Kimo unmasked – a misleading approach to chess

Kimo – the full truth

After extensive research over a number of days our editorial staff has come up with the following facts regarding the report posted on this site on April 1st.

1. The report described a new chess program that does not use brute force, but works with chess knowledge derived from 20,000 master games. This is true, such a program does indeed exist.

2. The report stated that in spite of some glaring defects the program was able to hold its own against the world's strongest programs. Examples and scores were cited. All of this turned out to be accurate.

3. Kimo, it was said, was created by a team of Russian programmers "hailing originally from Armenia and neighboring republics". This too was true. Kimo originated in the Azerbaijan city of Baku, but is now being developed in Moscow.

4. It was claimed that former world champion Mikhail Botvinnik directly contributed to Kimo's chess heuristics. This too turned out to be accurate – apparently early Kimo development took place in the "Botvinnik School", probably some kind of R&D company.

5. Yes, the box. Here we are uncertain about its actual existence, especially since we found a very similar packaging for a different (traditional brute force) program on Amazon. It would seem quite extraordinary that the Russians came up with so many identical design elements for their box. Someone in our company attributed the similarity to a "photo shop", but we have not been able to determine where or what that is.

6. The diagram positions included in the article actually occurred in test games played between Kimo and the programs mentioned (Deep Junior, Deep Fritz). Kimo played two matches in January and November 2003 and actually did hold each of these rival products to a draw. Further information on the matches may be found here and here.

7. The report said that Kimo ran on "massively parallel special-purpose hardware". This turns out to be accurate. The hardware is manufactured by a company called Nature and its generic brand designation is "brain". Apparently it contains 100 billion painfully slow processors working in a parallel network, with thousands of connections between individual processing nodes. We have no idea how such a system can be constructed.

8. The size of the opening book (30,000 positions) and the knowledge base (20,000 games) turned out to be roughly accurate. We have been informed that in spite of its gigantic hardware capacity Kimo is unable to store more positions or games (which we find hard to believe).

9. The full designation of the program turns out to be "Garry Kimovitch Kasparov". Apparently the sales department decided that you cannot market a product with such a cumbersome name, so they shortened it to "Kimo".

10. The report stated that the running expenses of Kimo were very high and that individual games played by the program could run a bill of tens of thousands of dollars. We contacted the Kimo agent and learned that this is indeed true. There is only one copy of the program available, and it has to be rented (or in their jargon "invited"). The costs for spectacular public events, including flying in the hardware first class, maintaining it in special air-conditioned suites and priming it to play against human or computer opponents, can cost anywhere between $30,000 and over a million dollars.

We apologise to all visitors of our news page who may have got a false impression of the program. We strongly recommend that any orders that were placed for the purchase of the program should be reconsidered in the light of our research results.

Some previous misleading reports in April

• New Fischer moves
• Fischer moves – a cruel hoax!
• A classic retractor
• The bionic chess interface
• 'Please don't touch my hippothingy!'

Feedback

Here are some of the many hundreds of letters we received after our review of Kimo was published. The first arrived minutes after the news item was uploaded, most arrived in the course of the publication day and on April 2nd. The following selection was conducted by a semi-automatic process and does not reflect on the quality of the messages that have been omitted.

Matiss Silis, Riga, Latvia
This is the best of your April fools article :) The toughest one to crack from the recent ones. Thought, my position pattern recognition still works and I could remember the positions from the Man vs. Machine matches. :)

Dave Clarke, Oregon, USA
Ha! Another brilliant April Fool article. Can I get Kimo installed on my own personal massively parallel computer? (Mine is cheap, it runs on Margaritas and unsound gambits, but it is badly in need of an upgrade). Thanks for the smile.

Devangshu Datta, Delhi India
Is this another of the wonderful April 1 Chessbase reports? I like the cute scattering of clues and in-jokes.

