So it was CTD028 playing in this tournament, on 4 CPU cores (AMD Opteron 270).
The first round CTD played Eowyn, a new engine by Walter Ravenek. He told me he started the engine in October last year, in Java, and did not expect to win much as he knew how much functions, search tricks and evaluation knowledge was still missing. In the game the QGA quickly went to an endgame where the advanced pawn of Eowyn turned out to be weak which ultimately cost the game.
The second round CTD played a engine that is a bit more known. CTD played a few crappy moves as it did not really seem to understand the position where it left the book. Nevertheless, it fought back and although it was a pawn down, because of the uneven colored bishops ending that appeared on the board there was a glimmer of hope of drawing Rybka again. In vain ...
The last round of the first day CTD played ZZZZZZ. ZZZZZZ was not in an 'active play' mood which meant that CTD succeeded in dictating the course of the game. In the final position the material was still even, but the mighty pair of bishops and the cramped position of ZZZZZZ's pieces meant that serious material loss was near.
The first round of the next day was against Diep. Vincent had some trouble with his book conversion function as it was thrown out of book after only a few moves, and got into a very bad position, even though it was a pawn up. CTD probably missed the win by not playing 19.Rg1 and allowing Diep to castle out of the worst trouble. The position remained nice for CTD, but in the end it could not avoid a trade down to a drawish endgame. It could choose: KRBKR or KBPK with the wrong rook pawn. It chose neither, shuffling pieces, trying to find a third option. That third option appeared to be a 50 moves draw with the material it had on the board at that time, so a draw was agreed.
The fifth round had a similar scenario in a sense that the mighty pair of bishops of CTD ruled the board, won a pawn, and then CTD allowed Hermann a trade down to a rook ending with an extra passer, and it did not manage to win that one either.
So I had high hopes for the sixth round. Two games with a very good if not won position that went to a draw, so the third game should give me a good/won position converted to a win. Problem number one was the opponent though: Deep Sjeng. The bookline was not what Gian-Carlo had in mind when he selected 1..e6 after 1.d4: It turned French with 2.e4. CTD engaged into a dynamic attacking line that looked good, but soon after leaving the book CTD probably made its first mistake with 18.g3 instead of the more natural 18.Ne2. 18.g3 lead to 19.g4 and the board ignited. Sjeng calculated the long force sequences much better due its use of singular extensions and CTD started to miss the best continuations, allowing Sjeng to take over the attack. Sjeng did not make a mistake there and converted the game to a win.
The next day started with a short game against Tzjezz. It had an original way of throwing CTD out of book, but I doubt this novelty should be repeated as it resulted in a mate being delivered about 20 moves later. But Tzjezz is improving. It will join the programmer's tournament in Leiden at the end of the month and it might start getting some points there.
The eight round was against Hiarcs. CTD/theBaron has a track record against Hiarcs that it wanted to maintain, but that proved to be a bit more difficult that I hoped. CTD chose a Petrov and since uneven colored bishops appeared on the board Harvey and I started to suspect this would be one more draw between CTD and Hiarcs. But as CTD thought it had the better position, it avoided the queen exchange that would probably have gotten it the draw, later sacced the exchange in an attempt to get to a drawn ending (and it probably was), but missed the dangers in the pawn ending that appeared a little later. With the safer 29..Bd4 the game may have been saved.
After 8 rounds, there were only 3 programs that CTD hadn't played yet: Joker, Hansdamf and the King. Either one of them could have been the last opponent and it turned out to be the King. CTD/theBaron never beaten the King and drew it only once (at least I remember only one time). It also turned out to be a crucial game in the tie-breaker between Diep and Hiarcs for third place. I was told that if the King would have won, Diep would be third, and if I won, Hiarcs would be third. As Diep beat Tzjezz in 7(!) moves, Vincent had a special interest in this game and sat close to it during most of it. But much to his regret, CTD got an edge just after leaving the opening book, won the exchange for a pawn and while the King did see some compensation for the exchange somewhere, neither of us could find it. The resulting rook versus bishop ending proved to be a win for CTD.
This meant that CTD became 5th in this tournament, 1.5 points behind number 3/4 and a full point ahead of number 6 (Hermann). As my goal was to get to the top 6, that goal was clearly reached. I look back at a succesfull and entertaining tournament and I hope to get a bit closer to the top engines in the next tournaments.
I will extend this report with some more games analysis and put it on my site later this week.
Richard.