International Computer Chess Association ( ICCA ) tournament

The ICCA considers itself the sanctioned body to determine the annual computer chess champion similar to FIDE. However, the ICCAs prestige has dropped considerably over the last several years. The key problem with the tournament is only a handful of chess programs participate. The main reason is the tournament is normally held in Europe and programs must be operated at the location, so many potential participants cannot get to the playing site in person. In addition, the cost and logistics of getting to the playing site. Furthermore, the ICCA has the bad habit of changing the rules at the last moment causing further problems for participants.

There are literally hundreds of computer chess engines available and yet only a small percentage participates in the tournament. The cure for this problem is quite simple, have the tournament on the Internet, preferably at the Free Internet Chess Server ( FICS ), this way many programs can participate. The ICCA should seriously consider this or face the fact that it is becoming a non-entity.

This year’s tournament had new rules about titled winners, which are rather confusing. The three titles are 1) Single-CPU 2) Multi-CPU and 3) Amateur. Amir Ban, Israel, the author of Deep Junior responded to the following questions;

Q1. "What did you think about the multi category championship?"

A1. "I said it very plainly in the players meeting before the tournament start: the multi category makes no sense, and I don't consider myself competing in it. I used a dual not to be in the category, but to give me a better chance to win the tournament.

There was discussing of which playoff to hold in case of ties, and the following possibility came up:

The tournament ends with three programs tied, ranked by tie-breaking point; (1) multi (2) single (3) multi. What playoff to hold? It was decided (1) will be declared tournament winner, (2) will win the single title, and (1) & (3) will playoff for the multi title, with the ridiculous possibility that (3) will win and (1) will remain as "only" the tournament winner.

I said if this came up, and I was (1), I would have no interest in winning the playoff game, and may just forfeit it.

Regarding the single event, it's questionable too, because several of their strongest single opponents were missing, having chosen to play dual."

Q2. "Do you feel your title has been tarnished to any degree that Shredder is the single CPU title holder?"

A2. "No. It would have been if ( single-CPU ) Junior7 had played in this tournament, but it didn't."

Q3. "How would you like to see the title organized?"

A3. "In such an event the only titles that make sense are overall champion and the amateur title. The multi and single titles are ICCA nonsense. The single title should be awarded in a single-only event. The multi title makes no sense at all, because any multi can be a single if he chooses to."

This year's tournament had eighteen participants playing nine games with the final standings as follows:

 1 Deep Junior                     8.0 44.5 Dual
 2 Quest (Deep Fritz)              6.0 47.0 Dual
 3 Shredder                        6.0 41.5 Single
 4 Gromit 3.9.5                    5.5 39.0 Single
 5 Chess Tiger 14.6 Gambit Tiger   5.0 50.5 Single
 6 Rebel                           5.0 47.5 Single
 7 Crafty 18.10X                   5.0 47.0 Dual
 8 Goliath                         5.0 44.5 Single
 9 Ferret                          5.0 40.0 Dual
10 Gandalf 5.0                     5.0 38.5 Single
11 ParSOS                          5.0 34.0 Dual
12 Diep                            4.0 38.0 Dual
13 Tao                             4.0 36.0 Single
14 IsiChess X                      3.5 36.5 Single
15 Ruy Lopez                       3.0 37.0 Single
16 Pharaon                         3.0 35.5 Single
17 SpiderGirl                      2.5 34.5 Single
18 XiniX                           0.5 37.0 Single

The separate blitz tournament title winner is the freeware program Goliath.

Many of you reading this article may be wondering why Deep Fritz 7 ( DF7 ) did not participate. In fact, DF7 is playing in Maastricht, but under the secret 'nom de guerre' Quest, however, not so secret. Worried that the prestige of the GM Kramnik match will be tainted if DF7 does anything other than win, which it did not do. Rumours have it that Quest is a DF7 version tweaked to play against computer chess programs.

Certainly the most talked about game of the tournament is between Ferret, authored by Bruce Moreland, USA and Gandalf 5, authored by Steen Suurballe, Denmark. This game is a tactical nightmare and quite interesting to play through.

Further information on the match and to download the games can be found at the following URL: http://www.cs.unimaas.nl/olympiad/

(1) Ferret - Gandalf
WMCCC - Round 2 Maastrict, 19.08.2001
[Annotator Sean Evans]

C92: Closed Ruy Lopez: 9...Be6, Karpov and Flohr-Zaitsev Variations Certaintly the most talked about game of the tournament is between Ferret, authored by Bruce Moreland, USA and Gandalf 5, authored by Steen Suurballe, Denmark. This game is a tactical nightmare and quite interesting to play through.

Game in PGN

1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 Nf6 5.0-0 Be7 6.Re1 b5 7.Bb3 d6 8.c3 0-0 9.h3 Bb7 10.d4 Re8 11.Nbd2 Bf8 12.a4 h6 13.Bc2 exd4 14.cxd4 Nb4 15.Bb1 g6+/= Diagram

The views expressed here do not necessarily reflect those of TWIC, Chess & Bridge Ltd or the London Chess Center.