The Daily Update: British Championship, Politiken Cup, Tromsø, Computers: UPDATE: US Open, World Correspondence Championship
1. In the British Championship, super-GM Michael Adams resumed his winning ways, quickly defeating GM Stephen Gordon with the Black pieces. Gordon had a small edge out of the opening, but his aggressive plan with 20.f5, 21.Ng4 and 22.f6 was perhaps a bit too aggressive. The resulting position was easier for Adams to play, but White was only slightly worse until he uncorked 29.Rd1?? This allowed a tactic straight out of a beginner's book.
It's always amazing how easy it is to walk into such things when one is worse or playing a stronger player; had Gordon been Black he'd have spotted it in a second himself. As for the solution, I'll trust all of you to figure it out on your own.
Almost all the other top boards drew, but Stuart Conquest beat Danny Gormally to remain a single point behind Adams. They'll play next round, with Adams enjoying White.
2. Politiken Cup: After 4 rounds in the open event, nine players still have perfect scores: Rodshtein, Landa, Hector, Postny, Carlstedt, Cmilyte, Hillarp-Persson, Brunello and Krasenkow. As for the Svidler-P.H. Nielsen match, it's looking grim for the national hero. After what looked like a narrow escape in game 1 with White, Nielsen lost very quickly in game two and again (but more slowly) in game three. 2.5-.5 Svidler, with three more rapid games tomorrow.
3. Tromsø: Another fine open event, the Arctic Circle Challenge, is taking place in Tromsø, Norway. There are lots of GMs (but no Magnus Carlsen), including Loek van Wely, who leads a group of 11 with 3/3. The event goes 10 rounds.
4. Computers: Martin Thoresen's latest engine-engine event is almost finished. This one was a drawfest, and while it's not yet clear who the winner will be, it is clear that it won't be Rybka 4, which drew all its games and can finish no better than tied for second. (I know that the sample size is too small to draw any serious conclusions, but it's still unusual not to see Rybka win an event.) [UPDATE: It's just the first cycle that's almost finished. It's a double round robin.]
UPDATE:
5. The U.S. Open is underway, which will be of interest primarily to U.S. readers. Here's the website.
6. The 22nd World Correspondence Championship is over, and it was won by Aleksandr Surenovich Dronov of Russia. He won on tiebreak over Juergen Buecker of Germany. Both players had 11.5/16, half a point ahead of Joop van Oosterom (a.k.a. Jeroen Piket, some say - English version here).
Reader Comments (18)
Dennis,
Thanks for following the tournament. I have a small correction, however: The format is a double round-robin. So it's only reaching the half-way mark in a few games (15 games). Then the colors are reversed for the second cycle, another 15 games. Please forgive me if I made a mistake making this clear earlier.
For fans of romantic chess, check out Naum's win with the black pieces over Shredder in Round 3. It was a fantastic mess of a game, with 3 queens on the board for a rather long time. Black's b2 pawn must have been the strongest pawn I've ever seen which never queened.
Best Regards,
Martin
LOL!
I wonder what the record is for the longest number of moves in a game with more than two queens on the board?
That's unusual, but as records go it's absolutely nothing. Go to http://www.xs4all.nl/~timkr/records/records.htm and click "Longest 4 queen sequence" and "Longest polygamy".
That is a cool website - thanks!
Does anyone know why they even play this event anymore. With the power of chess engines and a $1,200 for a Quad Core machine I don't see the test of skill that it took to play 10-15 years ago.
Larry, if you're referring to the correspondence world championship, computers have not ended the competition any more than they've put all GMs at the same level in their preparation. And there have been battles between strong correspondence players who aren't so strong over the board and computers, and the humans have won.
If you're not interested in the event, why comment on it?
Because non correspondence players don't understand correspondence. Even many correspondence players don't understand true correspondence. They wrongly believe that people blindly follow computer lines because computers are infallible and have obviously solved chess. Sure that's a bit of a stretch for satire and sarcasm but no where near as much an exaggeration as you might think. I can't tell you how many times I have been ridiculed for playing correspondence chess.
