Computer Chess Club Archives


Search

Terms

Messages

Subject: CCT4: Celes notes of the second weekend

Author: Johan Hutting

Date: 08:06:02 01/28/02


Round 7: Celes-Pyotr, 1-0
While I had fixed Celes hash implementation for this weekend the author of Pyotr
told me he had a problem with his... He said he had disabled all hashing and was
running with a max_depth of 4, ouch. After giving away a pawn on e5 Celes simply
outsearched Pyotr and won more material after which the game was more or less
over.

Round 8: Armageddon-Celes, 1-0
Armageddon was one of the engines Celes had at least a small chance against,
however after
[D]rn1q1rk1/p4pbp/1p2p1p1/2pp1b2/P1P1PPnN/3P2P1/1P4BP/RNBQ1RK1 b - e3 0 11
I was afraid celes lacked the searchdepth and had given away the position. After
dxe4, dxe4 Celes showed a pv of Bg7-d4 Kg1-h1 Ng4-f2 Rf1xf2 Bd4xf2 Qd1xd8 Rf8xd8
e4xf5 e6xf5 after which the rook on a8 is en prise, but instead of exf5 white is
forced to play Nc3 to prevent Rd1+.
After 12.dxe4 Bd4+ 13.Kh1 Nf2+ 14.Rxf2 Bxf2 15.Nc3 Bh3 16.Bxh3 Qxd1+ 17.Nxd1
Celes had a rook for two white pieces, once the position opened up it lost quite
fast.

Round 9: Amateur-Celes, 1-0
Celes managed to get into a position which usually quickly loses vs human
players, but we all know how computers handle these games :) Celes decided to go
straight for the white king in
[D]rn1q1rk1/1bpnbppp/1p2p3/3pP3/1P1P1B2/p1P2N2/P1QNBPPP/R4RK1 b - b3 0 13
13...g5 14.Bg3, h5 and after 15.h3 the pawns stayed at their position, altough I
frequently saw g4 and f5 fly by in the pv. In this position Celes remained
planless until Amateur started to put it's pieces together and attack the
kingside. After 25.Qe3 the score started to drop rapidly and Celes eventually
was mated.

Round 10: Celes-Nullmover, 0-1
Celes played ok up to move 19,
[D]3rr1k1/pp4pp/2n5/5p2/2RPn3/1P2BB2/5PPP/R5K1 w - f6 0 19
Here it played b4, locking in the rook, the point being that Nd6 fails on Bxc6.
[D]6k1/1p6/7p/1p1P2p1/2nB1p1P/3K4/5PP1/8 b - h3 0 33
Black took on h4 here, I think white is ok after that, as long as it plays f3,
Bc3 to stop the black pawns, then go pawngrab on f4. Unfortunately, Celes
decided to play Ke4 instead of Bc3 and quickly lost the bishop for a pawn.
I had already accepted Celes' loss and started observing some of the other games
until this position showed up:
[D]3k4/3PnK2/3q3p/8/7p/8/8/8 w - - 0 55
55.Kg7 Nf5+ 56.Kf7 Ne7 57.Kg7 Nf5+ 58.Kf7
> NullMover has disconnected
> NullMover reconnected
58...Kxd7 59.Kg8 Qg3+ 60.Kf7 Qg7# White checkmated 0-1
I'm sure you can figure out my opinion on this way of 'fixing' bugs...

Round 11: Celes-Pepito, 0-1
*sigh* in the last round I had to play Pepito, I'd rather have had Polarchess or
WarLord as my opponent but unfortunately I missed them both during the
tournament.
Celes went for a stonewall that was quickly blown open and made a tactical error
with Qd1-b3xb7 after which it lost one pawn and later on had to give up another
one. Pepito finished the game with some strong play.

Anyway, I'd like to thank Volker and Folgers, along with the programmers to make
a tournament such as this possible. I feel sorry for Thomas who seems to have
burned down the computer Quark ran on, although I'm also glad he managed to make
Quark play such a strong tournament (although it had luck vs monsoon IIRC).

Regards,
Johan



This page took 0.05 seconds to execute

Last modified: Thu, 07 Jul 11 08:48:38 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.