Does any engine find Ne5 in this position?

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

Moderator: Ras

Uri
Posts: 507
Joined: Thu Dec 27, 2007 9:34 pm

Re: Does any engine find Ne5 in this position?

Post by Uri »

Spock wrote:No, the machine had in fact been re-booted between the two runs
So how do you explain the difference between my Rybka and your Rybka?
Spock

Re: Does any engine find Ne5 in this position?

Post by Spock »

Uri wrote:
Spock wrote:No, the machine had in fact been re-booted between the two runs
So how do you explain the difference between my Rybka and your Rybka?
I don't know.
User avatar
smrf
Posts: 484
Joined: Mon Mar 13, 2006 11:08 am
Location: Klein-Gerau, Germany

Re: Does any engine find Ne5 in this position?

Post by smrf »

This is, what SMIRF MS-175i has been thinking:
  • 05:05.7 (15.01=) +5.666 25...Ne5 (Rf8) 26.Qf1 Bh3 27.Rd2 Bxf1 28.Rxf1 c4 29.Kb1 Qg3 30.Bxa7 Nd3 31.Bxd3 cxd3 32.Bd4 Re1+ 33.Rd1
    02:14.9 (14.01=) +5.582 25...Ne5 (Rf8) 26.Qf1 Bh3 27.Rd2 Bxf1 28.Rxf1 c4 29.Kb1 Qg3 30.Bxa7 Nd3 31.Bxd3 Re1+ 32.Rxe1 Qxe1+
    01:41.5 (13.01=) +6.084 25...Ne5 (Rf8) 26.Qf1 Bh3 27.Rd2 Bxf1 28.Rxf1 c4 29.Bxc4+ Nxc4 30.Rxh2 Nxe3+ 31.Kb3 Nxf1 32.Rxh4 g5
    01:39.8 (13.01+) +5.887 25...Ne5 (Rf8) 26.Qf1 Bh3 27.Rd2 Bxf1 28.Rxf1 c4 29.Bxc4+ Nxc4 30.Rxh2 Nxe3+ 31.Kb1 Nxf1 32.Rxh4 Re1+ 33.Kc2
    00:23.7 (12.01=) +5.559 25...Ne5 (=/=) 26.Qf1 Nc4 27.Bd2 Qxe2 28.Qxe2 Bf5+ 29.Qd3 Bxd3+ 30.Kxd3 Nxb2+ 31.Kc2 Nxd1 32.Rxd1 Re2
    00:13.9 (11.01=) +6.186 25...Ne5 (Rf8) 26.Qf1 Nc4 27.Bd2 Qxe2 28.Qxe2 Bf5+ 29.Qd3 Bxd3+ 30.Kxd3 Nxb2+ 31.Kc2 Nxd1 32.Rxd1 Re3
    00:11.7 (11.01+) +6.186 25...Ne5 (Rf8) 26.Qf4 Qxe2+ 27.Rd2 Qh5 28.Bxc5 Nc4 29.Kc1 Qxc5 30.Qxh4 Nxd2 31.Qxh7+ Kxh7 32.Kxd2
    00:07.2 (10.01=) +5.742 25...Ne5 (Rf8) 26.Qf4 Qxe2+ 27.Rd2 Qf3 28.Rf1 Qxf4 29.Rxf4 Be7 30.Rd7
    00:06.6 (10.01+) +5.643 25...Ne5 (Rf8) 26.Qh5 g6 27.Rg1 Bg4 28.Qxg6+ hxg6 29.Rxg4 Nxg4 30.Bxc5
    00:04.8 (10.01+) +4.687 25...Ne5 (Rf8) 26.Qh5 Bf7 27.Rh1 Qg2 28.Qxh7+ Kxh7 29.Rxh4+ Kg8 30.Re1 Bxa2 31.Bxc5 Qg6+ 32.Bd3
    00:01.2 (09.01=) +4.369 25...Ne5 (Rf8) 26.Qh5 Bf7 27.Rh1 Qg2 28.Qxh7+ Kxh7 29.Rxh4+ Kg8 30.Re1
    00:01.1 (09.01+) +3.910 25...Ne5 (Rf8) 26.Qh5 Bf7 27.Rh1 Qg2 28.Qxh4 Qxe2+ 29.Kc1 Qxe3+ 30.Kb1
    00:00.6 (08.01=) +3.598 25...Ne5 (Rf8) 26.Qh5 Bf7 27.Rh1 Bg6+ 28.Kb3 Qxh1 29.Rxh1 c4+ 30.Ka3 Be7+ 31.b4 Bxh5 32.Rxh5 Bxb4+ 33.cxb4
    00:00.3 (07.01=) +3.840 25...Ne5 (Rf8) 26.Qh5 Bf7 27.Rh1 Bg6+ 28.Kb3 Qxh1 29.Rxh1 Bxh5 30.Bxh5 g6
    00:00.1 (06.02=) +2.439 25...Ne5 (Rf8) 26.Qh5 g6 27.Rg1 Bf5+ 28.Qxf5 Qxe2+ 29.Bd2 Be1
    00:00.0 (05.03=) +2.379 25...Rf8 (Ne5) 26.Qh5 Bf5+ 27.Kc1 Nf4 28.Rd7 Nxh5 29.Rxg7+ Nxg7 30.Bc4+
    00:00.0 (02.23=) +1.420 25...c4 (a5) 26.Rd7
Uri
Posts: 507
Joined: Thu Dec 27, 2007 9:34 pm

