Kamsky-Yussupow, Tilburg 1992

[PauloSoare]

Eelco, Zappa Mexico II ,withouth endgame tablebases, analyses from the initial position.
Hardware: Q6600, HT= 4Gb.
Yesterday I made some analysis with computer help and f4 it´s a move that wins, unhapily I didn´t save the analysis.

Paulo Soares

10/26 0:00 +2.27 1.f4+ Kxf4 2.c6 Ke5 3.c7 Rh3+ 4.Kc2 Rh8 5.Rc6 Rc8 6.Kd3 d4 7.cxd4+ (602.712) 1925
10/26 0:00 +2.27 1.f4+ Kxf4 2.c6 Ke5 3.c7 Rh3+ 4.Kc2 Rh8 5.Rc6 Rc8 6.Kd3 d4 7.cxd4+ (740.296) 1974
11/30 0:00 +2.84 1.f4+ Kxf4 2.Ra8 Ke5 3.Re8+ Kf6 4.c6 Rh7 5.Kd4 Rh4+ 6.Kxd5 Rc4 7.Kd6 Kf7 8.Re7+ (1.390.362) 2224
11/30 0:00 +2.84 1.f4+ Kxf4 2.Ra8 Ke5 3.Re8+ Kf6 4.c6 Rh7 5.Kd4 Rh4+ 6.Kxd5 Rc4 7.Kd6 Kf7 8.Re7+ (1.468.615) 2185
12/34 0:00 +3.11 1.f4+ Kxf4 2.Ra8 Kf5 3.c6 Rh7 4.Kd4 Ke6 5.Kc5 Rh4 6.Re8+ Kf7 7.Rb8 Rc4+ 8.Kd6 d4 9.cxd4 Rxd4+ (2.191.935) 2262
12/34 0:01 +3.11 1.f4+ Kxf4 2.Ra8 Kf5 3.c6 Rh7 4.Kd4 Ke6 5.Kc5 Rh4 6.Re8+ Kf7 7.Rb8 Rc4+ 8.Kd6 d4 9.cxd4 Rxd4+ (2.347.123) 2241
13/37 0:01 +2.66 1.f4+ Kxf4 2.Ra8 Kf5 3.c6 Rh7 4.Kd4 Ke6 5.Kc5 Rh3 6.c4 Rc3 7.Kb6 Rxc4 8.Re8+ (3.794.774) 2335
13/37 0:01 +2.66 1.f4+ Kxf4 2.Ra8 Kf5 3.c6 Rh7 4.Kd4 Ke6 5.Kc5 Rh3 6.c4 Rc3 7.Kb6 Rxc4 8.Re8+ (4.316.642) 2361
14/38 0:02 +2.66 1.f4+ Kxf4 2.Ra8 Kf5 3.c6 Rh7 4.Kd4 Ke6 5.Kc5 Rh3 6.c4 Rc3 7.Kb6 Rxc4 8.Re8+ (4.945.778) 2397
14/38 0:02 +2.66 1.f4+ Kxf4 2.Ra8 Kf5 3.c6 Rh7 4.Kd4 Ke6 5.Kc5 Rh3 6.c4 Rc3 7.Kb6 Rxc4 8.Re8+ (5.502.623) 2395
15/38 0:05 +2.44 1.f4+ Kxf4 2.Kd4 Kf5 3.Kxd5 Rh3 4.Kc6 Ke5 5.Kxb5 Rxc3 6.Rd6 Rc1 7.Kb6 Rb1+ 8.Kc7 Rc1 (13.067.317) 2526
15/38 0:06 +2.44 1.f4+ Kxf4 2.Kd4 Kf5 3.Kxd5 Rh3 4.Kc6 Ke5 5.Kxb5 Rxc3 6.Rd6 Rc1 7.Kb6 Rb1+ 8.Kc7 Rc1 (16.686.838) 2573
16/38 0:08 +2.64 1.f4+ Kxf4 2.Kd4 Kf5 3.Kxd5 Rh3 4.Kc6 Ke5 5.Kxb5 Rxc3 6.Rd6 Rc1 7.Kb6 Rb1+ 8.Kc7 Ra1 9.Rd7 Ra7+ (23.432.133) 2682
16/38 0:09 +2.64 1.f4+ Kxf4 2.Kd4 Kf5 3.Kxd5 Rh3 4.Kc6 Ke5 5.Kxb5 Rxc3 6.Rd6 Rc1 7.Kb6 Rb1+ 8.Kc7 Ra1 9.Rd7 Ra7+ (26.463.994) 2662
17/46 0:18 +3.26 1.f4+ Kxf4 2.Kd4 Kf5 3.Rd6 Rh4+ 4.Kxd5 Rc4 5.Kc6 Rxc3 6.Kxb5 Ke5 7.Kb6 Rc1 8.Rd8 Rb1+ 9.Kc7 Rc1 10.c6 Ke6 11.Kb7 (51.102.446) 2812
17/46 0:20 +3.26 1.f4+ Kxf4 2.Kd4 Kf5 3.Rd6 Rh4+ 4.Kxd5 Rc4 5.Kc6 Rxc3 6.Kxb5 Ke5 7.Kb6 Rc1 8.Rd8 Rb1+ 9.Kc7 Rc1 10.c6 Ke6 11.Kb7 (56.504.599) 2794
18/46 0:27 +3.28 1.f4+ Kxf4 2.Kd4 Kf5 3.Rd6 Rh4+ 4.Kxd5 Rc4 5.Kc6 Rxc3 6.Kxb5 Ke5 7.Kb6 Rb3+ 8.Kc7 Ra3 9.Rd8 Ke6 10.c6 Ke7 11.Kb7 Kxd8 (79.654.856) 2926
