Who wins a match between the world champion and a computer?

[mschribr]

What do people think would be the score between the world champion and a computer? It has been 3 years since 1 of the top players has played a computer. That match ended in a draw.

[A.Presman]

Strongly dependent from the form of competition.
And of power of computer.
If the machine of the level that play in chess ( http://www.research.ibm.com/deepblue/meet/html/d.3.html )
and if the a time that chess programmer can allow themselves preparing the match would be given to draughts programmers.
I'm afraid that machine will be unbeatable. And with the shortening of time it will win at some moment.
But if someone will take the home PC and run the "box" version of a draughts software than human top player can use some specific weaknesses of computer and win.

[mschribr]

The match should duplicate the world champion match with the same time limits. The computer would be a fast PC computer. No custom made hardware. Both sides will have the same amount time to prepare, 3 months. I am not sure what is “shortening of time it will win at some moment”? The programmer can make changes up to the last minute. So the world champion can train with an older version of the program. The computer program would be the winner of the last computer tournament. I think that was TD King. Who would win the world champion or TD King?

[Alexander Kandaurov]

I have played and won two matches against computers programs in 64 board (Russian version), Magistr and Tundra.
Programs have 7 or 8 endgames database (sorry don'r remember), PC was Pentium IV 3200 with 2 Gb RAM
Time control - 30 min. per game. I won both matches +1 and .. was very happy
I think i was last one who did it

[mschribr]

I have played and won two matches against computers programs in 64 board (Russian version), Magistr and Tundra.

Are you saying the world champion would win? My question was about 10x10 International draughts. 10x10 International would be harder to win. Computers play the International draughts better because there are more programs trying to be the best in International. More competition leads to better programs. Are there annual computer tournaments for 64 board (Russian version), Magistr or Tundra?

[Alexander Kandaurov]

Are you saying the world champion would win?

I think chances in present time is 60-40 for man.

My question was about 10x10 International draughts.

Sorry, don't saw this in polls title

10x10 International would be harder to win. Computers play the International draughts better because there are more programs trying to be the best in International. More competition leads to better programs.

Sorry again. Are you programmer?
Let's compare 64 and 100 programs. Tunda has 10 pieces endgames database (full 5X5 and 5X5 with 1 king), practically it's mean that it "see" any position with 7X7 or 8X8 till ED (ending database). What about 100 programs?
From another points. We have 12X12 in start position. Try to receive winning for this short period! I'm not sure it's easy task.

[ildjarn]

Let's compare 64 and 100 programs. Tunda has 10 pieces endgames database (full 5X5 and 5X5 with 1 king), practically it's mean that it "see" any position with 7X7 or 8X8 till ED (ending database). What about 100 programs?

Full 6-piece databases are generally available (Harm Jetten sells them on his website). I think Stef Keetman (and probably others) did several 7-piece databases too.

[mschribr]

Are you saying the world champion would win?

I think chances in present time is 60-40 for man.

60-40 sounds like the wold champion has a small advantage. It would be an exciting match that could be a draw. I almost agree with you. I also think it is 60-40 but in favor of the computer. That also would be exciting draughts that also could end in a draw.

My question was about 10x10 International draughts.

Sorry, don't saw this in polls title

Sorry the poll did not say 10x10 international draughts. Being 10x10 international is the main draughts game of the FMJD I did not think it was necessary to state. I’m sorry for the misunderstanding.

10x10 International would be harder to win. Computers play the International draughts better because there are more programs trying to be the best in International. More competition leads to better programs.

Sorry again. Are you programmer?

I am a computer programmer, but not of games. Congratulations on winning 2 matches. You are a very good player to beat the computer.

Let's compare 64 and 100 programs. Tunda has 10 pieces endgames database (full 5X5 and 5X5 with 1 king), practically it's mean that it "see" any position with 7X7 or 8X8 till ED (ending database). What about 100 programs?
From another points. We have 12X12 in start position. Try to receive winning for this short period! I'm not sure it's easy task.

Are you saying beating the computer in 64 board (Russian version) is harder than beating the computer in international draughts? Because the computer uses 6 pieces ending database for international draughts compared to 10 pieces ending database for 64 board.

But the ending database does not come into play until near the endgame. In the middle game an inferior program will be outplayed so it will lose before it can use the endgame database. Today’s refined international draughts programs play the middle game at a high level. I believe this more than compensates for fewer piece endgame database.
Mark

[Hanco Elenbaas]

Match between former World Champion Aleksander Kandaurov (left, behind the black pieces)
and computerprogram Magistr-III, 4th game, Moscow january 2003


Former World Champion Aleksandr Kandaurov destroys another computerprogram in 2002

[mschribr]

I am a casual draughts player. I did not know I was speaking to a former world champion.

Where can I find more information about computers playing the 64 board (Russian version)? Have these programs played in any major human tournaments? Is there an annual computer 64 board tournament? Where can I find 64 board ratings?

I see the poll shows people are evenly divided. Half believe the world champion will win. Half believe the world champion will lose. There should be a match to decide who is better.
Mark.