• Your Russian knowledge-based program has the same name as Garry's patronymic, (Kimovich) and it hails from Armenia (just like GKK's mother did).
• It implements some original programming by Mikhail Moiseyevich - just like GK did after he went to the great man's school.
• Kimo masterfully beats Deep Junior etc in exactly the same positions in exactly the same ways that GK did. It blunders into exactly the same Rh5?? loss.
• It has a very high-class opening database of around 20K games - about the same that you'd expect a human 2800 player to have. Each massively parallel game costs $10,000 upwards - you wouldn't get GK to sit at the board for less.
• But it has never blundered in opening play vs a computer?? Really? What about the last game it played against IBM's Deep Blue when it forgot a standard anti-Caro Kan piece sac?

Nice story anyway like the rest of your hoaxes! Keep up the April 1 tradition!!

David Bush, Brightstar
Very entertaining article.I had just arisen from a deep sleep, read this article and about half-way through, seeing some familiar positions, realized that Kimo is Garry Kimovich Kasparov. The 20,000 game database was suspicious too! Thanks for the morning smile.

Adam Smith, Denver, USA
I would really like a copy of Kimo, it sounds like a computer program that would do exceptionally well against top humans. A program that would perform consistently at the top of the super GM touraments with maybe one ocassional lose to the young generation and missed wins turned into draws here and there. But why stop at using the program just for chess playing. I bet it could write some great books, and I think it might be able to use its resources in Russian politics (but we might have to wait and see how that turns out).

Stephan Taylor, Melbourne, Australia
Great news on KIMO. Maybe on April 2nd, you can get Kasparovs' opinion on this.

Carsten Hansen, Hackensack, NJ
A cleverly disquised April's fools' joke – Kimo very much sounds like Kasparov's middle name, but what really gave it away were the positions, especially the one from the recent match in New York between GK and X3D Fritz, and of course the identical match results. Keep up the good work!

Maxim Sorkin, Los Angeles, CA
Very nice hoax, guys! Keep up the good humor!

Sunny Nahata, State College, USA
Well, looks like this is the April Fools joke this year. Does Kimo stand for Kasparov? All games shown are from Kasparov vs. Deep Junior or Fritz. If this is a joke, this better be your last one coz I am coming to bash you guys!!

Carl Bicknell, London, England
April fools day has stuck again! Why did you say "millions of dollars" at the end, that gave it away immediately as being a hoax.

Ivan Markovic, Belgrade, Serbia & Montenegro
In his otherwise extremely interesting article, Mr. Friedel failed to mention that Kimo's games vs. Deep Fritz 8 & Deep Junior 8 were practically clones of the famous Kasparov games played against the very same silicon opposition. Given the fact that Kimo's equipped with a database containing all the games from World Champions, it is no wonder that it chose to follow Kasparov's moves (or patterns). Therefore, it'd be much more interesting to see it pit its chips (or that overcomplicated parallel stuff) against top humans: somehow I got the impression that it was purposefully created to highlight and exploit the obvious Fritz & Junior weaknesses that had already become well-known from the previous Man vs. Machine encounters (why would anyone create such a tiny opening book, if not to avoid playing non-familiar positions?). If that's the case, it is most likely that it wouldn't be particularly useful against top humans, known for their versatility and ingenuity. Naturally, the best possible answer to my challenge would be to see more Kimo's games, where it would prove its class and show some "creative" detours from well-trodden paths: so far, we have hardly seen anything more than a below par Kasparov copycat... Anyway, thanks for the article and keep up the good work! – International Master Ivan Markovic, Chess Informant Editor

Henrik Dinesen, Naestved
From the games against the other engines, one might suspect Gary Kasparov to be one of the programmers, as he's well known for making those good programs. I'm almost ready to buy this massively special hardware, since I think there should not be spared anything when it comes to good computergames!

Mike
When I looked at Deep Jr.-Kimo, I realized it's April's Fool. White plays Rxh5 and it's not so equal anymore. PS: Do not give out my e-mail address.

Junior Tay, Rivervale, Singapore
Nice one. So when is the program 'Kramo' coming out? I heard beta tests shows that it draws most of its games....