18/46 0:30 +3.28 1.f4+ Kxf4 2.Kd4 Kf5 3.Rd6 Rh4+ 4.Kxd5 Rc4 5.Kc6 Rxc3 6.Kxb5 Ke5 7.Kb6 Rb3+ 8.Kc7 Ra3 9.Rd8 Ke6 10.c6 Ke7 11.Kb7 Kxd8 (89.162.074) 2914
19/46 0:48 +3.30 1.f4+ Kxf4 2.Kd4 Kf5 3.Rd6 Rh4+ 4.Kxd5 Rc4 5.Kc6 Rxc3 6.Kxb5 Ke5 7.Kb6 Rb3+ 8.Kc7 Ra3 9.Rd8 Ke6 10.c6 Ke7 11.Kb7 Kxd8 12.c7+ (149.637.488) 3095
19/46 0:55 +3.30 1.f4+ Kxf4 2.Kd4 Kf5 3.Rd6 Rh4+ 4.Kxd5 Rc4 5.Kc6 Rxc3 6.Kxb5 Ke5 7.Kb6 Rb3+ 8.Kc7 Ra3 9.Rd8 Ke6 10.c6 Ke7 11.Kb7 Kxd8 12.c7+ (170.072.582) 3068
20/46 1:29 +3.31 1.f4+ Kxf4 2.Kd4 Kf5 3.Rd6 Rh4+ 4.Kxd5 Rc4 5.Kc6 Rxc3 6.Kxb5 Ke5 7.Kb6 Rb3+ 8.Kc7 Ra3 9.Rd8 Ke6 10.c6 Ke7 11.Rd5 Rc3 12.Re5+ Kf6 13.Kc8 Kxe5 (287.365.075) 3228
20/46 1:42 +3.31 1.f4+ Kxf4 2.Kd4 Kf5 3.Rd6 Rh4+ 4.Kxd5 Rc4 5.Kc6 Rxc3 6.Kxb5 Ke5 7.Kb6 Rb3+ 8.Kc7 Ra3 9.Rd8 Ke6 10.c6 Ke7 11.Rd5 Rc3 12.Re5+ Kf6 13.Kc8 Kxe5 (328.248.477) 3197
21/56 8:50 +4.00 1.f4+ Kxf4 2.Ra8 Kf5 3.c6 Rh7 4.Kd4 Ke6 5.Kc5 Rh3 6.c7 Rxc3+ 7.Kb6 Rxc7 8.Kxc7 d4 9.Rd8 Ke5 10.Kb6 b4 11.Kc5 b3 12.Re8+ Kf4 13.Kxd4 b2 14.Rb8 (770.705.858) 1453
21/56 9:54 +4.00 1.f4+ Kxf4 2.Ra8 Kf5 3.c6 Rh7 4.Kd4 Ke6 5.Kc5 Rh3 6.c7 Rxc3+ 7.Kb6 Rxc7 8.Kxc7 d4 9.Rd8 Ke5 10.Kb6 b4 11.Kc5 b3 12.Re8+ Kf4 13.Kxd4 b2 14.Rb8 (828.889.472) 1393
22/56 20:57 +4.05 1.f4+ Kxf4 2.Ra8 Kf5 3.c6 Rh7 4.Kd4 Ke6 5.Kc5 Rh3 6.c7 Rxc3+ 7.Kb6 Rxc7 8.Kxc7 d4 9.Rd8 Ke5 10.Kb6 b4 11.Kc5 b3 12.Re8+ Kf6 13.Kxd4 b2 14.Rb8 (1.502.745.286) 1195
22/56 22:22 +4.05 1.f4+ Kxf4 2.Ra8 Kf5 3.c6 Rh7 4.Kd4 Ke6 5.Kc5 Rh3 6.c7 Rxc3+ 7.Kb6 Rxc7 8.Kxc7 d4 9.Rd8 Ke5 10.Kb6 b4 11.Kc5 b3 12.Re8+ Kf6 13.Kxd4 b2 14.Rb8 (1.582.928.123) 1179
23/59 46:02 +4.11 1.f4+ Kxf4 2.Ra8 Kf5 3.c6 Rh7 4.Kd4 Ke6 5.Kc5 Rh3 6.c7 Rxc3+ 7.Kb6 Rxc7 8.Kxc7 d4 9.Rd8 Ke5 10.Kb6 b4 11.Kc5 b3 12.Re8+ Kf6 13.Kxd4 Kf5 14.Rb8 (3.044.409.770) 1102
23/59 48:25 +4.11 1.f4+ Kxf4 2.Ra8 Kf5 3.c6 Rh7 4.Kd4 Ke6 5.Kc5 Rh3 6.c7 Rxc3+ 7.Kb6 Rxc7 8.Kxc7 d4 9.Rd8 Ke5 10.Kb6 b4 11.Kc5 b3 12.Re8+ Kf6 13.Kxd4 Kf5 14.Rb8 (3.176.935.411) 1093
24/61 91:50 +4.11 1.f4+ Kxf4 2.Ra8 Rh6 3.Kd4 Kf5 4.Re8 Rh4+ 5.Kxd5 Rh3 6.Rf8+ Kg6 7.Kc6 Rxc3 8.Kxb5 Rb3+ 9.Kc4 Rb2 10.Rf3 Rc2+ 11.Rc3 Rh2 12.c6 Rh8 13.Rf3 Rb8 (5.840.217.372) 1059
24/61 99:23 +4.11 1.f4+ Kxf4 2.Ra8 Rh6 3.Kd4 Kf5 4.Re8 Rh4+ 5.Kxd5 Rh3 6.Rf8+ Kg6 7.Kc6 Rxc3 8.Kxb5 Rb3+ 9.Kc4 Rb2 10.Rf3 Rc2+ 11.Rc3 Rh2 12.c6 Rh8 13.Rf3 Rb8 (6.246.031.597) 1047

