Anudder test position--too easy for commercials

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

Moderator: Ras

kgburcham
Posts: 2016
Joined: Sun Feb 17, 2008 4:19 pm

Anudder test position--too easy for commercials

Post by kgburcham »

[d] r4rk1/pp2qpp1/2p1bn1p/8/1bP4Q/5N1P/PPB2PP1/R1BR2K1 w - - bm Bxh6 iq81 #182


Little Thought v1.03
14/52 4:41 +0.71++ 1.Bxh6 gxh6 2.Qxh6 Rfd8 3.Ng5 Bxc4 4.Bh7+ Kh8 5.Rxd8+ Rxd8 6.Bd3+ Kg8 7.Bxc4 Nh7 8.Nxh7 Kh8 9.a3 (193.545.039) 687
14/52 6:00 +1.31++ 1.Bxh6 gxh6 2.Qxh6 Rfd8 3.Ng5 Nh7 4.Nxh7 Kh8 (294.207.468) 816
best move: a2-a3 time: 7:05.438 min n/s: 868.857 nodes: 350.000.008
Top
kgburcham
Posts: 2016
Joined: Sun Feb 17, 2008 4:19 pm

Arasan 10.4

Post by kgburcham »

[d] r4rk1/pp2qpp1/2p1bn1p/8/1bP4Q/5N1P/PPB2PP1/R1BR2K1 w - -

Arasan 10.4 smp
Running on two processors

12.01 1:01 +0.56 1.Bxh6 gxh6 2.Qxh6 Rfd8 3.Ng5 Rxd1+ 4.Rxd1 Bc5 5.Bh7+ Kh8 6.Nxe6 fxe6 7.Bd3+ Kg8 8.h4 (85.319.773) 1379
13.01 1:10 +1.06 1.Bxh6 gxh6 (97.590.168) 1383
13.01 1:18 +1.18 1.Bxh6 gxh6 2.Qxh6 Rfd8 3.Re1 Bxe1 4.Rxe1 Qf8 5.Qxf6 Qg7 6.Qh4 Qxb2 7.Qh7+ Kf8 (109.368.763) 1388
14.01 1:49 +1.31 1.Bxh6 gxh6 2.Qxh6 Rfd8 3.Ng5 Rd6 4.Bh7+ Kh8 5.Bd3+ Kg8 6.a3 Bc5 7.Bh7+ Kh8 8.Nxf7+ Bxf7 9.Bd3+ Kg8 10.Qg5+ Kf8 11.Qxc5 (154.078.720) 1413
15.01 3:33 +1.56 1.Bxh6 gxh6 2.Qxh6 Rfd8 3.a3 Rxd1+ 4.Rxd1 Bc5 5.Ng5 Bxf2+ 6.Kxf2 Bf5 7.Bxf5 Qc5+ 8.Kf1 Qxf5+ 9.Nf3 Re8 10.Kg1 (305.061.548) 1427
16.01 11:03 +2.06 1.Bxh6 gxh6 (955.303.876) 1439
16.01 18:53 +2.18 1.Bxh6 Rfd8 2.Bg5 Rxd1+ 3.Rxd1 Rd8 4.Bh7+ Kf8 5.Bd3 Ke8 6.a3 Bc5 7.b4 Bd6 8.Be3 c5 9.Qh8+ Qf8 (1.615.359.713) 1425
17.01 25:10 +2.25 1.Bxh6 Rfd8 2.Bg5 Rxd1+ 3.Rxd1 Rd8 4.Bh7+ Kf8 5.Bd3 Ke8 6.Nd4 Bc8 7.a3 Bd6 8.b4 Kd7 9.Bd2 (2.158.526.658) 1428
18.01 42:18 +2.37 1.Bxh6 Rfd8 2.Bxg7 Kxg7 3.Qg5+ Kf8 4.Qh6+ Ke8 5.Qh8+ Qf8 6.Qxf6 Bxc4 7.a3 Rxd1+ 8.Rxd1 Be7 9.Qe5 Bd5 10.Rxd5 cxd5 11.Qxd5 Qg7 12.Qxb7 (3.596.898.933) 1417
18.01 42:18 +2.37 1.Bxh6 Rfd8 2.Bxg7 Kxg7 3.Qg5+ Kf8 4.Qh6+ Ke8 5.Qh8+ Qf8 6.Qxf6 Bxc4 7.a3 Rxd1+ 8.Rxd1 Be7 9.Qe5 Bd5 10.Rxd5 cxd5 11.Qxd5 Qg7 12.Qxb7 (3.596.898.933) 1417
best move: Bc1xh6 time: 42:18.312 min n/s: 1.417.050 nodes: 3.596.898.933
Top
User avatar
Ovyron
Posts: 4562
Joined: Tue Jul 03, 2007 4:30 am