Geir Skjøtskift, Oslo, Norway
The king on the Kimo box looks strangely familiar. Is it perhaps the king from the Fritz family of programs? And might this Kimo thingy be a figment of your imagination or perhaps its just 1st April. But it would have been nice though.

Michalis Kaloumenos, Kallithea, Greece
Frederic, I'm so glad that you do not forget tradition. This year, it was not Bobby Fischer, nor any rodents, but rather a chess monster. Please, I need an answer to the following questions: 1. Is it possible that the Armenian neighbouring republics are connected (somehow) to Azerbaijan? 2. The package stares at me with these eyes... Whose eyes are these? Next time these eyes come to Greece, I have to meet them and ask to sign a book about some grerat chess players of the past. 4. Deep Fritz was in my wish list. Not any more. I cannot accept to buy a program that has managed to achieve the same losing position after 45 moves of playing. Is it possible that Deep Fritz programmers allowed such a disaster to happen? (The same applies to the remaining positions, I just cannot recognise them.)

Seriously: I definately believe that such a successful way of thinking should be used for educational purposes rather than computer programming. As Dr. Tarrasch said, "Chess, like love, like music, has the power to make men happy". So, please, if you are ever able to achieve this goal, offer it to the world as a gift to humanity, instead of constructing the ultimate, the unbeatable opponent. Until next year's April 1st, I wish you all keep up good work,

Eric Smith, Republic, MO
I don't care how much it costs, I must have this program! Plus the 1 year membership to playchess is a real bargain. You will find a blank cashier's check enclosed, just fill in the price. Next-day Air is preferable. I am currently tearing apart my CD-ROM drive to modify it to fit Garry Kimovich inside. I am ordering a year's supply of Russian cuisine to be piped to my home, as well as downloading all ethnic Russian music in mp3 format from the year 1804 to present. Must make him happy in order to get maximum chess playing ability. You got me last year with the mouse implant thing (I was like "Honey, come look at this! This is amazing!"). I was even telling the guys at work about it the other night, and wondering what you would come up with this year. I'll confess that I read the whole article and was on my way to work before I went "Hey, I know where I've seen those games before!" Sweet!

Peter Johnson, Romford
I am an April Fool. Excellent spoof, only realised today. I fell for it hook line and sinker. Duh. its not on Amazon. Only runs on massively parrallel computers - pah! It was staring me in the face. Well done.

Charlie Linford, Brighton, England
A very humorous April Fools; I must admit, even though I initially looked at 'Kimo' and thought Kimovich, recognised Kasparov's eyes on the box, and even thought "that diagram looks very familiar". It wasn't until I got to the line in which Kaspy overlooked the mate that the penny finally dropped. I'm now going to pass this on to as many of my friends as I think will fall for it.

Jon Fred, Rockford, US
I wish to protest in the strongest terms the two month delay in the US distribution of Kimo. I have any amount of underused, massively parallel special-purpose hardware hidden away in my secret laboratory. Nowadays when I want to play Botvinnik I have to pay Madame Zorza $50 for a half-hour sceance, and it's rarely satisfactory. The old fellow seems so debilitated by, well, death, that he can barely remember the four-knights game. But just let some up and coming Armenian fellows invent a program we could really use, and before you can say "sharpen your pencils", the greedy folks who inflict Fritz and Shredder on us have bribed the bureaucrats and kept the product off the shelf "for two months." Sure. In two months, they'll probably say it was all a joke.

James Norman, Austin, TX
That kimo article was a joke, right? The position of it's tourney game with deep fritz (2-2 result) is exactly the same as kasparov-X3D fritz which also had a 2-2 overall result. It also held a 3-3 tie with deep junior, says it's from russian when it's really not, and "trained with" Mikhal Botvinnik. That sounds way too close to Kasparov! But, funny joke

Lars Dock, Stockholm
Kimo would be fun to test, if it only existed!!! Btw I liked the ninja on the boxcover!