Re: Does any engine find Ne5 in this position?

Post by Uri »

Spock wrote:I don't know.
Sorry I made a mistake. My mistake, I posted the wrong position. This is the position I wanted to post in the beginning of this thread:

[d]4r1k1/p5pp/4b1n1/1pp5/7b/2P2Q2/PPKBB2q/R2R4 b - - 0 25[d]

Here Rybka 3 plays c4 which is not the best move. Ne5 is stronger.
Kurt Utzinger
Posts: 169
Joined: Sun May 11, 2008 10:31 pm
Location: Switzerland

Re: Does any engine find Ne5 in this position?

Post by Kurt Utzinger »

Uri wrote:
Spock wrote:That was 4 cores, but it finds it instantly on 1 core as well
Perhaps it has the results from the 4 cores already stored in memory so it also shows Ne5 also on the single core. Or maybe you use a different engine setup than me.

Otherwise I can't explain why my Rybka 3 doesn't find it. Any idea?
In the meantime we all know that finding a solution within X-time happens more or less accidently by mp-engines and in the given position we should also not forget that it was a 1m+0s game ... and nobody will be able to reproduce the moves of such bullet games.
Kurt
Uri Blass
Posts: 10892
Joined: Thu Mar 09, 2006 12:37 am
Location: Tel-Aviv Israel

Re: Does any engine find Ne5 in this position?

Post by Uri Blass »

Uri wrote:
Spock wrote:I don't know.
Sorry I made a mistake. My mistake, I posted the wrong position. This is the position I wanted to post in the beginning of this thread:

[d]4r1k1/p5pp/4b1n1/1pp5/7b/2P2Q2/PPKBB2q/R2R4 b - - 0 25[d]

Here Rybka 3 plays c4 which is not the best move. Ne5 is stronger.
move is probably dependent on the time control and I suspect both moves are good enough to win.

New game
4r1k1/p5pp/4b1n1/1pp5/7b/2P2Q2/PPKBB2q/R2R4 b - - 0 1

Analysis by Rybka 3 1-cpu 32-bit :

