Winning starts with what you know
The new version 18 offers completely new possibilities for chess training and analysis: playing style analysis, search for strategic themes, access to 6 billion Lichess games, player preparation by matching Lichess games, download Chess.com games with built-in API, built-in cloud engine and much more.
Hard-boiled insiders will not completely agree with the headline. "Well fished" is political more correct, since a new chess program named Rybka is causing great excitement among chess engine fans. Rybka means little fish in Czech (as well as in Polish). But we must not think ornamental fish – it's more like a piranha.
If you are participating in the $16,000 Freestyle Chess Tournament next weekend (March 18/19), and if you are not an expert on computer chess, you need to decide which engine you will employ, whether one or many, and how you can use them in the game.
Rybka, Fritz – okay, and we shouldn't forget Shredder. Last year a majority of the engine players in the PAL/CSS Freestyle Tournament used Shredder 9. Its author, Stefan Meyer-Kahlen, is diligently working on the new version, No 10, right now. If it were already out, people wouldn't play so many incestuously Rybka-Rybka matches in the engines room on playchess.com (and you can expect something similar in the Freestyle Tournament).
But if you ask me, honestly, I would say that Fritz 9, Hiarcs 10, Fruit 2.2, Spike 1.1, Junior 9 and a dozen other programs are equally useful when it comes to whispering cunning moves in your ear. It is completely up to you, whether you let your engine play automatically (without your "help") or if play centaur style, like I do, trying to create a plan and checking the moves with your chess engine(s). "Right or wrong - it was my plan", you can say after your loss, and the engine player will answer to you: "nice plan, but the dollars are mine".
And one further free hint: Rybka is sometimes very generous about sacrificing its pawns. Often the program is often right with its evaluation. And I tell you no secret when I say that this program also still suffers from a serious lack of endgame knowledge in some types of endgame (not all). Work is in progress. Another point is its weakness in view of clever kingside attacks. In this respect, no matter of attacking or defending, I clearly prefer Fritz 9.
If this didn't already put you in the right mood, then think of the possibility that Gary might be among the kibitzers. Last year there were rumours that he might be part of the mysterious winning team, "ZackS", which was so easily able to defeat well-known grandmasters in the play-offs. Nobody could imagine that there was no strong grandmaster backing ZackS, but that two computer freaks and chess amateurs, with a rating below 2000, were all that stood behind this name. A real sensation, which was reported in Computerschach & Spiele and on the ChessBase.com news pages.
The ZackS story should encourage everyone who is afraid that they have no chances against the strong title holders in this tournament. Although the general trend indeed showed a superiority of the combination GM plus chess engine, especially in comparison to the stand-alone engines and weaker centaur teams, there is no rule without exception. But the fact that the GMs Lanka, Kosyrev ("tank1") and Dobrov were among the last four doesn't mean they had a picnic in the rounds before (as is the case in the early rounds of a open human tournament). Freestyle means also hard work for the title holders, from the first round on.
Gary Kasparov, the spiritual father of Advanced Chess, which he proposed in 1998, saw his views confirmed by the 1st Freestyle Tournament. He wrote about this in his New In Chess column (NiC Magazine 5/2005, p. 96f). “At first the results [of the PAL/CSS Freestyle tournament] seemed quite predictable. Even the strongest computers were eliminated by IMs and GMs using relatively weak machines to avoid blunders.”
Three semi-finalists were indeed grandmasters, working with computers, but the fourth player, ‘ZackS’, who eventually won the event, turned out to be two American amateurs, both rated under 1700. Kasparov points out that the ZackS team consistently played the Grunfeld and the Najdorf, using the machine to pick the right lines. “This could be the postmodern way to play the opening,” writes Kasparov, “using database statistics and the machine’s ‘instincts’. And they beat GMs with this technique.
Just to give you clue, I would say that the strength of good Advanced Chess players must be something around Elo 3000. Frederic Friedel was right in his conclusions last year: "The level of play may be the highest ever seen at these time settings. There cannot be a doubt that a human player, even one of the top players in the world, would have no serious chance in such a field."
New ideas sometimes have a hard time, before they become generally accepted. Only visionaries and the far-sighted will spot the potential. “Pro-am events are an excellent way to popularise a sport,” writes Kasparov. “Hand-wringing about random chess, computers, and other novelties is foolish. No one is trying to replace or destroy classical chess. We must welcome innovations that can attract a new audience and expand the horizons of the game.”
