
1.Bxd6 
  +/=  (0.60 !)   Depth: 9   00:00:01  92kN
1.Bxd6 
  +/-  (1.40 !)   Depth: 9   00:00:01  104kN
1.Bxd6 
  +-  (3.00 !)   Depth: 9   00:00:01  143kN
1.Bxd6 Qxd6[] 2.Rh3 
  +-  (3.27)   Depth: 9   00:00:02  222kN
1.Bxd6 
  +-  (3.47 !)   Depth: 10   00:00:03  282kN
1.Bxd6 
  +-  (3.67 !)   Depth: 10   00:00:04  325kN
1.Bxd6 Qxd6[] 2.Rh3[] Nb3+ 3.Nxb3[] Qf4+ 4.Kb1[] Qe5 5.Qh7+ Kf8 6.Nc5[] Bc6[] 7.Nxa6 d4[] 8.Qh8+ Qxh8[] 9.Rxh8+ Ke7[] 10.Rxe8+ Bxe8[] 11.Ne2 
  +-  (3.93)   Depth: 10   00:00:05  418kN
1.Bxd6 Qxd6[] 2.Rh3[] Nb3+ 3.Nxb3[] Qe5 
  +-  (3.93)   Depth: 11   00:00:06  498kN
1.Bxd6 
  +-  (4.13 !)   Depth: 12   00:00:09  768kN
1.Bxd6 Qxd6[] 2.Rh3[] Nb3+ 3.Nxb3[] 
  +-  (4.14)   Depth: 12   00:00:11  880kN
1.Bxd6 Qxd6[] 2.Rh3[] Nb3+ 3.Nxb3[] 
  +-  (4.25)   Depth: 13   00:00:16  1399kN
1.Bxd6 Nb3+ 2.Nxb3 Qxd6 3.Rh3 Qf4+ 
  +-  (4.25)   Depth: 14   00:00:28  2464kN
1.Bxd6 
  +-  (4.45 !)   Depth: 15   00:00:42  3578kN
Top
Terry McCracken
Posts: 16465
Joined: Wed Aug 01, 2007 4:16 am
Location: Canada

Re: Best Move Bxd6

Post by Terry McCracken »

Howard E wrote:[d] 3rr1k1/1bq2p2/p2b2pQ/1pnp1pP1/3N4/P1N2RB1/1PP4P/2KR4 w - - 0 1 bm Bxd6;
Nice position...Nxf5!! is hidden but crushes Black. Who were the players?

Terry
Top
Howard E
Posts: 261
Joined: Wed Mar 08, 2006 8:49 pm

Re: Best Move Bxd6

Post by Howard E »

The players were Prodeo(Ares) - Alaric.
Game 40/40 repeating
Athlon 2800
128mb each

[Event "*"]
[Site "*"]
[Date "2009.02.15"]
[Round "1.1"]
[White "Pro Deo 1.6 - Ares"]
[Black "Alaric 707"]
[Result "1-0"]
[TimeControl "40/2400:40/2400"]


{ERT 1.0.0.5 Autoflag=On Pondering=Off OperatorTime=300 mS Start positions=NoomenTestsuite2008.pgn}
1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. Bg5 e6 7. f4 Nbd7 8. Qe2
Qc7 9. O-O-O b5 10. a3 Bb7 11. g4 Be7 12. Bg2 {0.11/13 89} O-O {0.10/14 94} 13. Bh4
{0.17/13 74} Nc5 {0.07/14 82} 14. g5 {0.30/14 93} Nfxe4 {0.13/16 114} 15. Bxe4 {0.59/12 24} d5
{0.15/16 177} 16. Bxh7+ {0.73/13 137} Kxh7 {0.15/15 77} 17. Bg3 {0.39/14 141} Kg8
{-0.15/14 77} 18. Qh5 {0.46/13 79} g6 {-0.24/14 77} 19. Qh6 {0.99/14 62} Bd6 {-0.05/14 77}
20. Rhf1 {1.06/13 67} Rad8 {0.06/14 77} 21. Rf3 {1.25/14 167} Rfe8 {0.52/15 77} 22. f5
{3.46/14 126} exf5 {0.53/15 77} 23. Rxf5 {1.07/13 118} Bxg3 {-0.43/12 77} 24. hxg3
{6.39/13 59} Qe7 {6.23/14 202} {Score threshold reached} 1-0

move 23, prodeo does not play Bxd6 but wins anyway. Alaric must have blundered with Bxg3. I tweaked the final Ares version to catch Bxd6.
Top
User avatar
Graham Banks
Posts: 44611
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: Best Move Bxd6

Post by Graham Banks »

Howard E wrote:The players were Prodeo(Ares) - Alaric.
Game 40/40 repeating
Athlon 2800
128mb each