[jdart]

for Glaurung does a nullmove search that is 5 plies shorter than the normal search, in Toga Checkov it was the same, R=5, and verification search I believe was even more shortened, 7 or 8 plies.

The Toga and Glaurung I have looked at (Toga 1.4b5 and Glaurung v2) just do R=3 (so total depth reduction of 4). Toga also allowed double null moves. But maybe they are different now.

Arasan has consistently used adaptive null reduction with R=2 or 3 depending on depth. It also does a verification search in the endgame.

I have tried other reduction depths, along with other more aggressive search techniques but always have gotten worse results on my test suites. But Toga does well with its search. The problem is, everything interacts (scoring, extensions, null move, etc.) so it is hard to find the best combination.

[bob]

I'm not getting to 14 ply in 1 second on this, not even on a quad.

For some reason my branching factor isn't that high on this position. I have null move pretty much disabled for this because of the low material.

--Jon

what about reductions?

Here's crafty on my laptop. It is 4-5x faster on the 8-core machine I normally use for tournaments. It gets to 14 in a couple of seconds...

Code: Select all

              12->   0.70   2.45   1. Ra8 Rh4 2. f4+ Kxf4 3. Kd4 Kf3+
                                    4. Kxd5 Rc4 5. Kc6 Rxc3 6. Kxb5 Ke2
                                    7. Rd8 Rb3+ 8. Kc6 (s=3)
               13     1.04   2.32   1. Ra8 Ke6 2. Re8+ Kd7 3. Rf8 Rh2 4.
                                    Rf6 Ke7 5. Rd6 Rf2 6. Ke3 Rc2 7. Kd4
                                    Rf2 (s=2)
               13     1.64   2.38   1. f4+ Kxf4 2. Kd4 Kf3 3. Rb6 Ke2 4.
                                    Rxb5 Rh4+ 5. Kxd5 Kd3 6. Rb3 Rh5+ 7.
                                    Kd6 Kc4 8. Rb4+ Kxc3 <HT>
               13->   1.89   2.38   1. f4+ Kxf4 2. Kd4 Kf3 3. Rb6 Ke2 4.
                                    Rxb5 Rh4+ 5. Kxd5 Kd3 6. Rb3 Rh5+ 7.
                                    Kd6 Kc4 8. Rb4+ Kxc3 <HT> (s=2)
               14     2.31   2.32   1. f4+ Kxf4 2. Kd4 Kf3 3. Rb6 Ke2 4.
                                    Rxb5 Rh4+ 5. Kxd5 Kd3 6. Rb3 Rh5+ 7.
                                    Kd6 Kc4 8. Rb4+ Kxc3 9. Re4 Rf5

[jdart]

Arasan 10.4 on a Q6700, 3.3Ghz:

"Kamsky-Jussupow, Tilburg 1992" bm f4+
1 0.00 Ke3 +2.16 27
2 0.01 c6 +2.16 159
3 0.01 c6 +2.28 789
4 0.01 c6 +2.19 2065
5 0.01 Rb6 +2.25 9220
6 0.03 Rb6 +2.50 13991
7 0.03 Rb6 +2.53 28216
8 0.09 Rb6 +2.20 123556
9 0.14 Rb6 +2.22 224450
10 0.56 Rg6 +2.16 1075928
11 1.48 Rb6 +2.14 2917182
12 2.89 Rb6 +2.20 5101170
13 11.40 f4+ +2.38 17698415
14 13.03 f4+ +2.41 20704861
15 15.85 f4+ +2.44 24906190
16 21.28 f4+ +2.63 33715840
17 41.75 f4+ +2.75 63706832
1.51M nodes/second.
35844609 regular nodes, 54651434 quiescence nodes.
38372960 searches of hash table, 25091952 successful (65 percent).
10367291 hash entries inserted, 9958474 entries replaced, 1096539 inserts fai
.
null cutoffs: 4248770 (4%)
pruning: 63316416 moves (futility), 0 nodes razored
extensions: 7569992 check, 0 recapture, 1034075 pawn,
15238 threat, 43884 forced move.
79309 tablebase probes, 79308 tablebase hits
156 splits, average thread usage=2.29
position hash hit rate = 28.62%
pawn hash hit rate = 100.02%
king/pawn hash hit rate= 86.65%
king cover hash hit rate= 99.11%
lazy cutoff in eval = 19.95%
result: f4+ 60 seconds. score: +2.75 90496043 nodes. ++ correct
f4+ Kxf4 Kd4 Kf5 Kxd5 Kf4+ Kc6 Rh8 Kxb5 Rb8+ Rb6 Rc8 c6 Ke3 Rb7 Kd3 c4 Ra8 c7
8

solution times:
0 1 2 3 4 5 6 7 8 9

0 | 11.40

correct : 1/1
nodes to solution : 17.70M
depth to solution : 12.00
time to solution : 11.40 sec.
test complete

So not terrible but not really good either. NPS is lowered a bit because it is hitting TBs.

[PauloSoare]

Eelco, impressing. Zappa Mexico II, in the same conditions of my last post, but with 5 pieces endgames and some 6.
I was making the moves suggested by Zappa with analysis between 3 and 5 min. When Zappa found the mate, I was making the moves back to the beginning, and I came to the analysis below.