In order to help Freestyle Chess along we have made some modifications. Last year's event had too many days of play, so we have introduced a simplified modus, in co-operation with the main sponsor, the PAL Group in Abu Dhabi, which among chess players is already well-known because of its chess 'monster' Hydra.
Furthermore the event is part of a Freestyle Grand Prix, with three tournaments in 2006. The next tournaments are planned for June and September. In December their could be a further highlight, when the three winners of the year come together in order to play out the Super Freestyle Championship. Details about this exciting final will be supplied as they become available.
Back to next weekend, the 18th and 19th of March. Until now teams of 28 countries have registered, and we expect that there will be a total of 150–200 participants (three times more than in last year's main tournament). The latest news (and a real highlight) is that British GM Tony Kosten will work together with Mark Uniacke and the Hiarcs program as a team.
In order to hold one's own in the Freestyle Tournament you will indeed need a highly competitive hardware. With outdated hardware you will find it difficult to get even half a point, even if you have a FIDE rating of more than 2400. The optimum is a team of two or three PCs. This is how you can use them.
On your internet PC you should have your engine running. This is only possible in “Centaur” mode, which can be set in the menu "Edit – Playing mode" (Ctrl-M).
"Centaur" is the Playchess term for humans and machines working together. For players who are using only computers, without human intervention. it is appropriate to set "Computer". In this mode the machine can in fact play the entire game automatically. Note that you can only set Computer and Centaur in rooms where machine play or assistance is allowed.
On your other PC(s) you should be analysing possible continuations, going quickly to critical positions and checking their feasibility. Today’s chess engines are very strong, but if the search is guided by humans they become even stronger.
On at least one of your PCs (if not on all) you should have installed the five-piece tablebases (some users even have with selected six-piece tablebases installed on their hard drives) If you risk playing without tablebases that will cost you a fair number of points. Last year when I was in time trouble against a strong US grandmaster I got into an endgame with two knights (him) vs. pawn (me). I was shocked when my tablebases announced a mate against me in 79. Fortunately my opponent was playing without tablebases, and his knights wandered helplessly around the board for 50 moves. The game was drawn, and I greatly enjoyed this half point.
On principle a team of two players will have an advantage compared to a single player, provided that the two players work well together. That was one probably one main aspect of the ZackS success. If you start a long debate with your partner on simple questions then you are not suited to work as a team. A lot of practice is required, as was demonstrated bny ZackS. The two amateurs played countless games on the server and were well tuned to cooperate in their analysis.
This year the time controls are a bit shorter than last year: 45 minutes with 5 seconds bonus per move. The reason was to make it possible to have eight rounds played on one weekend, without too much stress. Planning the schedule we had to take into account that there may be games that last more than 100 moves, due to the extraordinary precision of computer assisted play. Such games often produce interesting endings, and it would be a shame to prevent them by avoiding a time increment per move.
The right psychological and practical attitude to the time controls is very important. I suspect that 45m+5s (compared to last year’s 60m+15s) will slightly improve the chance of pure engine players, as they usually come out of the opening with a time advantage, due to their big opening books. On the other hand automatic engines often waste time for moves that are completely self-evident (e.g. recapturing a piece). That can be of advantage for the centaur. Also, if you play a move which was expected by the engine, it will often reply immediately. So if you have the choice between two moves of nearly the same value, you should prefer the less probable move in order to force the engine to start a new calculation.
There are many less important points which must stay unmentioned by now. The best would be to gain some experience in the engine room of playchess.com, before the Freestyle Tournament gets started.
About the author: Arno Nickel, 54, is a correspondence chess grandmaster, ICCF rated 2604, and a chess publisher in Berlin. Nickel has been working in computer chess, especially in the context of correspondence chess, for about 15 years now. He played some exciting correspondence matches against computers, amongst others defeating an earlier version of Hydra in 2004/2005.
The
tournament begins on Saturday, March 18, 2006 at 14:00h CEST, the time in Berlin,
Paris, Rome and Madrid. That translates to 13:00h London, 8 a.m. New York,
16:00h Moscow. If you are uncertain about the time in your location you can
check this by clicking on the times of each round on the ChessBase.com
announcement.
