Author: Mike S.
Date: 09:40:19 02/21/02
After Chessmaster's / The King's recent successes with custom personality
settings, I wanted to try "desgning" some settings, too:
Enginematch P3/700, 5m3s, ponder off
123456789012
1 Yace 0.99.56 010½10011½01 6.0/12 36.00
2 CM_Brain beta 101½01100½10 6.0/12 36.00
I think this is a good result for CM_Brain, keeping in mind that Yace had
tablebase access which The King 3.12d doesn't use. This may have been decisive
in one or two games.
A quick win by CM_Brain, with Yace's evaluation data:
[Event "P3/700 5m3s"]
[Site "?"]
[Date "2002.??.??"]
[Round "?"]
[White "CM_Brain beta"]
[Black "Yace 0.99.56"]
[Result "1-0"]
[ECO "B08"]
[PlyCount "80"]
[EventDate "2002.??.??"]
1. e4 {0s} 1... d6 {0s} 2. d4 {0s} 2... Nf6 {0s} 3. Nc3 {0s} 3... g6 {0s} 4.
Nf3 {0s} 4... Bg7 {0s} 5. h3 {0s} 5... O-O {0s} 6. Be2 {0s} 6... c6 {0s} 7. a4
{0s} 7... a5 {0s} 8. O-O {0s} 8... Na6 {0s} 9. Re1 {0s} 9... Qc7 {0s} 10. Bc4 {
0s} 10... e5 {-0.26/8 14s} 11. dxe5 {8s} 11... dxe5 {-0.21/9 11s} 12. Be3 {10s}
12... b6 {-0.23/8 9s} 13. Qe2 {9s} 13... Nb4 {-0.29/9 13s} 14. Rad1 {9s} 14...
Bb7 {-0.24/9 9s} 15. Rd2 {9s} 15... c5 {-0.20/8 11s} 16. Red1 {7s} 16... Nc6 {
-0.24/8 18s CM starts the attack now:} 17. Nb5 {11s} 17... Qb8 {-0.42/8 10s}
18. Ng5 {8s} 18... Nd4 {-0.23/8 6s} (18... Nd8 19. Rxd8 $1 $18) 19. Bxf7+ {9s}
19... Kh8 {-0.54/9 8s} (19... Rxf7 20. Qc4 Qf8 21. Nxf7 Qxf7 22. Qxf7+ Kxf7 23.
Nd6+ Ke6 24. Nxb7 $16) 20. Qe1 {12s} 20... h6 {-0.15/8 10s} 21. Bxg6 {17s}
21... hxg5 {-0.05/8 10s} 22. Bxg5 {9s} 22... Bc6 {-0.15/8 9s} 23. f3 {10s}
23... Bxb5 {-0.22/8 18s} 24. Qh4+ {9s} 24... Kg8 {-0.13/9 9s} 25. axb5 {10s}
25... Ra7 {-0.59/9 11s} 26. c3 {6s} 26... Ne6 {-1.44/10 15s} 27. Rd6 {9s} 27...
Nxg5 {-1.45/9 5s} 28. Qxg5 {4s} 28... Re7 {-1.54/9 6s} 29. Bf5 {9s#} 29... Qc7
{-1.65/7 8s} 30. Be6+ {9s} 30... Kh7 {-1.88/8 6s} 31. Qh4+ {24s} 31... Kg6 {
-1.72/8 7s} 32. Rc6 {12s} 32... Qa7 {-3.71/8 12s} 33. Rdd6 {9s} 33... Rxe6 {
-4.44/8 6s} 34. Rxe6 {6s} 34... Qd7 {-5.39/9 21s} 35. Rcd6 {5s} 35... Qf7 {
-5.84/8 4s} 36. f4 $1 {8s} 36... exf4 {-6.77/9 7s} 37. e5 {7s} 37... c4 {
-8.74/9 23s} 38. exf6 {9s} 38... Bxf6 {-11.47/10 20s} 39. Rxb6 {8s} 39... Qxe6
{-12.45/11 10s} 40. Rxe6 {5s} 40... f3 {-12.45/10 3s} 1-0
After that, I adjusted the settings a bit, and matched it against Crafty 18.01
(wcrafty 18.1), which was using a custom book I had compiled for it longer ago
(it seemed ok):
Enginematch P3/700, 5m3s, ponder off
123456789012
1 CM_Brain ½½½1½1½10011 7.5/12
2 Crafty 18.01 ½½½0½0½01100 4.5/12
This would mean CM_Brain *could* be 90 Elo stronger than Crafty under these
conditions. The match includes one loss on time by Crafty though, which is
probably very unusual.
In this game, CM_Brain's rooks successfully invade Crafty's position:
[Event "Enginematch P3/700 MHz"]
[Site "?"]
[Date "2002.??.??"]
[Round "12"]
[White "Crafty 18.01"]
[Black "CM_Brain"]
[Result "0-1"]
[ECO "D19"]
[PlyCount "85"]
[EventDate "2002.??.??"]
1. d4 Nf6 2. c4 c6 3. Nf3 d5 4. Nc3 dxc4 5. a4 Bf5 6. e3 e6 7. Bxc4 Bb4 8. O-O
O-O 9. Qe2 c5 10. Na2 Nc6 11. Nxb4 Nxb4 12. dxc5 Bd3 13. Bxd3 Qxd3 14. Qd2 Qe4
15. Nd4 a5 16. f3 Qe5 17. Nb3 Rfd8 18. Qc3 Qc7 19. Bd2 Rd3 20. Qc4 Nfd5 21.
Bxb4 Nxb4 22. e4 Rad8 23. Rac1 Qf4 24. Nxa5 Rd2 25. e5 Qg5 26. Qg4 Qe3+ 27. Kh1
Nd3 28. Ra1 Nf2+ 29. Rxf2 Qxf2 30. f4 R8d4 31. Qf3 Rxa4 32. Rc1 Qxf3 33. gxf3
Rxa5 34. b4 Raa2 35. h4 Rh2+ 36. Kg1 Rag2+ 37. Kf1 Rc2 38. Rxc2 Rxc2 39. h5 Rb2
40. f5 exf5 41. Ke1 Rxb4 42. Ke2 Rh4 43. Kd3 0-1
I don't know if the original settings, or other custon settings would have done
noticeably better (or worse). - But if Chessmaster fans find the CM_Brain
settings interesting, maybe someone likes to use them for matches against some
pros (by using that inbetween and OPK stuff; I don't touch this :o).
CM_Brain.CMP Base Rating: 2724 CM8000.OBK
A/D (Attack/Defender) [ -8] M/P (Material/Position) [ -3]
S/P (Strength of Play) [ 100] C/C (Control of Center) [ 90]
Rand (Randomness) [ 0] MOB (Mobility) [ 100]
MSD (Max Search Depth) [ 99] K/S (King Safety) [ 130]
SS (Selective Search) [ 10] P/P (Passed Pawns) [ 115]
C/D (Contempt for Draw) [ 0] P/W (Pawn Weakness) [ 120]
OWN OPPONENT
Queen [ 92] Queen [ 90]
Rook [ 50] Rook [ 50]
Bishop [ 32] Bishop [ 31]
Knight [ 30] Knight [ 30]
Pawn [ 10] Pawn [ 10]
Transposition Table [32 MB]
Regards,
Mike Scheidl
http://meineseite.i-one.at/PermanentBrain/
This page took 0.01 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.