Kuro5hin.org: technology and culture, from the trenches
create account | help/FAQ | contact | links | search | IRC | site news
[ Everything | Diaries | Technology | Science | Culture | Politics | Media | News | Internet | Op-Ed | Fiction | Meta | MLP ]
We need your support: buy an ad | premium membership

The crème de la crème of the chess world

By Dlugar in Technology
Mon Sep 10, 2001 at 01:59:07 AM EST
Tags: News (all tags)

... in a show with everything but Yul Brynner.

Vladamir Kramnik, since he beat Garry Kasparov 8.5 to 6.5 in last year's World Chess Championships, now holds the title of the World Chess Champion. This October in Bahrain, he will play an eight-game series against the world's most powerful chess program, Deep Fritz.

The above link quotes Deep Fritz's impressive statistics:
"Put it this way, the Deep Blue supercomputer that beat Kasparov in 1997 weighed 1.4 tons, was over 6 feet high and needed 20 people to keep it running. The new Deep Fritz can run on a laptop. Even on a fast desktop machine it will be able to achieve the playing level of any incarnation of Deep Blue. I'll let you draw your own conclusions as to what it is going to do to Kramnik."
-- Frederic Friedel
To date, Deep Fritz has defeated Kasparov, Anand, Shirov ... all the major grandmasters with the sole exception of Kramnik. Deep Fritz has defeated Deep Blue and every other chess computer it has played against in these tournaments. It can analyze over six million moves a second. Is Kramnik toast?

Not everyone thinks so--not least of all Kramnik himself. "They are inferior in tactics," he says of chess-playing computers. "There is nothing else to say." By the age of five, Kramnik was already attending professional chess clubs, and at the age of eleven awarded the title of candidate master. Now at twenty-five years old, Kramnik, a former pupil of Kasparov, has overtaken his master and is quickly becoming the number one chess player in the world.

Kramnik has also been on his guard in accepting this million-dollar match with Deep Fritz, after noting Kasparov's mistakes with Deep Blue. It has often been said that Kasparov was "ambushed" by Deep Blue--the computer had access to all of Kasparov's past matches, while Kasparov was given no match history for Deep Blue. IBM programmers were able to reprogram Deep Blue between matches with Kasparov, thus creating an entirely new person for Kasparov to compete against every time. And some say that Deep Blue was simply a "programmed Kasparov-killer", specifically programmed to beat Kasparov but not a decent chess player otherwise.

Kramnik, in response to this, asked for a copy of Deep Fritz to "practice with" before the big match. According to the BGN Match Director Raymond Keene, he has received such a copy. In addition, only a few new analyses are allowed to be added between matches--nothing major. Also, rumor has it that Kramnik is far more skilled at thoroughly trouncing computer chess programs. All in all, it ought to be a very interesting and close match.

Meanwhile, is Bobby Fischer back?

The BBC is reporting that Bobby Fischer could be playing three-minute blitz games on-line anonymously against various Grandmasters. Nigel Short, a British Grandmaster who drew a speed chess series against Kasparov in 1995, believes firmly that the mysterious person he has played against is Fischer. Others are more skeptical, thinking it's merely a computer program.

I'm not so sure. Short claims that his opponent made moves that were very bizarre, very Fischer-esque, and not typical of a computer at all. In addition, the person gave various "hints" as to his identity. Not to mention the fact that, as one of the best speed chess players in the world, Short ought to be able to beat the program handily, since in blitz games computers typically only rate in the low 2000s. So perhaps a newly designed chess program with non-standard openings, armed by a human who has memorized Fischer trivia?


Voxel dot net
o Managed Hosting
o VoxCAST Content Delivery
o Raw Infrastructure


Kramnik vs. Deep Fritz, who will win?
o The wetware. 29%
o The software. 16%
o Chess? Why should I care? 14%
o Bobby Fischer. 38%

Votes: 54
Results | Other Polls

Related Links
o last year's World Chess Championships
o Deep Fritz
o he says
o BGN Match Director
o Others
o Also by Dlugar

Display: Sort:
The crème de la crème of the chess world | 28 comments (15 topical, 13 editorial, 0 hidden)
Good article (2.00 / 3) (#3)
by onyxruby on Sun Sep 09, 2001 at 09:25:14 PM EST

Nice well written article, I thought it was well laid out. It is nice to see some other people (here at K5) interested in the game.