1...Ng6-e5
³ (-0.33) Depth: 2 00:00:00
1...Qh2xe2
³ (-0.49) Depth: 2 00:00:00
1...Qh2xe2
³ (-0.46) Depth: 3 00:00:00
1...b5-b4
³ (-0.66) Depth: 4 00:00:00
1...b5-b4
³ (-0.62) Depth: 4 00:00:00
1...Re8-f8 2.Qf3-e3
³ (-0.68) Depth: 5 00:00:00
1...Ng6-e5
µ (-0.88) Depth: 6 00:00:00 10kN
1...Ng6-e5
µ (-1.08) Depth: 6 00:00:00 10kN
1...Ng6-e5 2.Qf3-e4 Be6-f5 3.Qe4xf5
µ (-1.20) Depth: 6 00:00:00 12kN
1...Ng6-e5 2.Qf3-e4 Be6-f5 3.Qe4xf5 Qh2xe2
µ (-1.20) Depth: 7 00:00:00 13kN
1...Ng6-e5
µ (-1.40) Depth: 8 00:00:00 19kN
1...Ng6-e5 2.Qf3-e4 Ne5-g4 3.Bd2-f4 Be6-b3+
µ (-0.78) Depth: 8 00:00:01 29kN
1...Re8-f8 2.Qf3-e3 Ng6-e5 3.Rd1-f1 Bh4-f2
µ (-0.86) Depth: 8 00:00:01 35kN
1...Re8-f8 2.Qf3-e3 Ng6-e5 3.Be2xb5 Be6-f5+ 4.Kc2-c1 c5-c4 5.Rd1-f1 Bh4-e7 6.b2-b4
µ (-0.75) Depth: 9 00:00:02 58kN
1...Ng6-e5
µ (-1.06) Depth: 9 00:00:02 75kN
1...Ng6-e5
µ (-1.26) Depth: 9 00:00:02 85kN
1...Ng6-e5 2.Qf3-e4 Ne5-g4 3.Rd1-h1 Be6-f5 4.Qe4xf5 Qh2xe2 5.Qf5-d5+ Kg8-h8 6.Rh1-h3 Ng4-e3+ 7.Rh3xe3 Re8xe3 8.Ra1-d1 Re3-e5 9.Qd5-a8+ Re5-e8 10.Qa8xa7 Qe2-c4 11.Kc2-c1
µ (-0.86) Depth: 9 00:00:04 135kN
1...Ng6-e5 2.Qf3-e4 Ne5-g4 3.Rd1-h1 Be6-f5 4.Qe4xf5 Qh2xe2 5.Qf5-d5+ Kg8-h8 6.Rh1-h3 Ng4-e3+ 7.Rh3xe3 Re8xe3 8.Ra1-d1 Re3-e5 9.Qd5-a8+ Re5-e8 10.Qa8xa7 Qe2-c4 11.Kc2-c1
µ (-0.86) Depth: 10 00:00:05 156kN
1...Ng6-e5 2.Qf3-e4 Ne5-g4 3.Rd1-h1 Be6-f5 4.Qe4xf5 Qh2xe2 5.Qf5-d5+ Kg8-h8 6.Rh1-h3 Ng4-e3+ 7.Rh3xe3 Re8xe3 8.Ra1-d1 Re3-e5 9.Qd5-a8+ Re5-e8 10.Qa8xa7 Qe2-c4 11.Kc2-c1
µ (-0.86) Depth: 11 00:00:07 206kN
1...Ng6-e5 2.Qf3-e4 Ne5-g4 3.Rd1-h1 Be6-f5 4.Qe4xf5 Qh2xe2 5.Qf5-d5+ Kg8-h8 6.Rh1-h3 Ng4-e3+ 7.Rh3xe3 Re8xe3 8.Ra1-d1 Re3-e5 9.Qd5-a8+ Re5-e8 10.Qa8xa7 Qe2-c4 11.Kc2-c1
µ (-0.86) Depth: 12 00:00:09 257kN
1...Ng6-e5 2.Qf3-e4 Ne5-g4 3.Rd1-h1 Be6-f5 4.Qe4xf5 Qh2xe2 5.Qf5-d5+ Kg8-h8 6.Rh1-h3 Ng4-e3+ 7.Rh3xe3 Re8xe3 8.Ra1-d1 Re3-e5 9.Qd5-d7 c5-c4 10.Rd1-f1 Re5-e8
µ (-0.97) Depth: 13 00:00:25 714kN
1...c5-c4
µ (-1.06) Depth: 13 00:00:43 1327kN
1...c5-c4 2.Rd1-f1 Qh2-e5 3.a2-a4 b5-b4 4.Qf3-e3 b4-b3+ 5.Kc2-c1 Qe5-c7 6.Qe3-d4
µ (-0.98) Depth: 13 00:00:49 1501kN
1...c5-c4 2.Rd1-f1 Qh2-e5 3.a2-a4 b5-b4 4.Qf3-e3 b4-b3+ 5.Kc2-c1 Qe5-c7 6.Be2-f3 Ng6-e5 7.Bf3-e4 Bh4-f6
µ (-0.98) Depth: 14 00:01:02 1858kN
1...c5-c4 2.Rd1-f1 Qh2-e5 3.a2-a4 b5-b4 4.Qf3-e3 b4-b3+ 5.Kc2-c1 Qe5-c7 6.Qe3-d4 Bh4-f6 7.Rf1xf6 g7xf6 8.Qd4xf6 Be6-d5 9.Be2-d1 Ng6-e5 10.Bd2-e3
µ (-1.14) Depth: 15 00:02:56 5193kN
1...c5-c4 2.Rd1-f1 Qh2-e5 3.Be2-d1 Bh4-f6 4.Kc2-c1 Re8-d8 5.Rf1-g1
µ (-1.14) Depth: 16 00:06:16 11122kN