20 1:14 +M41 1.f4+ Kf5 2.Rd6 Rh4 3.Kd4 Rh1 4.Rxd5+ Ke6 5.f5+ Kf6 (139.226.225) 1869 TB:21.977
20 1:43 +2.00 1.Ra8 Ke6 2.f4 Rh4 3.Rf8 Kd7 4.Rf5 Rh1 5.Rxd5+ Kc6 6.Re5 Rd1+ 7.Ke2 Rc1 8.Kd2 Rf1 9.Ke3 Re1+ 10.Kd2 Rf1 (196.534.240) 1902 TB:34.318

Paulo Soares

[Eelco de Groot]

Eelco, impressing. Zappa Mexico II, in the same conditions of my last post, but with 5 pieces endgames and some 6.
I was making the moves suggested by Zappa with analysis between 3 and 5 min. When Zappa found the mate, I was making the moves back to the beginning, and I came to the analysis below.


20 1:14 +M41 1.f4+ Kf5 2.Rd6 Rh4 3.Kd4 Rh1 4.Rxd5+ Ke6 5.f5+ Kf6 (139.226.225) 1869 TB:21.977
20 1:43 +2.00 1.Ra8 Ke6 2.f4 Rh4 3.Rf8 Kd7 4.Rf5 Rh1 5.Rxd5+ Kc6 6.Re5 Rd1+ 7.Ke2 Rc1 8.Kd2 Rf1 9.Ke3 Re1+ 10.Kd2 Rf1 (196.534.240) 1902 TB:34.318

Paulo Soares

Paulo thanks very much, that helped me a lot because so far Ancalagon's analysis seemed only to lead to drawn positions with ridiculous evals because of the passed pawns. And I started to believe maybe it was a draw...

Fantastic analysis from Zappa!

This is what I got with Build No. 4:

8/8/R7/1pPpk2r/8/2PK1P2/8/8 w - -

Engine: Ancalagon 1.08 Beta Build 4 (Athlon 2009 MHz, 256 MB)
by Tord Romstad, Eelco de Groot

2.00 0:00 +9.70 1.c6 Kd6 (107) 0

3.00 0:00 +1.96 1.c6 Kd6 2.f4 Kc7 (371) 1

3.00 0:00 +10.76 1.f4+ Kxf4 2.c6 Ke5 (635) 2

4.00 0:00 +10.49 1.f4+ Kxf4 2.c6 Rh7 3.Kd4 (1.293) 4

5.00 0:00 +11.13 1.f4+ Kxf4 2.c6 Rh7 3.Kd4 Kf5 4.Kxd5 (2.849) 10

6.00 0:00 +11.13 1.f4+ Kxf4 2.c6 Rh7 3.Kd4 Kf5 4.Kxd5 (5.974) 21

7.00 0:00 +8.72 1.f4+ Kxf4 2.c6 Rh7 3.Kd2 Ke5 4.Ke3 Kd6 (10.653) 35

7.00 0:00 +8.96 1.Rg6 Rh3 2.c6 Rxf3+ 3.Kc2 Rf2+
4.Kb3 Rf7 5.Kb4 (28.328) 86

8.00 0:00 +8.60 1.Rg6 Rh3 2.c6 Rxf3+ 3.Kc2 Rf7 4.Kb3 Rc7
5.Kb4 (37.669) 109

9.00 0:00 +7.49 1.Rg6 Rh3 2.Ke2 Rh2+ 3.Ke3 Kf5 4.Rb6 b4
5.c6 bxc3 6.c7 d4+ 7.Kxd4 (91.771) 202

10.00 0:00 +6.86 1.Rg6 Rh4 2.c6 Rc4 3.f4+ Kf5 4.Rd6 Rxc6
5.Rxc6 Kxf4 6.Kd4 Kf5 7.Kxd5 (269.802) 345

10.00 0:00 +7.29 1.f4+ Kxf4 2.Ra8 Rh6 3.Kd4 Kg3
4.Rg8+ Kf3 5.Rc8 Kg4 6.Kxd5 Rh5+
7.Kd6 Rh6+ 8.Kd7 (371.764) 390

11.01 0:01 +7.25 1.f4+ Kxf4 2.c6 Rh6 3.Kd4 Rxc6
4.Rxc6 Kg5 5.Ke5 d4 6.Kxd4 Kg4 7.Re6 (617.476) 444

12.01 0:02 +9.31 1.f4+ Kf5 2.Kd4 Rh1 3.c6 Rd1+ 4.Kc5 Rd3
5.Kb4 d4 6.c4 bxc4 7.Kxc4 Ke4 8.f5 Rc3+
9.Kb4 Kxf5 (1.269.930) 444