The moon is covered with the results of astronomical odds.

News Flash! This just in ... (5.00 / 5) (#6)
by Dlugar on Sun Sep 09, 2001 at 09:40:49 PM EST

Regarding the "side story" of Bobby Fischer's return, I found some more information (*cough, cough* at the other site):


my game with Kramnik on ICC (3.00 / 3) (#8)
by eLuddite on Sun Sep 09, 2001 at 09:50:30 PM EST

  1. e4    f6
  2. Bc4   a3
  3. Qh5   b4

    [Commentator's: Kramnik reacts with his usual combative style, attcking the White Bishop with his pawn and preparing for an expansion on the Queen's side. This should be an interesting game. In his two previous encounters with the black pieces in this match, Kramnik played the disappointing 3. ... Ra7. ]

  4. Qf7++
Commentator's: Shocking! eLuddite has revealed himself as Robert Fischer, noted paranoiac and general all-round nutter.

God hates human rights.

Kinda funny, but... (none / 0) (#28)
by Megumi on Tue Sep 25, 2001 at 08:36:02 PM EST

1.e4 f6
2.Bc4 a3

a3? Do you mean a6 maybe?

3.Qh5 b4

Again, I know that you are joking, but did you actually consider the moves that you're writing down here? 3.Qh5+ makes b4 illegal. Plus, I think that you really meant b5 anyway.

Apparently, it's best to leave chess-humor to the professionals. :)

-- Megumi
[ Parent ]

Short vs. Fischer (3.00 / 1) (#14)
by calibraxis on Sun Sep 09, 2001 at 10:40:54 PM EST

Does anyone know where Short's blitz match is against Fischer? The moves, not just articles. Short definitely has most of the moves memorized, even if he didn't log them.

Never mind (4.00 / 1) (#16)
by calibraxis on Sun Sep 09, 2001 at 11:10:35 PM EST

I just read the link at Dlugar's earlier post. Painful. Fontaine vs. Fischer had humor written all over it.

I don't even care if the games were fake. Brilliant chess is still brilliant chess. And we are past the Soviet heyday.

Great article, Dlugar.

[ Parent ]
Spelling (2.25 / 4) (#18)
by zauc on Mon Sep 10, 2001 at 06:14:50 AM EST

Hi, just a frenchie comment : it's créme not crème.

No it isn't (none / 0) (#19)
by (void*) on Mon Sep 10, 2001 at 06:31:10 AM EST

crème is fine. Have a look in a dictionary if you need to.

[ Parent ]
sorry :-P (none / 0) (#27)
by Dlugar on Mon Sep 10, 2001 at 11:08:49 AM EST

I just copy-pasted from the song/musical. I didn't write it--it was a bunch of Scandinavians (ABBA, IIRC). But apparently both ways work?


[ Parent ]
My money's on Kramnik (none / 0) (#21)
by codemonkey_uk on Mon Sep 10, 2001 at 07:06:02 AM EST

For one reason, and one reason only. He has a copy of the program.

Chess is Kramnik's life, winning this game is going to be everything to him. If there is a weakness in the AI (and there will be one), he will find it. I belive he has enough time to find a winning strategy.

For this game to be fair, he should have had access to Deep Fritz's game history, and vice versa, but nothing more.
"The most savage controversies are those about matters as to which there is no good evidence either way." - Bertrand Russell

one thing... (4.00 / 1) (#24)
by speek on Mon Sep 10, 2001 at 09:30:40 AM EST

For this game to be fair, he should have had access to Deep Fritz's game history, and vice versa, but nothing more.