[A.Presman]

Popular Russian newspaper Sport Express published after the World championship 2005 an interview with the winner Alexei Chizhov.
http://www.sport-express.ru/art.shtml?111713

Alexei is not only 10-time world champion ( and so the strongest draughts player all time ) but also the computer freak. Therefore his opinion is very interesting

Can you fight the strongest computer program?
Why should I? I won't loose, but I can't win because computer defends phenomenally. He sees all my threats many moves before. And to play blitz games agains computer has no sense, he will never make a serious mistake and certainly will win. Draughts players with self respect use computer as database and sparring partner, first of all for studying of end games. Nobody is planning fight him.

- À âû ìîæåòå áîðîòüñÿ ñ ñèëüíîé êîìïüþòåðíîé ïðîãðàììîé?
- À çà÷åì ñ íåé áîðîòüñÿ? Ïðîèãðàòü ÿ åé íå ïðîèãðàþ, íî è âûèãðàòü íå óäàñòñÿ, òàê êàê êîìïüþòåð ôåíîìåíàëüíî çàùèùàåòñÿ. Îí æå âèäèò âñå óãðîçû çà ìíîãî õîäîâ âïåðåä. À â áëèö ñ êîìïüþòåðîì èãðàòü ïðîñòî áåññìûñëåííî, îí íè÷åãî íå çåâíåò è îáÿçàòåëüíî ïîáåäèò. Óâàæàþùèå ñåáÿ øàøèñòû èñïîëüçóþò êîìïüþòåð â êà÷åñòâå áàçû äàííûõ è êàê ñïàððèíã-ïàðòíåðà, â îñíîâíîì äëÿ èçó÷åíèÿ ýíäøïèëÿ. Áîðîòüñÿ ñ íèì íèêòî è íå ñîáèðàåòñÿ.

[mschribr]

Thank you for posting and translating the quote from Chizhov.

Chizhov thinks such a match will be a draw while others polled think there will be a winner. His opinion is based on his expert knowledge but it is only an opinion. He knows the match will end in draw so why bother playing. He is overconfident. He also is practical in seeing no benefit in playing a tournament against the computer. What about playing or us watching an exciting draughts match at the highest level?

Has the World Draughts Federation considered asking a company like Intel, AMD or Dell to sponsor a tournament? The tournament could have a prize and promote their computer. Would Chizhov decline an invitation to play? I think a computer company would be interested if the draughts program ran on their computer. The computer company could lend a very fast computer. Many people are interested in man against machine. Draughts will get publicity. The computer company will get publicity.

Chizhov attitude to the computer is different from other great champions.

Kasparov the greatest chess player believes chess requires the greatest amount of imagination. A computer has no imagination so it should not be able to beat the human brain. Kasparov also said all of his opponents were not up to his level. He wanted a stronger challenger. Then he could show us how great he really is.

Tinsley the greatest checkers player also thought the human brain is superior to the computer brain. He also wanted a stronger challenger. None of the players of his time were near his ability.

If we compare draughts to other classic strategy games like Shogi and Go. In these games the world champion can easily beat the computer. Does that mean draughts is an easier game requiring less imagination? What does it mean when Chizhov is not willing to accept a greater challenge? Has Chizhov reached his absolute best?
Mark

[Peter van Heun]

It is interesting to have a match between human and computer. There is a in general curiousity to compare man and machine. In particular on a mental level. Therefore mind games are a well suited. The match between Kasparov and Deep Blue of IBM did generate a lot of publicity. Kasparov also gained financial benefit. (I do not know if he also get positive appreciation for his chess games.)

(International) draughts could also benefit from a match between the world champion draughts player and "the best" computer program. Rejecting the challenge would surely mean no positive publicity. The conditions, preparation, end game databases, time limt etc., should be carefully chosen.

[mschribr]

The conditions, preparation, end game databases, time limt etc., should be carefully chosen.

What would be the ideal conditions, preparation, end game databases, time limt etc.?

[Peter van Heun]

The conditions, preparation, end game databases, time limt etc., should be carefully chosen.

What would be the ideal conditions, preparation, end game databases, time limt etc.?

I'm not sure. I'm no expert in this field. Experts are needed to determine the correct conditions. Although I have some ideas:

• 1. A reasonable amount of money should be payed to the participants. Both people managing the computer(program) en player. What is reasonable? I don't know but I guess at least E 10.000,- each.
  2. At least three months of preperation. I think this is common between humans. (I doubt if this condition is fare towards the computer. How should the computer program prepare?)
  3.No use of databases. It should be a challenge between man and machine at the time present. Calculation times before actualy playing, used to create databases, is not a part of the challenge. (Preparation in advance is a different type of condition.)
  4. The usual time limit of 25 moves within one hour is applicable. Other time limits are also possible as long as it not to fast.
  5. One game per day.
  6. At least ten games total. Maximum twenty games total.
  7. A quiet playing room.
  8. Ensured electrical power.
  9. Orther conditions....?

[/b]