13.01 0:04 +10.47 1.f4+ Kf5 2.Kd4 Rh7 3.c6 Ke6 4.f5+ Kxf5
5.Kxd5 Rh3 6.Rb6 Rd3+ 7.Kc5 Rxc3+
8.Kxb5 Ke6 9.Kb4 Rc1 (1.993.675) 476

14.01 0:08 +6.17 1.f4+ Kxf4 2.Kd4 Kf5 3.Rb6 Rh3 4.Rd6 Rh4+
5.Kxd5 Rh3 6.c6 Rxc3 7.Kd4 b4 8.Rh6 Kg5
9.Rh5+ Kxh5 (4.375.710) 497

15.01 0:17 +6.29 1.f4+ Kxf4 2.Kd4 Kf5 3.Rd6 Rh3 4.c6 Rh4+
5.Kxd5 Rc4 6.Rd7 Rxc3 7.Rf7+ Kg5
8.Ra7 Rxc6 9.Kxc6 b4 10.Kc5 (8.463.668) 497

16.01 0:25 +8.01 1.f4+ Kxf4 2.Kd4 Kf5 3.Rd6 Rh4+
4.Kxd5 Rh3 5.Kc6 Ke5 6.Kxb5 Rh7
7.Rg6 Kf5 8.Rb6 Rh5 9.Rd6 Ke5 (12.262.796) 483

17.01 0:41 +10.39 1.f4+ Kxf4 2.Kd4 Kf5 3.Rd6 Rh3 4.c6 Rh4+
5.Kxd5 Rc4 6.Rd8 Rxc3 7.Rf8+ Kg6
8.Rb8 Kf7 9.Kd6 Rd3+ 10.Ke5 Re3+
11.Kd4 Ke6 (19.858.852) 476

18.01 0:51 +10.52 1.f4+ Kxf4 2.Kd4 Kf5 3.Rd6 Rh3 4.c6 Rh4+
5.Kxd5 Rc4 6.Rd8 Rxc3 7.Rf8+ Kg6
8.Rb8 Kf6 9.Rb7 Rc1 10.Kd6 Rd1+
11.Kc5 Ke6 (25.270.094) 487

19.01 1:52 +11.45 1.f4+ Kxf4 2.Kd4 Kf5 3.Rd6 Rh3 4.c6 Rh4+
5.Kxd5 Rc4 6.Rd8 Rxc3 7.Rb8 Kf6
8.Kd6 Rd3+ 9.Kc7 Ke6 10.Rxb5 Rc3
11.Rb8 Ke7 (55.478.971) 492

20.01 2:31 +11.35 1.f4+ Kxf4 2.Kd4 Kf5 3.Rd6 Rh3 4.c6 Rh4+
5.Kxd5 Rc4 6.Rd8 Rxc3 7.Rb8 Kf6
8.Kd6 Rd3+ 9.Kc7 Ke6 10.Rxb5 Rc3
11.Rh5 Re3 12.Rh6+ Ke7 (75.417.447) 499

21.01 4:42 +11.62 1.f4+ Kxf4 2.Kd4 Kf5 3.Rd6 Rh3 4.c6 Rh4+
5.Kxd5 Rc4 6.Rd8 Rxc3 7.Rb8 Kf6
8.Kd6 Rd3+ 9.Kc7 Rb3 10.Kb7 b4 11.c7 Ke7
12.Rh8 Rb1 13.Rh7+ Ke6 14.Rh6+ (142.469.332) 503

22.01 17:53 +14.49 1.f4+ Kxf4 2.Kd4 Kf5 3.Kxd5 Kf4+
4.Kc6 Ke5 5.Ra2 Rh8 6.Re2+ Kf5 7.Kb7 Rh7+
8.Kb6 Kf6 9.Kxb5 Rb7+ 10.Ka6 Rg7
11.c6 Kf5 12.Ka5 Kf4 13.Kb6 Rg6
14.c4 (550.432.996) 512