(so k, 18.01.2009)
ernest
Posts: 2053
Joined: Wed Mar 08, 2006 8:30 pm

Re: Does any engine find Ne5 in this position?

Post by ernest »

Uri Blass wrote:Analysis by Rybka 3 1-cpu 32-bit :
Sorry, can't reproduce... :?
What was your hash?
kgburcham
Posts: 2016
Joined: Sun Feb 17, 2008 4:19 pm

Re: Does any engine find Ne5 in this position?

Post by kgburcham »

[d] 4r1k1/p5pp/4b1n1/1pp5/7b/2P2Q2/PPKBB2q/R2R4 b - -

Zappa Mexico II

10/31 0:01 -1.38 1...c4 2.Rf1 Qe5 3.Qe3 Bc8 4.Qxe5 Rxe5 5.Kd1 Bh3 (1.862.112) 1727
10/31 0:01 -1.38 1...c4 2.Rf1 Qe5 3.Qe3 Bc8 4.Qxe5 Rxe5 5.Kd1 Bh3 (2.109.545) 1709
11/31 0:01 -1.35 1...c4 2.Rf1 Qe5 3.Qe3 Bf5+ 4.Kd1 Be4 5.b3 Rd8 6.Kc1 (3.155.210) 1802
11/31 0:02 -1.35 1...c4 2.Rf1 Qe5 3.Qe3 Bf5+ 4.Kd1 Be4 5.b3 Rd8 6.Kc1 (3.939.927) 1827
12/38 0:04 -1.41 1...c4 2.Rf1 Qe5 3.Qh5 Bh3 4.Qxe5 Rxe5 5.Rf3 Bf5+ 6.Kd1 Be4 7.b3 Bxf3 8.Bxf3 (8.192.286) 1949
12/38 0:04 -1.41 1...c4 2.Rf1 Qe5 3.Qh5 Bh3 4.Qxe5 Rxe5 5.Rf3 Bf5+ 6.Kd1 Be4 7.b3 Bxf3 8.Bxf3 (9.288.835) 1955
13/38 0:08 -1.33 1...c4 2.Rf1 Qe5 3.b4 Bd5 4.Qh5 Bg2 5.Bf3 Bxf1 6.Rxf1 Qxh5 7.Bxh5 Bf6 8.Bf3 (16.595.744) 2058
13/39 0:10 -1.33 1...c4 2.Rf1 Qe5 3.b4 Bd5 4.Qh5 Bg2 5.Bf3 Bxf1 6.Rxf1 Qxh5 7.Bxh5 Bf6 8.Bf3 (21.974.204) 2041
14/39 0:15 -1.33 1...c4 2.Rf1 Qe5 3.a4 b4 4.Qe3 b3+ 5.Kd1 Rd8 6.Qxe5 Nxe5 7.Rf4 Rxd2+ 8.Kxd2 Bg5 9.Raf1 (31.383.477) 2057
14/39 0:18 -1.33 1...c4 2.Rf1 Qe5 3.a4 b4 4.Qe3 b3+ 5.Kd1 Rd8 6.Qxe5 Nxe5 7.Rf4 Rxd2+ 8.Kxd2 Bg5 9.Raf1 (38.161.716) 2068
15/49 0:26 -1.35 1...c4 2.Rf1 Qe5 3.a4 Bd5 4.Qh5 Qe4+ 5.Kd1 Rf8 6.Rxf8+ Kxf8 7.Bf1 Be6 8.Qc5+ Kg8 9.Qe3 (54.825.138) 2082