True, provided they aren't allowed to change Deep Fritz's programming - thus making that game history useless.

al queda is kicking themsleves for not knowing about the levees
[ Parent ]

Agreed (none / 0) (#26)
by codemonkey_uk on Mon Sep 10, 2001 at 10:07:46 AM EST

IMHO, reprogramming should be a big no-no. If an programming team want to make adjustments, the adjusted AI should have to work its way up the rankings again.

AI's that learn by themselves, are, on the other hand, a compleatly different matter! :)
"The most savage controversies are those about matters as to which there is no good evidence either way." - Bertrand Russell
[ Parent ]

From an email (4.00 / 2) (#22)
by calibraxis on Mon Sep 10, 2001 at 08:20:38 AM EST

I am so intruiged by this whole Fischer thing. This is an email I sent a friend. Maybe you might be interested in it.

Three possibilities:

  1. Short vs. Fischer
  2. Short vs. machine + human
  3. Short vs. human
For 1:
Fischer learned that there are blindspots for speed chess players. Where is the best place to get speed if you're playing blitz 3-minute games? From the opening! So, what happens if Fischer plays strange openings? The opponent plays as if it were the middle game. Fontaine played meaningless middle games too early against Fischer, mobilizing his pieces without clear understanding of the position's tactics.

Remember Fischer vs. Byrne when he was 13? Same thing happened there, after Fischer sacced his queen!

Fontaine didn't play chess. He just moved the pieces quickly, with some rules of thumb.

For 2 and 3:
I will say these two are identical. The chances of a program or amateur showing extreme ability are probably close. I mean sure, we can quibble, but I don't care that much. There have been many notable amateurs who have beaten the old Soviets, but were scientists or in other fields.

Today's advatage that professionals have against amateurs is openings knowledge. A person can read all the games, but it soon would become a lifetime profession. But a person smart enough to play amazing chess could get away with not knowing openings by not playing them. That person would just have to be smart enough to not need minor opening advantages.

I hear blitz tiebreakers in normal tournaments called "Lotteries." Opening advantages probably overrated in blitz.

Not as dominant as you'd think (4.50 / 4) (#23)
by Salamander on Mon Sep 10, 2001 at 09:21:34 AM EST

If you look at the results from the recent Cadaqués chess tournament here the first thing you'll see is Deep Fritz beating its nearest competitor by 3.5 points out of 100. However, if you leave out results vs. Gandalf 4.32g - which clearly did not belong in this tournament - the margin shrinks to a measly 0.5/80. That's really too close to say whether Deep Fritz or Gambit Tiger (which came second) is better. In fact, there's a good argument that Gambit Tiger is basically a better program, based on the fact that it achieved this result while lacking the massive opening and endgame tablebases that Deep Fritz uses. With the simple addition of large-tablebase support to GT, it seems highly likely that it would beat the pants off of DF.

GT is also known for its aggressive play. I think it would be much more interesting to see how well that works against defensive specialist Kramnik, compared to a match against bland old Fritz.

For more info... (3.00 / 2) (#25)
by maveness on Mon Sep 10, 2001 at 09:59:07 AM EST

The just-out October 2001 issue of Wired has an article entitled "Deathmatch 2001:This Time It's Personal" by Rudy Chelminski that goes into great detail about human v. computer chess and the forthcoming Brain Games competition which features Kramnik v. Fritz.

Latest fortune cookie: "The current year will bring you much happiness." As if.

The crème de la crème of the chess world | 28 comments (15 topical, 13 editorial, 0 hidden)
Display: Sort:


All trademarks and copyrights on this page are owned by their respective companies. The Rest © 2000 - Present Kuro5hin.org Inc.
See our legalese page for copyright policies. Please also read our Privacy Policy.
Kuro5hin.org is powered by Free Software, including Apache, Perl, and Linux, The Scoop Engine that runs this site is freely available, under the terms of the GPL.
Need some help? Email help@kuro5hin.org.
My heart's the long stairs.

Powered by Scoop create account | help/FAQ | mission | links | search | IRC | YOU choose the stories!