Tuesday, July 17, 2012

Bellyaching about Engines


        I saw it again.  On one of the forums there was a discussion about correspondence play, specifically playing for the ICCF which allows engine use.  It was mentioned by a couple of people that engine use was cheating and there was no point in playing engine assisted games because it would be very boring anyway.
        First of all if a site allows the use of engines for analysis it is not cheating any more than the use of opening books and databases is cheating on sites that say you can consult them.  I don’t understand the thought process that says even if something is within the organization’s guidelines and you do it, it’s still cheating.  I even heard one guy say that he considers using books in correspondence chess cheating. How can that be?!  Anyway, he probably cheats, too. When he plays correspondence chess, I bet he uses a lot more time than OTB players and I’d go even further and say he touches the pieces and moves them around which is cheating in over the board chess. Hey!  If it’s cheating in tournament chess, it’s got to be cheating in correspondence, right?  What they don’t realize is that OTB chess and correspondence chess are not the same thing. 
      Maybe they don’t want to use engines in their games, and that’s OK.  If that’s the way they feel then I have no problem with it and the solution for them is simple…don’t play on ICCF or any site that allows engine use…problem solved.
      I don’t like chess problems, but I know one guy who does; in fact he’s a Master Solver with the US problem solving organization.  If that’s what he likes to do then I don’t have a problem with it and I’d never tell him it’s a waste of time to mess around with positions that will never come up in real play.  They also have problems where pieces move in ways they don’t move in regular chess.  So what?  That’s what those guys enjoy. I don’t, so I don’t get involved, but neither do I belittle an aspect of the game they like.
      Some time back the ICCF had an interview with an OTB GM who was just getting into CC.  He said he did it because he was not interested only in the competitive nature of the game, but appreciated the search for hidden and deep ideas and thought he could be more creative in correspondence games without the tension found in OTB.
       I found it interesting that he mentioned a point I’ve made here is some posts…most engines do not understand positions with a material imbalance. He also commented that even with computers, one can never be lazy, adding that sometimes the search for the best move can take days and even then there is no guarantee you will find the proper answer.
       In his correspondence games he uses all engines available. The search for the best move, which is what he is trying to find, should, in his opinion, meet no boundaries. He pointed out that blindly following engine suggestions will usually lead to disaster and that in the end the best players will win anyway
       On ICCF where deep engine analysis drives most of your opponents' moves, it becomes paramount to be well-prepared in the opening. His procedure is to create a database of all ICCF games and then extract the first 40 moves into a new opening book. This gives him an opening overview which he can use to maximize his chances against the best engines. He also added that he has found several examples of top GMs building their opening lines based on ICCF games.
       He believes preparation is just as important (if not more so) than having an engine churning out moves. The main aspect is opening preparation. Without solid preparation, you can lose the game before your engine can save it.  Simply using the moves an engine comes up with is usually not enough to win; you have to use everything at your disposal, including human intuition.
       If a player is what is known if the CC world as a “postman” and just delivers what the engine says is the best move after a minute’s thought, then yes, it would be boring and there wouldn’t be much point in playing. But, how can a 1500 second guess an engine?  They probably can’t but what they can do is do a lot of research, compile opening books and databases of top level games that will allow them to get good positions for the engines to work on.
       Then using different engines they will often find there are different alternatives and different evaluations of the same position.  Also in some positions engines are fickle and will keep changing their mind on the evaluation. It is then up to the player to decide which move is best. What does a 1500 do in these situations?  He keeps on analyzing and researching, running shootouts and engine tournaments from the position in question until he thinks he’s found the best move then he plays it.  What if it turns out it isn’t the best move and his judgment was wrong?  He loses just like he would in any OTB game.
      Is playing chess this way enjoyable?  Depends on the individual.  If it isn’t, play on sites that disallow engines. Of course if you’re playing on a site where they are prohibited and you run into an engine user then you have a right to complain.  Otherwise keep your cybermouth shut. Thank you.

No comments:

Post a Comment