15/49 1:02 -1.35 1...c4 2.Rf1 Qe5 3.a4 Bd5 4.Qh5 Qe4+ 5.Kd1 Rf8 6.Rxf8+ Kxf8 7.Bf1 Be6 8.Qc5+ Kg8 9.Qe3 (128.336.864) 2041
16/51 1:34 -1.29 1...c4 2.Rf1 Qe5 3.Qh5 Bh3 4.Bf3 Qxh5 5.Bxh5 Re5 6.Bxg6 Bxf1 7.Bxh7+ Kxh7 8.Rxf1 g5 9.b3 Kg6 10.bxc4 bxc4 (195.067.192) 2056
16/51 2:48 -1.55 1...Ne5 2.Qe4 Ng4 3.Rh1 Bf5 4.Qxf5 Qxe2 5.Qd5+ Kh8 6.Rh3 Ne3+ 7.Rxe3 Rxe3 8.Qxc5 Re8 9.Rd1 Rd8 10.Qe3 Qg2 11.Kc1 Qd5 12.a3 a5 13.Rg1 (344.698.198) 2047
16/51 3:11 -1.55 1...Ne5 2.Qe4 Ng4 3.Rh1 Bf5 4.Qxf5 Qxe2 5.Qd5+ Kh8 6.Rh3 Ne3+ 7.Rxe3 Rxe3 8.Qxc5 Re8 9.Rd1 Rd8 10.Qe3 Qg2 11.Kc1 Qd5 12.a3 a5 13.Rg1 (390.724.077) 2037
17/53 5:43 -1.52 1...Ne5 2.Qe4 Ng4 3.Rh1 Bf5 4.Qxf5 Qxe2 5.Qd5+ Kh8 6.Rh3 Ne3+ 7.Rxe3 Rxe3 8.Qxc5 Re8 9.Rd1 Rd8 10.Qe3 Qg2 11.Kc1 Qd5 12.Qe2 Bg5 13.Bxg5 Qxg5+ (703.556.424) 2051
best move: Ng6-e5 time: 6:02.578 min n/s: 1.811.919 nodes: 736.922.486
Uri Blass
Posts: 10892
Joined: Thu Mar 09, 2006 12:37 am
Location: Tel-Aviv Israel

Re: Does any engine find Ne5 in this position?

Post by Uri Blass »

ernest wrote:
Uri Blass wrote:Analysis by Rybka 3 1-cpu 32-bit :
Sorry, can't reproduce... :?
What was your hash?
My hash was 256 mbytes
I can use 512 mbytes but I decided to use 256 mbytes usually because 512 mbytes often cause significantly slow down when I do other things.
Uri
Posts: 507
Joined: Thu Dec 27, 2007 9:34 pm

Re: Does any engine find Ne5 in this position?

Post by Uri »

Uri Blass wrote:My hash was 256 mbytes
I can use 512 mbytes but I decided to use 256 mbytes usually because 512 mbytes often cause significantly slow down when I do other things.
I use only 64 MB hash. How many threads/cores did you use?