[Event "*"]
[Site "*"]
[Date "2009.02.15"]
[Round "1.1"]
[White "Pro Deo 1.6 - Ares"]
[Black "Alaric 707"]
[Result "1-0"]
[TimeControl "40/2400:40/2400"]


{ERT 1.0.0.5 Autoflag=On Pondering=Off OperatorTime=300 mS Start positions=NoomenTestsuite2008.pgn}
1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. Bg5 e6 7. f4 Nbd7 8. Qe2
Qc7 9. O-O-O b5 10. a3 Bb7 11. g4 Be7 12. Bg2 {0.11/13 89} O-O {0.10/14 94} 13. Bh4
{0.17/13 74} Nc5 {0.07/14 82} 14. g5 {0.30/14 93} Nfxe4 {0.13/16 114} 15. Bxe4 {0.59/12 24} d5
{0.15/16 177} 16. Bxh7+ {0.73/13 137} Kxh7 {0.15/15 77} 17. Bg3 {0.39/14 141} Kg8
{-0.15/14 77} 18. Qh5 {0.46/13 79} g6 {-0.24/14 77} 19. Qh6 {0.99/14 62} Bd6 {-0.05/14 77}
20. Rhf1 {1.06/13 67} Rad8 {0.06/14 77} 21. Rf3 {1.25/14 167} Rfe8 {0.52/15 77} 22. f5
{3.46/14 126} exf5 {0.53/15 77} 23. Rxf5 {1.07/13 118} Bxg3 {-0.43/12 77} 24. hxg3
{6.39/13 59} Qe7 {6.23/14 202} {Score threshold reached} 1-0

move 23, prodeo does not play Bxd6 but wins anyway. Alaric must have blundered with Bxg3. I tweaked the final Ares version to catch Bxd6.
Hi Howard,

are you prepared to share your final Ares setting and provide some test results to show how much of an improvement it is?
Just wondering.

Cheers, Graham.
gbanksnz at gmail.com
Top
Howard E
Posts: 261
Joined: Wed Mar 08, 2006 8:49 pm

Re: Best Move Bxd6

Post by Howard E »

Yes I will share it. I have not changed it for about a couple of weeks now. The mx4 settings did not hold up too well so I tried to vindicate some of the ideas from that multi-personality one. Actually I may better have called mx4 prodeo - lobotomy, since it seemed blind to many tactics. Ares is more balanced, but I've only tested it using the Noomen2008 30 game set.
Top
swami
Posts: 6662
Joined: Thu Mar 09, 2006 4:21 am

Re: Best Move Bxd6

Post by swami »

Analysis by bright-0.4-cct1:

Code: Select all

1.Bxd6 Rxd6 
  -/+  (-0.78)   Depth: 1/2   00:00:00
1.Bxd6 Rxd6 
  -/+  (-0.78)   Depth: 2/8   00:00:00
1.Bxd6 Rxd6 2.h4 
  =/+  (-0.66)   Depth: 3/8   00:00:00
1.Bxd6 Qxd6 2.Rh3 Qf4+ 3.Kb1 
  -/+  (-0.72)   Depth: 4/19   00:00:00
1.Bxd6 Qxd6 2.Rh3 Qe5 3.Qh7+ Kf8 4.Qh6+ Qg7 5.Rf1 
  =/+  (-0.35)   Depth: 5/19   00:00:00
1.Bxd6 Qxd6 2.Rh3 Qe5 3.Qh7+ Kf8 4.Qh6+ Qg7 5.Rf1 Kg8 
  =/+  (-0.47)   Depth: 6/19   00:00:00  36kN
1.Ncxb5 axb5 2.Nxb5 Bxg3 3.Nxc7 Bxc7 4.Rh3 Bf4+ 5.Kb1 
  =  (0.14)   Depth: 6/19   00:00:00  73kN
1.Ncxb5 axb5 2.Nxb5 Bxg3 3.Nxc7 Bxc7 4.Rh3 Be5 5.Qh7+ Kf8 6.Kb1 d4 
  =  (0.16)   Depth: 7/21   00:00:00  132kN
1.Ncxb5 axb5 2.Nxb5 Bxg3 3.Nxc7 Bxc7 4.Rh3 Be5 5.Qh7+ Kf8 6.Kb1 Re6 7.Re3 
  +/=  (0.30)   Depth: 8/30   00:00:00  344kN
1.Ncxb5 axb5 2.Nxb5 Bxg3 3.Nxc7 Bxc7 4.Rh3 Be5 5.Qh7+ Kf8 6.Kb1 Re6 7.Re3 d4 
  =  (0.03)   Depth: 9/32   00:00:00  811kN
1.Ncxb5 axb5 2.Nxb5 Bxg3 3.Nxc7 Bxc7 4.Rh3 Be5 5.Qh7+ Kf8 6.Kb1 d4 7.b4 Ne4 8.Re1 
  =  (0.00)   Depth: 10/32   00:00:00  1880kN
1.Ncxb5 axb5 2.Nxb5 Bxg3 3.Nxc7 Bxc7 4.Rh3 Be5 5.Qh7+ Kf8 6.Re3 Bxb2+ 7.Kxb2 Na4+ 8.Kb3 Rxe3+ 9.Kxa4 Bc6+ 10.Ka5 Rxa3+ 11.Kb6 
  =  (0.07)   Depth: 11/36   00:00:00  4340kN
1.Rxf5 gxf5 2.g6 Bxg3 3.Qh7+ Kf8 4.hxg3 Re5 5.g7+ Ke7 6.g8Q Rxg8 7.Qxg8 Ne4 8.Nde2 Qb6 9.Kb1 
  +/=  (0.35)   Depth: 11/41   00:00:02  11211kN
1.Rxf5 gxf5 2.g6 f6 3.Bxd6 Rxd6 4.g7 Rdd8 5.Nxf5 Qb8 6.Qg6 Qf4+ 7.Kb1 d4 8.Nh6+ Qxh6 9.Qxh6 dxc3 10.Rxd8 Rxd8 11.Qxf6 
  +/=  (0.49)   Depth: 12/51   00:00:08  36444kN
1.Rxf5 gxf5 2.g6 fxg6 3.Qxg6+ Qg7 4.Qxg7+ Kxg7 5.Nxf5+ Kf7 6.Nxd6+ Rxd6 7.Bxd6 Ne4 8.Bb4 Ke6 9.Ne2 Ke5 10.Nd4 Rh8 11.Nf3+ Kf4 
  +/=  (0.62)   Depth: 13/51   00:00:14  69202kN
1.Bxd6 Qxd6 2.Rh3 Nb3+ 3.Nxb3 Qf4+ 4.Kb1 Qe5 5.Nc5 Bc6 6.Nxa6 d4 7.Ne2 Bg2 8.Qh7+ Kf8 9.Qh8+ Qxh8 10.Rxh8+ Kg7 11.Rxe8 Rxe8 12.Nxd4 
  +-  (3.70)   Depth: 13/51   00:00:17  85088kN
1.Bxd6 Qxd6 2.Rh3 Nb3+ 3.Nxb3 Qf4+ 4.Kb1 Qe5 5.Nc5 Bc6 6.Nxa6 d4 7.Nb4 Bg2 8.Qh7+ Kf8 9.Qh8+ Qxh8 10.Rxh8+ Kg7 11.Rxe8 Rxe8 12.Rxd4 
  +-  (3.95)   Depth: 14/51   00:00:19  98385kN
1.Bxd6 Qxd6 2.Rh3 Nb3+ 3.Nxb3 Qf4+ 4.Kb1 Qe5 5.Nc5 Bc6 6.Nd7 Rxd7 7.Qh7+ Kf8 8.Qh8+ Qxh8 9.Rxh8+ Ke7 10.Re1+ Kd6 11.Rhxe8 d4 12.Rd1 Kc5 
  +-  (3.91)   Depth: 15/57   00:00:23  120mN
1.Bxd6 Qxd6 2.Rh3 Nb3+ 3.Nxb3 Qe5 4.Nc5 Bc6 5.Nxa6 d4 6.Qh7+ Kf8 7.Qh8+ Qxh8 8.Rxh8+ Ke7 9.Rxe8+ Bxe8 10.Ne2 f6 11.Nxd4 fxg5 12.Nc7 Kf6 13.Ndxb5 
  +-  (3.91)   Depth: 16/57   00:00:33  186mN
1.Bxd6 Qxd6 2.Rh3 Nb3+ 3.Nxb3 Qf4+ 4.Kb1 Qe5 5.Nc5 Bc6 6.Nxa6 d4 7.Nb4 Bf3 8.Qh7+ Kf8 9.Rxf3 dxc3 10.Rxd8 Rxd8 11.Rxc3 Rd1+ 12.Ka2 Qe6+ 13.Rb3 
  +-  (4.02)   Depth: 17/67   00:01:03  353mN
1.Bxd6 Qxd6 2.Rh3 Nb3+ 3.Nxb3 Qf4+ 4.Kb1 Qe5 5.Nc5 Bc6 6.Nxa6 d4 7.Nb4 Bf3 8.Qh7+ Kf8 9.Rxf3 dxc3 10.Rxd8 Rxd8 11.Rxc3 Qg7 12.Qh6 Qxh6 13.gxh6 Rd1+ 14.Ka2 
  +-  (4.25)   Depth: 18/67   00:02:11  717mN
Top

Return to “Computer Chess Club: General Topics”