Derek Jones, Aylesbury, England
The dummy box with the words "Runs on massively parallel special hardware" gives it away immediately as an April 1st hoax. In your annual series of April 1st hoax articles I still rate the Bobby Fischer "take back one move per game" the best.

Ashton Anderson, Calgary
I was reading the Kimo article with supreme interest until about halfway down when I realized I was the victim of another Chessbase April Fools Day attack. How could I be so stupid? "Kimo"? Knowledge-based chess program out of nowhere? Anyways, thanks for the early morning wakeup call; maybe I shouldn't go write that Social exam?

Grima, Xavier, Malta
The game positions gave it all away. Cool disguise of Kaspy's eye as a ninja.

David M. St. Pierre, Salem, IL USA
I read with interest the article on the new chess program KIMO. However, the espression "runs on massively parallel special purpose hardware" on the package and the dateline of 01.04.2004 made me immediately suspicious that this was an April Fool's joke for 2004. I would definitely need further corroboration before I believe in the veracity of this review.

Patrice Vezeau, Montreal
I instantly recognize ur Kemo aka Kasparov game...

Wallace Hannum, Berkeley, CA
I found your article about the new program "Kimo" very interesting. But I have a sort of odd question. The eyes on the box's cover look a whole lot like Kasparov's. In fact, I'm rather sure of it. Can they use Kasparov's image on their box and not credit or pay him? Also, even the name "Kimo" has an obvious connection with Garry.

Wallace Hannum, Berkeley, CA
D'oh! I sent you an email recently about the program "Kimo" which has Kasparov's eyes on the cover. Now I realized the date. You guys totally had me for over an hour. You cheeky monkeys! Thanks for a great April Fool's. I love this tradition!

Count Zamoyski, Sopot, Poland
You had me for a while! Then I saw that closed position from the Kasparov-Deep Fritz game. LOL

Jonathan O'Connor, Leixlip, Ireland
Very amusing. Sadly your previous article on Russian patronmymics gave the game away. Still Garry Komovich should enjoy the ribbing.

David, Moore, Sicamous, Canada
A wonderful bit of writing Frederic! This year you "fooled" me completely, or at least until the Conclusions section. I was puzzled when first looking at the picture of the box as to why anyone would market a program which requires "massively parallel hardware", and the position Kimo vs Deep Junior was naggingly familiar to me, but I just kept on reading and taking it in! Thanks for the nice start to the first day in April.

Terry McCracken, Shelburne, Canada
Did you know NASA has confirmed life on Mars? April Fools' to you as well!:) It's a nice little story, but I didn't know Kimo is an alias for Kasparov:) He's even on the box...hiding like a Ninja;-)

Chris Taylor, Nottingham England
With all the over-used chess progs, it will be nice to have a little Kimo-therapy! At last some program with 'NO FOOLING.'

Mark Galecki, Sunnyvale, U.S.A.
In your article on the program Kimo, you forgot to mention that the full name of the program is Kimovich, that it excels in its tournament mode and that it is actively involved, in a team with a human female programmer, in creating offspring programs with similar capabilities!

Chris Harper, Chelmsford
Sorry but you haven't fooled me with this April fool!! Kimo is actually Garry KIMOvich Kasparov!!! And you didn't fool my friend either who knows nothing about chess but knows extraordinary amounts about everything else. The box photo gave it away though, having a PC logo on it but then saying "runs on massively parallel special-purpose hardware"! I also recognised the chess positions. But I like your sense of humour! Keep up the good work!

Navin Sawalani, Indiana, USA
LOL. Hilarious. I didn't realize that you were referring to Garry KIMOvich Kasparov until I saw that Qf3 move with the Fritz game. I thought to myself: didn't Kasparov do that? But I don't think he lost - he drew. But then I saw the one against X3D Fritz, and that was a giveaway. Then I remembered the date. April 1st! U guys are famous for April 1st jokes, like that retractor chess. Anyway, good one Frederic. Nice touch with Kasparov's eyes on the Kimo box. LOL!