Re: Anudder test position--too easy for commercials

Post by Ovyron »

Freeware Bright 0.2c:

CPU0: AuthenticAMD x86 Family 6 Model 10 Stepping 0 2133 MHz
<snip>
10/30 0:02 +0.60 1.Qg3 Nh5 2.Qe5 f5 3.a3 Bc5 4.Re1 Rae8 5.Be3 Bxe3 6.Rxe3 (2.016.091) 0
11/34 0:05 +0.56 1.Qg3 Nh5 2.Qe5 Nf6 3.a3 Bc5 4.b4 Bb6 5.Bf4 Rfd8 6.c5 (4.664.455) 0
12/36 0:17 +0.37 1.Qg3 Kh8 2.a3 Bc5 3.b4 Bb6 4.c5 Bc7 5.Ne5 Bg4 6.Re1 Rfe8 (14.824.767) 0
12/44 0:44 +0.57 1.a3 Bc5 2.Re1 Nd7 3.Bg5 Nf6 4.Bxf6 Qxf6 5.Qxf6 gxf6 6.b4 Bd6 7.c5 Bf4 8.Rad1 (36.375.922) 0
13/44 0:57 +0.61 1.a3 Bc5 2.Re1 Rae8 3.Bf4 Bd6 4.Nd4 Bxf4 5.Qxf4 Qc5 6.Nxe6 Rxe6 7.Rxe6 fxe6 8.Rd1 (44.693.811) 0
13/49 1:43 +1.39 1.Bxh6 gxh6 2.Qxh6 Rfd8 3.Ng5 Bd6 4.Bh7+ Kh8 5.Re1 Re8 6.Rad1 Bf4 7.Bf5+ Kg8 8.Bxe6 fxe6 9.Rxe6 (75.283.365) 0
Top
Will Singleton
Posts: 128
Joined: Thu Mar 09, 2006 5:14 pm
Location: Los Angeles, CA

Re: Anudder test position--too easy for commercials

Post by Will Singleton »

Not so easy for some good progs. In fact, this might be a good position to test the proposition that the better the program, the worse it performs.

Fruit 2.2.1 not found in 10 minutes
Rybka 1.0 not found in 10 minutes
(note the symmetry)

Shredder 7.0.4 2:34 (2cpu)
Amateur 2.86 1:55

amd 2.2 ghz
Top
Spock

Re: Arasan 10.4

Post by Spock »

Found by Rotor 0.4 by Jan Brouwer
(10 mins on QX 6700)