23.01 38:33 +15.49 1.f4+ Kxf4 2.Kd4 Kf5 3.Kxd5 Kf4+
4.Kc6 Ke5 5.Ra1 Rh8 6.Kxb5 Rb8+
7.Kc6 Rb3 8.Re1+ Kf6 9.Kc7 Rxc3
10.c6 Kg7 11.Kb6 Rb3+ 12.Kc5 Rc3+
13.Kd6 Rd3+ 14.Ke7 (1.169.566.036) 505

best move: f3-f4 time: 45:20.657 min n/s: 506.596 nodes:
1.378.260.000

I have not really checked all the positions but I think now some of the final positions seemed actually won! Evals are very high though again in this version.. The Black Rook seemed not to defend the back rank in those positions I saw, not sure if White can actually force the Rook to go elsewhere and still manage one of the pawns to promote without being captured, that would be great if the evaluation function could find that out.

Thanks Paulo!

Eelco

[Eelco de Groot]

It's a silly Blitz game I suppose but Ancalagon managed to finally win a game against Naum 3.1

[Event "0 Minutes/Game + 2 Seconds/Move"]
[Site "Engine Match"]
[Date "2008.07.16"]
[Round "2"]
[White "Ancalagon 1.08 Beta Build 7"]
[Black "Naum 3.1"]
[Result "1-0"]
[ECO "B33 - Sicilian/Sveshnikov Variation"]