David Levine, NY, USA
Ha-ha! you guys have done it again. i had actually forgotten all about the annual april fool's day article, and was a good quarter-way into the article before the "test game" positions started looking eerily familiar. Good job, Frederic, you've certainly topped the neural chip bit from last year with your rather funny look at kimo(vich).

Michael Babigian, Sacramento, USA
Nice one. Very amuzing. Who came up with the name? Happy April Fools.

Len, NH
All well and good, though Kaspy might not like that he cannot "compete commercially with the other engines" ;)

Paul Leyderman, Chicago, IL
Very cute! I didn't spot a hoax until I got to Gary's position vs. Deep Fritz 8! Details like "a team of Russian programmers (hailing originally from Armenia and neighboring republics)", Mikhail Botvinnik's contributions, and Gary Kimovich's eyes behind the mask are greatly appreciated.

Murali Vajapeyam, Boston, United States
As per tradition, once again a nice April Fool's Day joke by the chessbase team. And this was so convincing! Unfortunately you guys gave it away completely by using the (in)famous Kasparov games against Deep Junior and Deep Fritz. Were it not for that I would have believed Kimo for real... (Botvinnik "directly contributed to Kimo" - hahaha). As the article seem to imply, it is the case that GMs have virtually no chance against computers in blitz games? ("Kimo will not be able to occupy the top spots in blitz rating lists"). Or is it that since Kasparov is not a program he can't be on the list?

Predrag Miletic, Belgrade
Good one guys...

Timothy Chou, New York
very interesting article! Hopefully we will see Kimo around in the future!

Ian Olsen, Kadena AB, Okinawa Japan
Happy April Fool's Day to you to! I was buying your story about the new chess playing program Kimo until I saw your position from a "Kimo vs Deep Fritz 8" game, which looks suspiciously similar to a position that Kasparov had in his last computer match.

Omer Tunali, Sunnyvale CA
Nice joke...

Justo, Argentina-France
I see, this is a joke. Kimo is Kasparov. The same positions as those of his matches against the machines. Parallel hardware (the brain). Same style. Adittional clue: in the box of the program is written PC-CD Rom. But you are saying that this program doesn't operate on PCs.

Sébastien Benoit, Montréal
One of the positions supposedly played by Kimo is one that was reached in the 3rd game of the Kasparov - X3D match. I even recognised Kasparov's eyes on the picture of the box of the "new product". Very good April 1st joke!

Omer Tunali, Sunnyvale CA
The third game played between Kasparov and Deep Fritz was apparently in the 20,000 "high quality" game database that Kimo analyzed and deduced heuristics about chess. It is such a coincidence that Kimo(vitch) and Fritz ended up playing the same position :-) Nice joke, and Happy April Fool's day.

Cilmar Thomé, Porto Alegre, Brazil
Excellent program! The only one thing I dislike is the name. I would rather call it Kimovich. Or Garry.

Merim Bilalic, Oxford
The two examples of the long games Kimo played against Fritz and Junior are remarkably similar to the two only decisive games Kasparov and Fritz played in their last match. Do you know the reason?

Larry S. Tamarkin, Levittown, NY, USA
Not quite as funny as some of your previouse April 1st jokes, but I did enjoy Kasparov being made fun of as a chess program, with a relatively small opening book, but only of top quality games. Additionally the point about the "custom-built massively parallel hardware, and it also requires extensive care and maintenance. All of this is exorbitantly expensive compared to contemporary cash-and-carry PCs. We estimate that individual games played by Kimo can run a bill of tens of thousands of dollars". This all had me laughing out load as I thought that a number of my chess programs were marketed to me on the basis that they could immitate the 'personality' of Kasparov (or Tal, Fischer, Karpov, Judith Polgar & others), by their settings - So what could be funnier then a program being designed to immitate a human imitating a computer imitating a human?

Meir Lakhovsky, Redmond, WA, USA
Ahh yes Garry Kimovich, isn't it. That was a nice box though, cool design, it had me fooled until the poitions. The first position looked farmiliar, I thought that maybe it had studied Kasparov's games and used his oppenings and got similar positions, but the second game just gave it away. Happy April fool's day to you too and Keep up the jokes!

