Houdini, by Robert Houdart, is playing with a brand new version in Season 9 of the Top Chess Engine Championship. After 32 rounds Houdini 5 dev is one of the hot favorites for qualifying for the Superfinal.
Robert Houdart kindly agreed to an interview regarding the performance of Houdini and the situation in TCEC Stage 3
Chessdom: Dear Robert, now after 32 rounds Houdini is second in the TCEC Stage 3 standings. It is a point ahead of Komodo and two points behind Stockfish. How do you evaluate the chances for Houdini to get into the Superfinal?
Robert Houdart: First of all, thank you for organizing the TCEC tournament, probably the strongest chess event ever! It’s amazing to see the involvement of so many engine developers producing new releases at every stage; the superb web interface in which the games are played and the live chat that makes every game a vibrant experience.
To answer the question, Stockfish has had a very good run and its lead appears to be unsurmountable. For second place it will be a coin flip between Houdini and Komodo. With so few games remaining, anything can happen!
Chessdom: Is the performance of Houdini 5 dev matching your expectations? Have you found points where it still needs improvement?
Robert Houdart: The performance is about as expected – competitive with the top of the rating lists. In the games I’ve seen Houdini has been playing very well in the opening and early middle game, but then sometimes failed to convert its advantage in the late middle game. We’ll definitely try to improve that aspect.
Chessdom: Is there a chance to see Houdini 5.0 in the final?
Robert Houdart: If Houdini qualifies for the finals I will provide an updated engine –the current development version is already 15 Elo stronger than the version that is playing in Stage 3. We will continue to develop Houdini up to the final release of Houdini 5 in October, qualifying for the TCEC final would be a nice (albeit rather distracting :) ) bonus for an engine that is still very much in development.
Chessdom: Stockfish, Houdini, and Komodo are the hot favorites for the Superfinal. Has any of the other opponents impressed you at this stage?
Robert Houdart: A notable performance comes from Fire, it’s been playing simple, straightforward chess and been impressively solid. On the other side of the scale is Jonny, it played great in the previous stages but appears to have lost some of its mojo in Stage 3.
Chessdom: We cannot avoid this question, TCEC Stage 3 started with an incident that provoked a replay of a game which Houdini won. Do you find the decision of the organizers fair?
Robert Houdart: TCEC took the right decision to replay the infamous “Game 0”. Engine authors expect that the engines play with the settings they request, any deviation from that rule would jeopardize the tournament’s ability to act as a catalyst for engine development.
On the other hand it should be noted that the adjustments applied to Komodo were really tiny. So even if the game result doesn’t count in the standings, the game itself remains a valid data point in estimating the performance of the engines.
Chessdom: Stockfish looks unstoppable in the last rounds. It is the only engine that defeated Houdini. You stated earlier that you “would not have Houdini 5 dev competing, if you do not believe there is a chance for first”. Has the situation changed in the past 32 rounds?
Robert Houdart: From the games in Stage 3 you can see that the three top engines are close in strength – playing mostly draws between them and regularly beating the other engines. Any of the engines that qualify for the final will have a genuine shot at winning the tournament.