Fritz, a tool for chess players

FRITZ COMPUTER chess programme has achieved tremendous success in the nine years since it was first invented. The latest is a 4-4 draw it made against four of the top six players in the world.

The father-figure in chess programming is Frans Morsch, a 47- year old Dutchman from Dronten. He is a small made enthusiastic researcher, who has revolutionised commercial programmes to new levels.

Fritz is a former all-category world champion among computers and now tops the rating among them too. Backed by the largest chess company, Chess Base, it promises much more in future.

Frans Morsch in an exclusive chat with TheHindu gave his views about the programme:

Question: What background did you have before taking up programming?

Answer: I studied Physics and as a hobby I had electronics and chess. This was in the late seventies and that time we got the microprocessors. I built my own computer and wrote a chess programme for it.

Q: What language was the programme written on?

A: It was an assembler programme built in the early eighties. It was a small and primitive programme. But I was able to sell that programme to a manufacturer of Chess computers, Mephisto in Munich. They needed a very small programme for a low-cost computer. That started my chess career with computers. That version was the Mephisto Mondial, my first chess programme. I think it came out in 1984. The whole series of chess computers started with Mephisto, later on with another company which went bankrupt. Then I did a lot for Saitek.

That time I started to do a programme for experiments which run on a personal computer. Chess Base, Hamburg was interested in that programme in 1992. That was the first Fritz programme. That was not very strong. The next versions were stronger. I am now at version six. With each version we were able to give a good boost in playing strength. So, now it is on top of the rating list.

Q: The performance here has been good with black, not so good with white. Is there some reason behind this?

A: Not necessary. Humans play very cautiously. Even here, (Leko- Fritz, second game was in progress) it is a Sicilian, they try to close up the position. They are very careful. They show a lot of respect for Fritz. So it is difficult to take advantage of it.

Q: How many hours a day do you work on Fritz engine?

A: I have normal work time at home. I have been a professional for 16 years now. Only on Fritz for the last two years. I have also done a lot of Saitek programmes. There is not much activity for normal chess computers. Now I am full time for Fritz on the personal computer.

Q: How do you see the future ahead for Fritz?

A: I think there will always be a demand for a strong chess programme on the personal computer. I have always been able to improve quite a bit for the next versions. We already have enough for the next (Fritz-7) release.

Q: Do you think that at some point the sales of Fritz will catch up with Chess Master series, which sells more?

A: Chess Master is for a different kind of market. It is for the lower end market and appeals a different kind of customer. For example, the colours of boards and the graphics are all different and made to look attractive. We aim more at serious chess players. Chess Master is a game and Fritz is a tool for a serious chess player with lots of database functions, analysis functions, etc.

Q: When you make the engine, do you make Fritz with another computer as opponent or as a human opponent in mind?

A: We test a lot against computers because it is important to beat the competitor. If you test only against computers, you get bloodless sort of chess. They only play part of the game. They don't play their entire game. So we always try to play against strong humans. It is very difficult.

Q: Can you list the achievements of Fritz over the years?

A: The most important victory was the all-category World Computer Championship at Hong Kong in 1995. We did exceptionally well in the blitz tournament at Munich 1994 (beating Kasparov, Anand, Kramnik, etc.). Last year we won the Masters Tournament in Frankfurt in rapid chess and the year before that we won the Ordix Open rapid as well.

Q: Now you are leading the elite Grandmasters 4- 3?

A: I don't know (He did not like Fritz's position against Leko, which ultimately Leko won). This is a competition against Kasparov.

Q: How?

A: The idea is to see who does best against these five strong players, Kasparov or Fritz. Because Kasparov did not want to play against Fritz.

Q: Are you sure this evaluation would be the right one? Players come with different attitudes and goals playing against Kasparov and Fritz.

A: Absolutely. I would love to play against Kasparov. He refused to play Fritz.

Q: So a lot will depend on the match with Shirov?

A: Shirov is a very tactical player. Of course. But I do not know what he is going to do. If he is going to resort to anti-computer chess it will be difficult for Fritz. But if he plays his tactical game, in a rapid chess, then we have good chances of beating him. If you have seen the previous game (against Morozevich), Fritz was very good at defending his position. In the first game against Kramnik it had a difficult position. It defended really very well. It gradually improved its position.