1. e4 {book 0s} c5 {book 0s} 2. Nf3 {book 0s} Nc6 {book 0s}
3. d4 {book 0s} cxd4 {book 0s} 4. Nxd4 {book 0s} Nf6 {book
0s} 5. Nc3 {book 0s} e5 {book 0s} 6. Ndb5 {book 0s} d6
{book 0s} 7. Nd5 {book 0s} Nxd5 {book 0s} 8. exd5 {book 0s}
Nb8 {book 0s} 9. c4 {book 0s} Be7 {book 0s} 10. Bd3 {book
0s} O-O {book 0s} 11. O-O {book 0s} f5 {book 0s} 12. c5
{book 0s} a6 {-0.47/11 3s} 13. Nxd6 {book 0s} Bxd6
{-0.42/12 4s} 14. cxd6 {book 0s} Qxd6 {-0.39/13 5s} 15. b3
{book 0s} Nd7 {-0.14/11 2s} 16. Be2 {+0.29/10 2s} b5
{+0.19/11 2s} 17. a4 {-0.11/10 2s} Bb7 {+0.17/10 1s}
18. axb5 {-0.25/10 6s} axb5 {+0.24/12 1s} 19. Rxa8
{+0.68/11 2s} Rxa8 {+0.12/12 1s (Bxa8)} 20. Bxb5 {+0.88/11
3s} Bxd5 {+0.08/12 2s (Nf6)} 21. Bc4 {+0.96/11 3s} Nf6
{+0.03/13 1s} 22. Qf3 {+1.64/10 2s} Qe6 {0.00/12 2s (Qc6)}
23. Rd1 {+0.56/11 5s} e4 {+0.09/12 1s} 24. Qc3 {+2.50/11
4s} Bxc4 {-0.05/11 6s} 25. bxc4 {+2.66/10 2s} Kf7 {-0.02/11
3s (f4)} 26. c5 {+3.62/9 2s} Ra7 {-0.09/11 4s (Rc8)}
27. Bb2 {+4.94/10 3s} Rd7 {-0.09/10 2s (Rc7)} 28. Rxd7+
{+3.50/10 3s} Qxd7 {-0.11/12 1s} 29. Qc1 {+2.13/11 3s} Qc6
{-0.05/11 2s} 30. Be5 {+2.25/10 2s} Nd5 {+0.01/12 3s (Ng4)}
31. h3 {+2.33/10 2s} g6 {0.00/11 2s (Ke8)} 32. Qc4
{+3.92/11 2s} Ke6 {0.00/13 2s (Ke8)} 33. Bf4 {+2.37/11 2s}
h5 {-0.10/13 2s (Qa8)} 34. Kh2 {+4.68/11 2s} Kd7 {-0.10/13
2s (Kf7)} 35. Bd6 {+3.78/11 2s} Qa8 {-0.09/13 1s (Kc8)}
36. Qd4 {+4.90/11 1s} Qg8 {-0.25/13 2s (Kc6)} 37. Qa4+
{+7.54/10 1s} Ke6 {-0.19/14 1s} 38. Qa1 {+6.49/10 2s} Qf7
{-0.19/12 2s (Kd7)} 39. Qa6 {+6.37/11 2s} Nb4 {-0.10/12 2s}
40. Qc4+ {+6.29/11 2s} Nd5 {-0.19/13 1s} 41. Bg3 {+5.23/11
2s} Qf6 {-0.12/12 2s (Qa7)} 42. c6 {+6.29/11 2s} Qc3
{-1.66/12 1s} 43. Qa6 {+8.56/12 2s} Kf7 {-1.99/12 1s (f4)}
44. Qb7+ {+11.23/10 1s} Kf6 {-1.99/13 2s (Ne7)} 45. Qc8
{+10.58/10 1s} Qd3 {-3.00/11 1s (Qc5)} 46. Qf8+ {+10.78/9
2s} Ke6 {-3.00/13 2s} 47. Qd6+ {+11.23/9 1s} Kf7 {-1.43/1
0s} 48. Bh4 {+13.45/10 1s} g5 {-4.01/12 3s} 49. Bxg5
{+17.07/10 2s} Qc4 {-10.79/13 2s (h4)} 50. Qd7+ {+21.09/12
2s} Kg6 {-11.16/15 2s} 51. Qe6+ {+22.80/12 2s} Kg7
