| CG.com Masters - Machines Invitational (2011) |
|
The 2nd Chessgames.com Masters vs. Machines Invitational was organized by User: chessmoron in 2011, four years after the first edition. The master team was stronger than in 2007, with Sangeeta Hosea making the moves after consulting with Bill Wall and Eric Schiller. Aylerkupp / Rybka, Kutztown46 / Stockfish and Golden Executive / Houdini represented the machines. The engines had no opening books and no tablebase access; the operators (<AylerKupp>, <kutztown46> and <Golden Executive>) were allowed to help the engines on moves 1-8, after which the moves were selected by the engines and relayed by the operators. The schedule was one round per week, starting on April 16; rounds 5 and 6 were both delayed by an additional week to accommodate everyone involved. The games were played at Yahoo Chess, both sides having 105 minutes for the whole game with no increments or added time. Eric Schiller missed the first two games, in which the humans were represented only by Hosea and Wall. All engine vs. engine games were drawn, the games of the masters deciding the final result. The first round already showed the masters were more competitive than had been expected, holding AylerKupp/Rybka to a draw as Black, although Hosea spent two minutes analysing an illegal move at one point. The masters scored a second draw in round 2 and then took the tournament lead in round 3, defeating Golden Executive/Houdini as Black in a Marshall Attack. The decisive game of the tournament came in round 4, the masters facing AylerKupp/Rybka for the second time. AylerKupp/Rybka outplayed the masters positionally, developing a dangerous kingside attack that gained strength after the masters selected the wrong plan on moves 21 and 22. Though the human team defended well after that, Rybka continued to play highly accurate chess and ultimately converted its advantage in a rook endgame; all remaining games were drawn, and AylerKupp/Rybka won the tournament by half a point. The masters' main advantage compared to the engines was in opening preparation; both Hosea and Schiller had very deep preparation in many lines, displayed particularly in the round 3 win against Golden Executive/Houdini and the round 5 draw against kutztown46/Stockfish. 1st AylerKupp / Rybka ** =1 == == 3.5/6
=2nd Hosea / Schiller / Wall =0 ** == 1= 3.0
=2nd kutztown46 / Stockfish == == ** == 3.0
4th Golden Executive / Houdini == 0= == ** 2.5
Previous edition: CG.com Masters vs. Machines Invitational (2007)
|
|
| page 1 of 1; 12 games |
 |
 |
|
|
| Jun-15-11 | | chessmoron: Faster time control this time (105 minutes) rather than 150 minutes. No Opening Books and No tablebases
Thanks to Sangeeta Hosea (aka User: Open Defence) Bill Wall and Eric Schiller Thank you operators:
User: kutztown46 User: AylerKupp User: Golden Executive |
|
| Jun-15-11 | | Open Defence: thanks <chessmoron> for organizing it... the defeat against Rybka hurt a lot, we were out of our depth in that game I really like our play in the last game |
|
Mar-10-12
 | | maxi: Was this played over the web or were you, machines and all, at some physical location? |
|
| Mar-10-12 | | galdur: Interesting. When´s the next one ? |
|
| Mar-10-12 | | brankat: <maxi> This was played on Yahoo. <gaidur> The first one took place 4 years earlier (2007). I happened to be a part of the team then. That's probably why the result was considerably worse :-) :-) CG.com Masters vs. Machines Invitational (2007) So, I assume the next one may have to wait until 2015. |
|
Mar-10-12
| | maxi: The Black side of the Marshall Defense the humans won was very nicely played in last year's tournament. I wonder in particular about the moves 23...h6, 24...Bb7 and 25...Re4. Tremendous. Were these moves figured out over the board?. I already know that 21...Qd7 was a suggestion by Shamkovich. I would guess 23...h6 requires courage, as it creates a weakness. |
|
Mar-10-12
 | | theagenbiteofinwit: Crazy that the decisive games were won by black. |
|
| Mar-10-12 | | brankat: <maxi> <I wonder in particular about the moves 23...h6, 24...Bb7 and 25...Re4. Tremendous. Were these moves figured out over the board?> All the games (both tournaments) were live consultation games. Real time. Three hours for Masters, 2 hours for the engines. |
|
| Jun-12-16 | | Open Defence: In the 2011 tournament, the time controls were the same for the engines and humans (2 Masters + 1 Patzer) You can read the detailed analysis posted on the game pages to see how the humans determined their moves, some move were argued fiercely :) |
|
Jul-10-16
 | | AylerKupp: Hard to believe that 5 years have passed since the last tournament. I hope that <chessmoron> decides to organize a 3rd Masters vs. Machines Invitational and that several masters express an interest in participating. I, for one, would be very interested in participating since I greatly enjoyed the last tournament although in the interest of fairness and opportunity it might also be fun to have new operators. It would also certainly be interesting to see how the masters would do against the 3 currently strongest engines; Komodo 10, Stockfish 7, and Houdini 4. Another possibility would be to match Houdini and Stockfish against Rybka 4.1 to see how far they have advanced against Rybka and the masters particularly when running in newer hardware Five years is a long, long time in the computer business, both software and hardware. I still have the same 4-core, 32-bit computer that I used with Rybka five years ago, so that would make it even more interesting although I suspect that it would then not be any contest in engine vs. engine games. We would have to find another venue to play the games since Yahoo chess is defunct. The final intriguing thought would be a Grandmaster vs. Machines Invitational. I wonder if Caruana, Nakamura, and So would be interested in playing a series of consultation games in a tournament against the latest engines under a similar set of rules if they could find time in their busy schedules. I, for one, would be more than willing to rearrange my schedule to accommodate theirs. I've suggested those 3 grandmasters because they're not only ranked in the top 10 in the world but, since they all reside in the US, the time zone differences for them could be more easily accommodated. Maybe <chessgames.com> could take an interest in trying to make this happen. |