Code: Select all

 10/16	 0:01 	+0.26 	1.Nd4 Rad8 2.Nxe6 Rxd1+ 3.Bxd1 Qxe6 4.Be3 Bd6 5.Bf3 Qe5 6.Rd1 (2.386.259) 2272
 11/18	 0:02 	+0.17 	1.Nd4 Rad8 2.Be3 Bd7 3.Nb3 a6 4.c5 Nd5 5.Qxe7 Nxe7 6.Be4 (4.954.447) 2264
 12/19	 0:09 	+0.13 	1.Nd4 Rfd8 2.Qg3 Bd6 3.Qb3 Bc8 4.Nf5 Bxf5 5.Bxf5 Bh2+ 6.Kh1 Rxd1+ 7.Qxd1 Qe5 (24.600.021) 2469
 12/19	 0:11 	+0.20 	1.Be3 Bd6 2.Bxh6 gxh6 3.Rxd6 Qxd6 4.Qxf6 Rfd8 5.Ne5 Qc5 6.Bd3 Rac8 (28.249.796) 2481
 13/20	 0:19 	+0.16 	1.Be3 Bc5 2.Nd4 Bd7 3.Qg3 Ne4 4.Bxe4 Qxe4 5.Nf5 Bxf5 6.Bxc5 Rfd8 7.Qc3 (48.902.678) 2531
 13/21	 0:21 	+0.17 	1.Nd4 Rfd8 2.Qg3 Bd6 3.Qb3 Bc8 4.Be3 Qc7 5.Nf5 Bh2+ 6.Kf1 Re8 7.Qb4 Ne4 (53.374.464) 2538
 13/21	 0:33 	+0.19 	1.Bf4 Rad8 2.Nd4 Bc5 3.Nxe6 Qxe6 4.Rxd8 Rxd8 5.Rc1 Bd6 6.Be3 Qe5 7.Bd4 (84.952.772) 2562
 14/23	 0:50 	+0.17 	1.Bf4 Rad8 2.Bd3 Bd6 3.Bxd6 Rxd6 4.Qf4 Rfd8 5.Qe3 a6 6.c5 Rd5 7.Rac1 Qf8 (130.801.937) 2570
 14/23	 0:57 	+0.24 	1.Nd4 Rfd8 2.a3 Bc5 3.Be3 Bd7 4.b4 Bb6 5.Re1 Nd5 6.Qxe7 Nxe7 7.Rad1 Bc7 (147.952.476) 2572
 15/27	 1:29 	+0.26 	1.Nd4 Rfd8 2.Be3 Bd7 3.a3 Bc5 4.b4 Bb6 5.Re1 Re8 6.Bd2 Qd8 7.Rxe8+ Bxe8 8.c5 Bc7 (228.562.406) 2565
 15/27	 2:23 	+0.40 	1.a3 Bc5 2.Re1 Nd7 3.Qxe7 Bxe7 4.Nd4 Bc5 5.Nxe6 fxe6 6.Re2 Bd4 7.Rd2 e5 8.Re2 (366.479.755) 2562
 16/26	 2:30 	+0.39 	1.a3 Bc5 2.Re1 Nd7 3.Qxe7 Bxe7 4.Nd4 Bc5 5.Nxe6 fxe6 6.Re2 Bd4 7.Rd2 e5 8.Re2 Nc5 (385.827.871) 2564
 17/31	 5:00 	+1.15 	1.a3 Bc5 2.Bxh6 Ng4 3.Qxe7 Bxe7 4.Bg5 Nf6 5.Bd3 Rad8 6.Be3 c5 7.Bf4 Rxd3 8.Rxd3 Bxc4 9.Rdd1 (765.984.627) 2552
 17/32	10:13 	+2.00 	1.Bxh6 Rfd8 2.Bg5 Rxd1+ 3.Rxd1 Rd8 4.Rxd8+ Qxd8 5.Nd4 Bd6 6.Nxe6 fxe6 7.Qh7+ Nxh7 8.Bxd8 Nf6 9.Bg6 Be5 10.b4 Kf8 (1.559.852.777) 2542
best move: Bc1xh6 time: 11:52.129 min  n/s: 2.542.340  nodes: 1.559.852.777



....
Top
User avatar
Zach Wegner
Posts: 1922
Joined: Thu Mar 09, 2006 12:51 am
Location: Earth

Re: Anudder test position--too easy for commercials

Post by Zach Wegner »

Will Singleton wrote:Not so easy for some good progs. In fact, this might be a good position to test the proposition that the better the program, the worse it performs.