More than ten years ago - but I guess things haven't changed fundamentally in the meantime - I discussed this with my former clubmate Joachim Neumann who was and is a top level correspondence player, currently #5 on the ICCF rating list. As far as I remember he said something like "of course I use engines to double-check, but most if not all ideas and decisions are my own". Most of the positions he showed us during very inspiring club evenings were from complicated endgames where engines are still relatively weak.
So maybe the nature of correspondence chess has changed - more strategic and positional, less tactical. Even longer ago, some 20 years, I had won a correspondence game with a fairly simple combination against a guy who was (then as a student), and still is working for Chessbase - for sure this won't happen again (and not only because I no longer play correspondence chess). However, correspondence chess still has a right to exist.
Just for those following the engine competition: The website has been moved to http://www.tcec-chess.org
Best Regards,
Martin
Dennis,
Actually I always held a high interest in correspondance chess and played myself mostly over the internet back in the 1990's. I enjoyed the fact that the top level games can be a wonderful blend of deep strategies and tactics. Where else could you see Grandmaster level players slug it out in opening like a Two Knights - Wilkes-Barre. As a player I was able to actively explore a new or difficult opening against a strong opponent without the worry I would get the order mixed up or win or lose because of a silly blunder in trying to make time control.
My comment was directed at the fact chess is a way for two people to engage in a contest of mental skills. However, with such strong computers part of that contest has been transferred to a computer program as part of the mix. Checking tactics with a program to my mind defeats the essence of player vs player competition. Yes computers are not infallable but as they have reached the level of near perfect tactical calculation which take away the purpose of playing person vs person chess and testing each one's skills. If the purpose is to see if you and your program can out think your opponent's computer program then that is a totally different contest.
Except for Advanced Chess over the board play is much different. You can not legally check out your moves with the latest version of Rybka when you are actually engaged in a game. Yes over the board play has involve programs more and more but when you sit down at the board you still are on your own.
"My comment was directed at the fact chess is a way for two people to engage in a contest of mental skills. However, with such strong computers part of that contest has been transferred to a computer program as part of the mix."
OK, but that argument also applies to corr. chess the old way - consulting books on openings and endgames had the very same effect.
Also, people have been using computers to assist in their correspondence games - not always legally - for a couple of decades. It made a difference even back then, though then, as now, computer use was insufficient in itself to defeat strong opponents.
Martin, thanks for the alert on the Shredder-Naum game - that was wild!
Dennis, how and/or under which circumstances would computer assistance during correspondence games be illegal, or rather unfair or immoral? Maybe if the opponents had made a deal (exchanged words of honor) that they won't use engines - I heard of such deals or propositions, probably not at the very highest level. Maybe if one player uses superior software or hardware which isn't available to the public including his opponent [was Topalov cheating during his match against Anand?].
But generally spoken, isn't it just another difference between correspondence and OTB chess? Beyond the most obvious and desired one, spending much more time on each move, these include [all DURING the game]
- (as mentioned by MNb) use of opening books, databases and endgame manuals
- getting help from others (maybe this is immoral, but such an unwritten rule cannot be enforced)
- moving the pieces around while analysing
If, hypothetically, all computer help during correspondence chess was declared illegal, just like doping is illegal in other sports, what would be the implications? Could "doping authorities" search the house of a correspondence GM and disqualify him as well as ban him for one or two years if they find a chess engine? If I sound ridiculous it's done on purpose, pushing forward my proposed analogy.
BTW, 23.Bc5!! in Van Oosterom-Nimtz (your Dutch language link) may have been Piket's idea - but it's also a typical engine move.
Using books in correspondence chess is not the same as using a computer engine and a set of table bases. First books contain outright errors (both before computers and now) or even misleading analysis. Second because when literature was based on human efforts only there was lots of room for discovery of new ideas that could take a game out of book and back into the realm of head to head competition rather quickly.