The event will be staged in a special tournament room (“Hydra Chess – CSS Freestyle Tournament”) of the Playchess server. Note that you should set your playing mode to “Centaur” when you are in the Freestyle room (this is done in the menu "Edit – Playing mode" or by pressing Ctrl-M). If you leave “Human” mode switched on you will not be able to start an analysis engine. “Computer” mode is for people who want their engines to play automatically, without human intervention.
Due to the large number of participants it is necessary to enforce some additional rules for this event. In summary: all participants are responsible for understanding and being able to adequately operate the Playchess client software; they are responsible for establishing an adequate and stable Internet connection to the server; for finding the tournament room; for their punctual appearance on the server in time for the start of each round; for not attempting to use unfair methods to gain advantage over the opponent.
In particular we will be implementing the following rules.
1. All participants must be present in the tournament room of the server 30 minutes before the start of round one (Saturday, March 18, 2006, 14:00h CET, i.e. they must appear in the room at 13:30h CET). Participants must stay connected until the tournament director has included them in the list of players ("invited" them). After a player has seen that he is on the participants' list he can leave the room or log out, but must be present in the room at least five minutes before the start of the game.
2. Each round starts on the hour. Players are required to be in the tournament room and ready to start their games at least five minutes before the hour (i.e. at 13:55h, 15:55h, 17:55h, etc.). The games will start punctually, and any player who is not present in the tournament room at the time of the start will automatically lose that game. It is not possible to start individual games at a later time and include them in the tournament results.
|
3. VERY IMPORTANT: if at any stage of a game your opponent has technical problems, e.g. communication lag or disconnects, you MAY NOT claim a win by clicking on the “Claim win on disconnect" button on the top right of the board window. You can consult the tournament director (by chat), but may have to remain connected and at the board until your opponent's time runs out. Note that if anyone claims a win by using the "Claim win on disconnect" button, even if that player is awarded the win by the server, the tournament director will change the result to a loss for player who has manually claimed the win.
Please remember that due to the very large number of participants we have to rely on your cooperation to successfully stage this event. If you feel you have been unfairly treated you can write to the organisers, who may offer you a free place in the next Freestyle tournament to compensate for the injustice. As a rule it will be difficult to find satisfactory solutions while the event is under way. Remember: the decision of the tournament director is final.
One more request: please play the entire tournament, even if after some rounds you feel that you do not have a chance to win the main money prizes.
The large number of participants makes it necessary to keep the exact tiebreak mode open and flexible. Depending on the number of players who have equal scores within the qualification slot the tournament director will choose a tiebreak modus that is practical and fair. In any event the players will be required to play tiebreak games, but the date, time controls and exact rules will be announced by the tournament director after the number of players that need to play in the tiebreak is known.
At the end of the main tournament there will be an extensive report on the official web site. We would like you to contribute to this report by sending us some information on yourself and the course of the tournament as seen by you. Please give us an exact description of the hardware and software you had, and how you used them in the event.
If you have played an interesting or important game, please consider sending it to us with light annotations. A short text description of your general impressions would also be appreciated. Please do not forget to include your name and place of residence. We will invite all players who send us interesting material to the next PAL/CSS Freestyle tournament.
Address your correspondence to redaktion@computerschach.de.
With best wishes
Dieter Steinwender
All times given in the following schedule are in Central European Time (CET), which is used in most European countries, e.g. Germany, France, Spain, Italy, Scandinavia, etc. CET is GMT + one hour. For instance 14:00h (or 2 p.m.) CET translates to 13:00h London, 8 a.m. New York and 4 p.m. Moscow. Click on the times given in the schedule below to convert to your local time.
| Main Tournament (Sat.-Sun., March 18-19, 2006) | ||
|
Round 1 |
Saturday, 18.03.2006 |
|
| Round 2 |
Saturday, 18.03.2006 |
|
| Round 3 |
Saturday, 18.03.2006 |
|
| Round 4 |
Saturday, 18.03.2006 |
|
| Round 5 |
Sunday, 19.03.2006 |
|
| Round 6 |
Sunday, 19.03.2006 |
|
| Round 7 |
Sunday, 19.03.2006 |
|
| Round 8 |
Sunday, 19.03.2006 |
|