|
Jul-10-16
 | | perfidious: <AK> Not having played tournament chess since 2001, I would be a mite rusty, but am willing to hear more and may have a go. |
|
Jul-10-16
| | AylerKupp: <perfidious> Not to worry. Under the rules of the last tournament very little chess knowledge was needed other than in the opening, since our job as operators consisted of (a) waiting for our opponent's engine to make its move without falling sleep, (b) entering the opponent's move into our chess GUI for our engine to determine it's reply, and (c) waiting for our engine to make its move without falling sleep, and (d) entering our engine's move into the chess site. In the opening most of us, I think, prepared a small opening repertoire to save some engine time. And, if we ever got out of our "book", we just let the engine take over. We also had some training games both against other operators and the masters. These, in addition to providing some experience in using the various items of software, uncovered some deficiencies in, at least my, thinking. For example, there was a lag between the time our opponent's move was presented and the time we woke up from our slumber to detect our opponent's move, some time to enter our opponent's move into our engine, a lag between the time our engine made its move and the time we woke up from our slumber to detect our engine's move, and some time to enter our engine's move into the chess cite software. Since our engine was not aware of these additional time delays, we couldn't give it the full time allowed. If we had it would have taken it all, and would have lost by time forfeit. The amount o time we reserved for these time delays was a judgment call; too much and we would penalize our engine by giving it less time to calculate, too little and we increased our risk of a time forfeit. This was particularly troublesome for Rybka who tended to use a lot of time early on and get into time trouble. Sometimes we had to detect whether our engine was using too much time on what seemed like an obvious move, in which case we forced our engine to move. I approached that problem by running the engine at PV=2; if there was a large gap in the evaluations between PV=1 and PV=2 I forced Rybka to make its best move, provided that the gap was not decreasing. If the gap was small or practically non-existent I just let Rybka run a little bit longer and hope for the best. And unusual things happened that required quick thinking. Early during one game I got a message that I had lost on time. Huh? WHF? I restarted the game, forced the same moves for both engines, and managed to recover. And there was the infamous last game between Rybka and Stockfish when Rybka, was a pawn up, short of time, and at the 49th move without pawn moves or captures when Stockfish's or its operator could claimed a draw by the 50-move rule. Rybka evaluated its position as advantageous since it was a pawn up and decided to sacrifice a pawn since its evaluation of the position after a pawn sac, while much worse than before, was higher than zero which would have been the 50-move limit result. But of course Rybka didn't know about the various time lags and delays so it thought it had more time than it effectively did. So I had to force Rybka to play bullet chess in an attempt to avoid losing on time but its play, limited to a search depth of 1 or 2 play, greatly deteriorated. Even so, I/Rybka would have lost on time if not for the graciousness of Stockfish's operator, <kutztown46>, who offered us a draw. I rejected it the first time but succumbed to greed on the second offer. And as a result Rybka won the tournament, ½ point ahead of Stockfish. If I had not accepted the draw then Rybka would have lost the game and Stockfish would have won the tournament. So, yes, it was an exciting and entertaining experience. |
|
| Oct-16-16 | | Open Defence: Nice idea <AylerKuppp> |
|
Jun-18-22
 | | Messiah: Terrible tourney! |
|
|
|
|
NOTE: Create an account today
to post replies and access other powerful features which are available only to registered users.
Becoming a member is free, anonymous, and takes less than 1 minute! If you already have a username,
then simply login login under your username now to join the discussion.
|
Please observe our posting guidelines:
- No obscene, racist, sexist, or profane language.
- No spamming, advertising, duplicate, or gibberish posts.
- No vitriolic or systematic personal attacks against other members.
- Nothing in violation of United States law.
- No cyberstalking or malicious posting of negative or private information (doxing/doxxing) of members.
- No trolling.
- The use of "sock puppet" accounts to circumvent disciplinary action taken by moderators, create a false impression of consensus or support, or stage conversations, is prohibited.
- Do not degrade Chessgames or any of it's staff/volunteers.
Please try to maintain a semblance of civility at all times.
NOTE: Please keep all discussion on-topic.
This forum is for this specific tournament only. To discuss chess or this site in general,
visit the Kibitzer's Café.
|
Messages posted by Chessgames members
do not necessarily represent the views of Chessgames.com, its employees, or sponsors.
All moderator actions taken are ultimately at the sole discretion of the administration. |
Spot an error? Please suggest your correction and help us eliminate database mistakes!
Copyright 2001-2023, Chessgames Services LLC
|
Kibitzing