As for computer programs now not being strong enough to beat strong players, I have to disagree. And if the point is that the two players knowing the other is probably using a chess engine have to play the game to minimize their effect, then that is perverting the game as well. My initial question was what is the point of the world correspondence championship if there is no legal or moral restraints for using software to help you play the game? It certainly can not be what it was before, to determine which human was best at deep analysis and the development of chess strategy. If it is to see who can out play a computer then that is a different contest altogether.
Thomas & Larry: It really isn't necessary to disagree with me as a matter of principle! Believe it or not, I sometimes know what I'm talking about.
Thomas: US Chess Federation correspondence chess rules explicitly forbade using chess engines to assist players in finding moves. You may or may not agree with this policy (ideally, in my view, there would be events where engine use was permitted and those where it wasn't, though there's always the great difficulty of preventing people in the no-computer events from cheating), but it was their policy. In fact, it might still be their policy. How it was supposed to be enforced, well, who knows.
Larry: I'm talking about correspondence chess, you know. Of course in OTB battles engines are more than a match for even the best humans. But correspondence players have consistently beaten engines over the years in correspondence games. In 1999 there was a famous challenge match between Stephen Ham - a 1600 OTB player! - and engines, and he barely lost it, 2.5-1.5. (http://www.correspondencechess.com/campbell/ham/ham.htm) That same year OTB expert Volker Jeschonnek defeated the engines 2-0 in a similar match. (http://www.correspondencechess.com/marconi/volker.htm) And in 2004, Arno Nickel, a correspondence chess grandmaster but only FIDE-rated 2128 in OTB play, was able to beat the dreaded Hydra 2-0. (http://www.correspondencechess.com/campbell/interviews/n050620.htm) He used desktop engines there, but if you read that interview you'll see that it wasn't the superiority of his home copy of Fritz that helped him win but his positional understanding. Deep strategy is still primarily the domain of human thinking, not computer calculation.
So yes, obviously, using engines changes the challenge, but it by no means turns the human into a mere appendage to the engine. I haven't played correspondence chess for years, but when I analyze with the computer it is clear that my judgment (not my calculating ability) is still often superior to the engine's.
Dennis, I never disagree with you just for the sake of disagreeing - even if I sometimes have my own opinion. In the given case, I actually agree with you regarding the main point that computers don't make correspondence chess redundant - see my first post in this thread. My question was rather out of curiosity and I couldn't know that you know what you're talking about, i.e. that the US Chess Federation explicitly forbids computer assistance. Thanks for this clarification! However, to the best of my knowledge (dated, but based on conversations with top players) this is not the case elsewhere.
I just checked on the ICCF (International Correspondence Chess Federation) webpage: They don't mention and ban computer assistance explicitly. A clause "Games shall be played in accordance with the FIDE Laws of Chess where applicable" could be interpreted in such a way, or it could merely indicate "no need to repeat en passant and castling rules here". And they go to great length regarding if and when ongoing games can be posted on the Internet (only with the opponent's consent, always delayed by three moves with respect to the live game).
There may be two different philosophies:
1) Computer assistance isn't a major problem anyway, and/or a rule that cannot be enforced doesn't make sense and may only punish those who abide by the rule.
2) Most players will consider it a matter of honor to abide by the rule. A rule makes sense even if some players break it. "Cheaters" will be caught sooner or later, one way or another.
Actually, if the USCF rule literally states that "using chess engines to assist players in finding moves" is forbidden, this may still be ambiguous. Is it OK to use engines to double-check your own analyses? If so, what if the engine suggests a strong move the player hasn't thought about, either replacing the candidate move or further down in one of the variations? Is a player supposed to forget this information immediately??
Thomas,
The common-sense answer, which can often be skirted by those who are really interested in having their way, fair play be damned, is that you should not check your analyses with a computer either. But one thing I have wondered about, and which does seem worth considering even by honest-minded players, is just how far you should be able to check opening prep. If I'm both a tournament and a correspondence player, then if I play super-sharp, deep lines in the Najdorf or Semi-Slav, then what am I supposed to do: stop exploring those lines with my engines until I'm as sure as can be that the correspondence games have passed the point of any possible theoretical significance? That seems like a gray area to me.