Computer Chess Club Archives




Subject: CCT4: Celes' games

Author: Johan Hutting

Date: 12:13:06 01/21/02

Since everyone seems to do reports, here's mine :)
(I hope the diagrams show up ok)

Round 1: Celes - Bringer, 0-1
Although I knew I wouldn't win this one all went fine up to move 9
[D]rnb1k2r/pp1pbppp/4p3/q1pPPn2/8/2P2P2/PP4PP/RNBQKBNR w
Celes could have gotten a nice position here by playing 9.d6, instead it played
9.Nd2, probably missing the threat Qa5-a4-h4 and Ng3 due to lack of searchdepth.
After it did spot the danger my search freaked out and it sacrificed material
though there was no need to... I had seen enough after the 14th move and decided
to use the 0.40 engine for the next game.

Round 2: Chezzz - Celes (0.40), 1-0
Chezzz took Celes out of book with 1.a3, after a while the position seemed worse
for Celes but Chezzz seemed planless in the closed position. Celes decided to
help a hand by opening the position and quickly lost material after doing so :(

Round 3: Celes - Averno, 0-1
I loaded an older version of 0.50 since I wasn't too happy with 0.40's
performance in the game vs chezzz. Even though I tested this build 2 weeks ago
vs blikskottel in a 60 10 match that it won with +7 =2 -1 apparently this
version has the same bug that killed celes in the first game:
[D]r3k2r/pp2qppp/2p2n2/5b2/3Pp3/2N1P3/PPQN1PPP/R3K2R w
here Celes played 13.f3?? instead of 13.0-0 which virtually made the game

Round 4: Djenghis - Celes, 0-1
Djenghis seems to have some bigger problems then Celes :)
it sacrificed 2 pawns and Celes just traded pieces to get into a winning
[D]r4rk1/pp5p/3p1p2/2b1n3/4R1b1/P4N2/1B1K2PP/5B1R b
was the interesting position from this game, Celes assumed it won a piece by
Nxf3+ but after 25.gxf3, Bxf3 26. Bc4+, Kh8 27. Re6 the position looked far from
simple. Celes played d5 here, but I think Bxa3 followed by d5 is better (more
pawns, GOOD :)
Just when I thought Celes would just finish the ending easily I ran into the
infamous searchbug again:
[D]8/2r4p/1R6/4k3/7P/1pp5/8/1K6 b
I figured Celes spotted a winning position after 45...Re7, but it just raped the
position and probably is a plusdraw. 45...c2+ clearly wins and b2 is probably a
win as well (rip the h-pawn and continue walking with my h-pawn). In the final
position White has Kd1, Rh2 and white can't take on h4 due to Rb4+ *sigh*.

Round 5: Tinker - Celes, 1-0
No bugs in this game, just lack of searchdepth costing Celes a rook and a pawn
vs a rook... I played on for a while since the position was closed, but once the
rooks entered trough the c-line the game was over.

Round 6: Celes - Qalat, 0-1
I hoped to steal at least half a point from Qalat since it only searches about 1
ply deeper. Celes nicely followed the plan of putting a piece on d4 and covering
it with pieces plus an attempt at doubling on the c-line. After move 32 I
already kibitzed expecting a draw once the queens would go off which happened at
move 35... unfortunately the 'bug from hell' decided to show up again:
[D]8/4kppp/8/2b5/3r1P2/4K1PP/4B3/1R6 w
On the previous moves Celes predicted this reaction and planned Rb5 and Bd3 as
equalizers (Rb5 repetition, Bd3 trading the rooks)... until this happened:
26800890 <first :   6     1     67    316119  Be2-d3 Rd4-b4 Ke3-f3 Rb4xb1 Bd3xb1
h7-h5 Bb1-e4
26800890 <first :   7    31     67    316144  h3-h4 h7-h6 Rb1-b5
26818234 <first :   8  -622   1801   9253930  h3-h4 Rd4-b4 Ke3-f3 Rb4xb1 Be2-d3
Rb1-b3 Kf3-e4 Rb3-b2 f4-f5
>ICS: kibitz unless some search bugs show up..\012
26846375 <first : nps:514000 D:8 TTp:27,8 Qp:81,9 Tns:23707666 t:46,156 mo:48,8
26846375 <first : move h3h4

*sigh*, I hope it's "fixed" next week...

This page took 0.06 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.