{-11.87/15 1s} 52. Bh6+ {+23.00/12 2s} Kh7 {-11.87/15 1s}
53. Bf8 {+24.92/12 1s} Qd4 {-8.02/12 1s} 54. Qxf5+
{+26.19/11 1s} Kg8 {-11.87/14 2s} 55. Bd6 {+25.52/11 2s} e3
{-11.87/13 2s} 56. Qe6+ {+25.35/10 2s} Kh7 {-12.07/14 1s}
57. Qf5+ {+26.31/11 2s} Kg8 {0.00/20 1s} 58. fxe3
{+28.66/12 2s} Nxe3 {-14.11/13 2s} 59. Qf8+ {+40.17/11 1s}
Kh7 {-3.39/1 0s} 60. Qf7+ {+40.74/13 2s} Kh8 {-14.80/13 3s
(Qg7)} 61. Qxh5+ {+89.17/13 1s} Kg8 {-15.00/14 2s} 62. Qg6+
{+89.17/14 2s} Qg7 {-15.00/14 2s} 63. Qe8+ {+93.25/12 0s}
Kh7 {-15.00/13 2s} 64. Qxe3 {+93.20/10 0s} Qf6 {-M28/9 0s}
65. Qe7+ {+93.40/12 1s} Qxe7 {-M27/8 0s} 66. Bxe7
{+93.40/11 0s} Kg6 {-M26/7 0s} 67. c7 {+M10/12 1s} Kf5
{-M25/6 0s (Kf7)} 68. c8=Q+ {+M8/8 0s} Ke4 {-M24/5 0s}
69. Qc4+ {+M6/6 0s} Ke5 {-M22/5 0s (Kf5)} 70. Kg3 {+M4/6
0s} Kf5 {-21.05/1 0s} 71. Qg4+ {+M3/6 0s} Ke5 {-M2/1 0s}
72. Kf3 {+M2/6 0s} Kd5 {-M1/1 0s} 73. Qe4# {+M1/6 0s} 1-0

[d]8/4B3/8/3k4/4Q3/5K1P/6P1/8 b - -

Eelco

[Ovyron]

So, what are your plans on releasing Ancalagon for the future? I'm certainly interested in testing it.

[Tony Thomas]

Elco, isnt there a possible problem with your time control? I know that you are expecting both engines to use the opening book, but what if engine one plays an opening like H2-H4, do you think that the opponent will be able to make a reply in 0 seconds? Or is there a way to circumvent that problem?

[Eelco de Groot]

So, what are your plans on releasing Ancalagon for the future? I'm certainly interested in testing it.

Hi Ulysses,

Thanks for the interest, I am looking at some of these games and I fear there is still a very long way to go. I said in an earlier post that the goal was to get within 200 points of Rybka 3 but 200 points closing distance away from Rybka 2.3.2a is already out of reach at the moment. I wrote to somebody else that maybe in autumn Ancalagon would be playing a bit better, I hope to have something ready by then, it would be nice if some improvements could benefit Glaurung also when Tord has more time after a busy period in summer.

Best, Eelco