Fruit 2.2.1 not found in 10 minutes
Rybka 1.0 not found in 10 minutes
(note the symmetry)

Shredder 7.0.4 2:34 (2cpu)
Amateur 2.86 1:55

amd 2.2 ghz
I think you might be on to something!!

Code: Select all

Depth          Time     Score   PV (nodes)
  5/11        0.015     +0.85   1. a3 Ba5 2. Be3 Bc7 3. Kh1 (7614)
  5/11        0.028     +1.55   1. Bxh6 gxh6 2. Qxh6 Bc5 3. Ne5 (13869)
[ 5/11]       0.031     +1.55   1. Bxh6 gxh6 2. Qxh6 Bc5 3. Ne5 (15426)
  6/11        0.041     +1.53   1. Bxh6 gxh6 2. Qxh6 Bc5 3. Ne5 Rac8 (20824)
[ 6/11]       0.057     +1.53   1. Bxh6 gxh6 2. Qxh6 Bc5 3. Ne5 Rac8 (29608)
  7/16        0.106     +1.53   1. Bxh6 gxh6 2. Qxh6 a5 3. Ne5 a4 4. Kh1 (54113)
[ 7/16]       0.120     +1.53   1. Bxh6 gxh6 2. Qxh6 a5 3. Ne5 a4 4. Kh1 (64036)
  8/18        0.230     +1.53   1. Bxh6 gxh6 2. Qxh6 Bc5 3. Bd3 Rac8 4.
                                Rdc1 Bd6 (124855)
[ 8/18]       0.271     +1.53   1. Bxh6 gxh6 2. Qxh6 Bc5 3. Bd3 Rac8 4.
                                Rdc1 Bd6 (147796)
  9/19        0.650     +1.63   1. Bxh6 gxh6 2. Qxh6 Bc5 3. Re1 Bb6 4. b4
                                Bc7 5. Ne5 (343588)
[ 9/19]       0.699     +1.63   1. Bxh6 gxh6 2. Qxh6 Bc5 3. Re1 Bb6 4. b4
                                Bc7 5. Ne5 (376122)
 10/23        1.554     +1.65   1. Bxh6 gxh6 2. Qxh6 Rab8 3. a3 Bc5 4. Ng5
                                Bd7 5. Bh7+ Kh8 6. Bf5+ Kg8 7. Bxd7 Nxd7
                                (827342)
[10/23]       1.724     +1.65   1. Bxh6 gxh6 2. Qxh6 Rab8 3. a3 Bc5 4. Ng5
                                Bd7 5. Bh7+ Kh8 6. Bf5+ Kg8 7. Bxd7 Nxd7
                                (926750)
 11/23        5.463     +2.07   1. Bxh6 gxh6 2. Qxh6 Rad8 3. Rxd8 Rxd8 4.
                                Ng5 Bd2 5. Bh7+ Kh8 6. Bf5+ Kg8 7. Bxe6
                                fxe6 8. Qg6+ (2883592)
[11/23]       5.673     +2.07   1. Bxh6 gxh6 2. Qxh6 Rad8 3. Rxd8 Rxd8 4.
                                Ng5 Bd2 5. Bh7+ Kh8 6. Bf5+ Kg8 7. Bxe6
                                fxe6 8. Qg6+ (3017082)
 12/25       14.377     +2.49   1. Bxh6 gxh6 2. Qxh6 Rad8 3. Ng5 Rxd1+ 4.
                                Rxd1 Rb8 5. Bh7+ Kh8 6. Bf5+ Kg8 7. Nxe6
                                fxe6 8. Bc2 (7543296)
[12/25]      14.562     +2.49   1. Bxh6 gxh6 2. Qxh6 Rad8 3. Ng5 Rxd1+ 4.
                                Rxd1 Rb8 5. Bh7+ Kh8 6. Bf5+ Kg8 7. Nxe6
                                fxe6 8. Bc2 (7660317)
Top
Spock