Sameer Wadkar, USA
Nice April Fool Joke!! I was waiting for this years joke.

Francisco Chavez, Mexico City
Nice, very nice! I confess that the first time I read through your Kimo article I found some data weird (the several thousand dollars required to run Kimo, for instance) but totally fell for it. Then, the positions seemed familiar... and there was also something familiar with that name "Kimo". Then, suddenly all the pieces fit and I realized I was beeing fooled!! Actually, once you understand it, the whole story makes a lot of sense. Happy fool`s day!! (We have a similar tradition here in Mexico but it is on the 28th of December)

Steven Hill, Adelaide, Australia
I was initially excited to read about a program that uses more chess knowledge but... Though it's already April 2nd in Australia by the time I read this I was not the April Fool. I immediately noticed the example games were very Kasparov like - infact, taken exactly from the Kasparov games against the programs mentioned.

Stephen Goodman, Falls Church, VA, USA
Enjoyed your piece on the remarkable program Kimo 1.0. I have heard a rumour that the program, although notably unpopular with other leading chess programs, fancies itself quite a skilled politician and may seek the Russian presidency in 2008. Care to comment?

Chris Meyer, Chicago, USA
I only have one question about Kimo: can mice control it telepathically?

Torstein Hall, Asker, Norway
I had hoped for a better 1. April joke than Kimo. Remember, this kind of jokes are supposed to be funny. Or is the funny part that you put a Kasparov game as a computer. I do not think he is massivly paralell yet, but probaly very expencive.

Ashwin Phatak, Pune, India
I almost believed it, but then saw the date: 1st April!

Erik van Dijk, Hoofddorp, Holland
Very funny. At first I thought: This could be a big step in computer chess... Then I recognized the test-game 'kimo'-Deep Fritz!

Stephane Desnault, Paris France
Well done! It took the absurd special purpose hardware mention to dismiss the article as a joke. The notion of an intelligent chess program using only 20 000 games was just too alluring I guess. I'll start saving, and in precisely one year after this article I think I'll be able to buy the program and the hardware.

Johannes Steckner, Zuerich
Your report on the new Program Kimo is interesting: its playing features strongly remind me of those of Gary Kimovich Kasparov, even its name. Maybe no pure coincidence on the 1st of April.... Thank you for this very well worked out story!

Gregory Topov, Canada
Ok, it's April 2nd now, so you can tell us that the article about Kimo was an april fool joke. I liked the one from www.academicchess.com about Fischer playing Deep Blue better though.

Timothy Webster, White Plains, NY, USA
Nice work with the April Fool's joke about Kimo. You had me going there for a while. The "massively parallel" part was a dead giveaway. By the way, did you choose the name "Kimo" after Kimo Nakajimi? A British paper played an April Fool’s joke on its readers by telling them that the Japanese marathon runner Kimo Nakajimi’s lack of understanding of English led him to continue running a marathon that he thought lasted for 26 days, instead of 26 miles.

Radovan Grujic, Prigrevica, Serbia
This article is very original 1. April joke, you have very good sense of humor. I was very surprized when I saw the title of the article, because I knew that all the modern chess playing programs use "brute force" calculation algorhytms, so it was a big surprise to see that "Kimo" uses different, "intelligent" approach. I immedietly noticed that eyes on the picture next to article look very similiar to Kasparov's eyes, so it was very suspicious to me, and when I saw the position from Kasparov-Deep Fritz 8 match, it was clear that the article is an april joke! "Kimo" is a Garry KIMO-vich Kasparov! Very good joke.

Henk Nijmegen, Netherlands
Until I read: 'Mikhail Botvinnik (...) directly contributed important elements that are today part of Kimo's chess heuristics', I almost believed it. Botvinnik tested his program only on positions from his own games and the program always found the bestmove, surprise, surprise, the move Botvinnik played. It was a joke. And your story, also a nice joke. Or am I wrong?