Q: What did we learn new from Fritz which we did not see last time?

A: In the previous event (Masters 1999) there was not much anti- computer chess. Most players played their normal tactical style. Lot of Sicilians came up. Fritz really did well in that tournament and won. Now, we see that it has weaknesses if you close up the position as you see is happening right now. All the pawns are fixed. Fritz has some weaknesses there. We have to work hard there. This is much more difficult.

Q: Is it possible to eradicate these weaknesses and also tell Fritz that two minor pieces are always stronger than a rook and pawn?

A: Oh yes. We have a lot of this in the programme. But chess is so complicated and there is so much happening. It is so much knowledge to acquire. We can never have the exhaustive knowledge of a Grandmaster. Chess is simply too complicated for them.

Q: At which stage can we expect Fritz to beat Grandmasters in rapid chess and classical chess?

A: We are playing against the top ten here. We have a plus score against them. So it plays at a very high level in rapid chess. Tournament (classical) chess is a different matter. It is much more strategical. Humans make less mistakes. We don't have enough games at the moment to prove ourselves or explore this possibility.

Q: How do you see Fritz's chances against Junior or Hiarcs?

A: I think Junior is the second best at the moment. Hiarcs has fallen back a little. He (Mark Uniacke of Hiarcs) is concentrating on his family and has a full time job.

Q: But Anand used Hiarcs in Advanced Chess at Leon to defeat Judit Polgar and Shirov who used Junior?

A: Vishy is a Hiarcs fan.

Q: When can we expect Fritz for multi-processor computers in the market?

A: We have a version running here. He is using eight processors. I think we will release it with Fritz-7 as a special professional version. I am not sure what Chess Base is going to do here. But I think it will be released this year.

Q: Do you make separate versions for players like Kasparov, Anand and others and one for the market?

A: (After a big silence) No, not yet. We consult top players like Anand to learn what kind of analysis functions they want for their opening preparations. Because we aim at making Fritz a tool for chess players.

Q: In the world top 100, which percentage of players do you think are using Fritz?

A: I am not sure. Fritz is tactically the strongest programme at the moment. I always tell players to use it for tactical analysis. For strategical assessment you should follow your own judgment.

Q: Why did he fail to understand the attack on his king by Kramnik?Is there some priorities you can set to safeguard the king?

A: There is lot of coding there. It did not work in this game. It saw the attack much too late. It thought it can prevent the rook manoeuvre with the bishop. I think it might have overlooked the sacrifice of the exchange. It uses a sort of selective scheme. It is the normal selection method. This may have caused it to miss that sacrifice.

Q: How did you get the name Fritz?

A: That was a Chess Base invention. I call the experimental version Quest. Fritz is a first name for Germans and it came from their marketing agents.

Q: Chess Base markets three other top programmes, Junior, Hiarcs and Nimzo.

A: My programme Fritz is the main engine. The sales of other engines are lower. Everybody buys Fritz, only a few buy the others. You cannot make an income from that kind of sales of other engines.

Q: By way of income, how does chess programming pay in comparison with other programming, like say Word Perfect?

A: Chess programming is still healthy business for not too many programmers. I always look if there are amateurs with talent. I saw a few, Hiarcs was an amateur, Junior was an amateur.

Q: Do you have assistants in the programming end?

A: I get lots of advice from Thomas Luther. He is the main adviser for Chess Base. He does a lot of test games and sends it to me with comments. That is very helpful.

Q: How many hours do you work?

A: I don't work more than 40 hours a week. Sometimes I like my work and sit behind my computer and work on weekends too.

Q: Is it work, research or how do you define your work?

A: It is lot of research, trying things.

Q: How good is your chess?

A: I played for a club at about 2000 Elo.

Q: What's your advice for future chess programmers?

A: Don't give up your job. (laughs) I don't need a competition! A lot of amateurs gave up their jobs eventually. Some have been very successful. Like, Junior, Hiarcs, Chess Tiger, etc. These guys find it difficult generating income. If it is extra money it is okay. But, don't give up your job.

ARVIND AARON

Send this article to Friends by E-Mail

Section  : Sport
Previous : Ramesh scores facile win
Next     : Walsh, Ambrose breath fire and brimstone