Re: Arasan 10.4

Post by Spock »

Found by Rybka 1.0 x64, but took longer than Rotor did

Code: Select all

 10.00	 0:01 	+0.36 	1.Qg3 Kh8 2.a3 Bc5 3.b4 Bb6 4.c5 (281.821) 210
 11.01	 0:02 	+0.37 	1.Qg3 Kh8 2.a3 Bc5 3.b4 Bb6 4.c5 Bc7 (579.763) 211
 12.01	 0:05 	+0.33 	1.Qg3 Kh8 2.a3 Bc5 3.b4 Bb6 4.Ne5 Rfe8 5.c5 (1.219.479) 213
 13.01	 0:14 	+0.35 	1.Qg3 Kh8 2.a3 Bc5 3.b4 Bb6 4.Ne5 Rfe8 5.c5 Bc7 (2.993.501) 213
 14.01	 0:56 	+0.42 	1.Qg3 Kh8 2.Be3 Rad8 3.Ne5 Nh5 4.Qf3 Nf6 5.Bxa7 Bxh3 (10.984.983) 199
 14.03	 1:41 	+1.39 	1.a3 Bc5 2.Bxh6 gxh6 3.Qxh6 Rfd8 4.Re1 Bd6 5.Nh4 Qf8 (18.932.126) 193
 15.01	 3:04 	+1.85 	1.a3 Bc5 2.Bxh6 gxh6 3.Qxh6 Rfe8 4.Re1 Qf8 5.Qh4 Bd7 (33.396.012) 185
 16.01	 4:13 	+1.82 	1.a3 Bc5 2.Bxh6 gxh6 3.Qxh6 Rfe8 4.Re1 Qf8 5.Qh4 Bxf2+ (46.968.825) 190
 17.01	 6:37 	+1.77 	1.a3 Bc5 2.Bxh6 gxh6 3.Qxh6 Rfe8 4.Re1 Qf8 5.Qh4 Bxc4 (74.807.126) 192
 17.06	16:35 	+2.29 	1.Bxh6 gxh6 2.Qxh6 (193.640.278) 199
best move: Bc1xh6 time: 16:45.333 min  n/s: 199.238  nodes: 195.452.928

..
Top
Will Singleton
Posts: 128
Joined: Thu Mar 09, 2006 5:14 pm
Location: Los Angeles, CA

Re: Anudder test position--too easy for commercials

Post by Will Singleton »

Zach Wegner wrote:
Will Singleton wrote:Not so easy for some good progs. In fact, this might be a good position to test the proposition that the better the program, the worse it performs.

Fruit 2.2.1 not found in 10 minutes
Rybka 1.0 not found in 10 minutes
(note the symmetry)

Shredder 7.0.4 2:34 (2cpu)
Amateur 2.86 1:55

amd 2.2 ghz
I think you might be on to something!!

Code: Select all

Depth          Time     Score   PV (nodes)
  5/11        0.015     +0.85   1. a3 Ba5 2. Be3 Bc7 3. Kh1 (7614)
  5/11        0.028     +1.55   1. Bxh6 gxh6 2. Qxh6 Bc5 3. Ne5 (13869)
[ 5/11]       0.031     +1.55   1. Bxh6 gxh6 2. Qxh6 Bc5 3. Ne5 (15426)
  6/11        0.041     +1.53   1. Bxh6 gxh6 2. Qxh6 Bc5 3. Ne5 Rac8 (20824)
[ 6/11]       0.057     +1.53   1. Bxh6 gxh6 2. Qxh6 Bc5 3. Ne5 Rac8 (29608)
  7/16        0.106     +1.53   1. Bxh6 gxh6 2. Qxh6 a5 3. Ne5 a4 4. Kh1 (54113)
[ 7/16]       0.120     +1.53   1. Bxh6 gxh6 2. Qxh6 a5 3. Ne5 a4 4. Kh1 (64036)
  8/18        0.230     +1.53   1. Bxh6 gxh6 2. Qxh6 Bc5 3. Bd3 Rac8 4.
                                Rdc1 Bd6 (124855)
[ 8/18]       0.271     +1.53   1. Bxh6 gxh6 2. Qxh6 Bc5 3. Bd3 Rac8 4.
                                Rdc1 Bd6 (147796)
  9/19        0.650     +1.63   1. Bxh6 gxh6 2. Qxh6 Bc5 3. Re1 Bb6 4. b4
                                Bc7 5. Ne5 (343588)
[ 9/19]       0.699     +1.63   1. Bxh6 gxh6 2. Qxh6 Bc5 3. Re1 Bb6 4. b4
                                Bc7 5. Ne5 (376122)
 10/23        1.554     +1.65   1. Bxh6 gxh6 2. Qxh6 Rab8 3. a3 Bc5 4. Ng5
                                Bd7 5. Bh7+ Kh8 6. Bf5+ Kg8 7. Bxd7 Nxd7
                                (827342)
[10/23]       1.724     +1.65   1. Bxh6 gxh6 2. Qxh6 Rab8 3. a3 Bc5 4. Ng5
                                Bd7 5. Bh7+ Kh8 6. Bf5+ Kg8 7. Bxd7 Nxd7
                                (926750)
 11/23        5.463     +2.07   1. Bxh6 gxh6 2. Qxh6 Rad8 3. Rxd8 Rxd8 4.
                                Ng5 Bd2 5. Bh7+ Kh8 6. Bf5+ Kg8 7. Bxe6
                                fxe6 8. Qg6+ (2883592)
[11/23]       5.673     +2.07   1. Bxh6 gxh6 2. Qxh6 Rad8 3. Rxd8 Rxd8 4.
                                Ng5 Bd2 5. Bh7+ Kh8 6. Bf5+ Kg8 7. Bxe6
                                fxe6 8. Qg6+ (3017082)
 12/25       14.377     +2.49   1. Bxh6 gxh6 2. Qxh6 Rad8 3. Ng5 Rxd1+ 4.
                                Rxd1 Rb8 5. Bh7+ Kh8 6. Bf5+ Kg8 7. Nxe6
                                fxe6 8. Bc2 (7543296)
[12/25]      14.562     +2.49   1. Bxh6 gxh6 2. Qxh6 Rad8 3. Ng5 Rxd1+ 4.
                                Rxd1 Rb8 5. Bh7+ Kh8 6. Bf5+ Kg8 7. Nxe6
                                fxe6 8. Bc2 (7660317)

Are the results below that of ZCT? My copy doesn't replicate those results.
Top
User avatar
Zach Wegner
Posts: 1922
Joined: Thu Mar 09, 2006 12:51 am
Location: Earth

Re: Anudder test position--too easy for commercials

Post by Zach Wegner »

Yes, it's the current development version (top secret!!), just one processor though. I went through some old versions, and the first one to get this quickly was 0.3.2475. There are a few new things in this version that might help: different (very simple) pawn eval, higher queen mobility, different definition of game phase, and unlimited search extensions. I don't really know what the real issue is here. I'd think extensions, but the selective depth is the same with .2474.
Top
Cubeman
Posts: 644
Joined: Fri Feb 02, 2007 3:11 am
Location: New Zealand

Re: Anudder test position--too easy for commercials

Post by Cubeman »

PocketFritz3 with Hiarcs 12.1 for some strange reason finds 1.Bxh6 in less than 1 second.And gives score (1.16) at Depth=6 00:00.
It later climbs up to (2.19) at D=11 01:31.
Hiarcs does well in these castled king sacrifice positions.
Top

Return to “Computer Chess Club: General Topics”