"Full-width means never having to say you're sorry"

[sje]

The above quote comes from one the the Spracklens (Kathe, I think). It was from sometime back in the early 1980s.

[mongrel]

It's funny that recently I am also trying to unbalance my program in order to get better results (and the playing style more pleasant to look at) - the only difference being that it is done at the level of about 2000-2100 Elo, and not at the top. Thank You for pointing a historical example that it actually might work to some extent.

General direction is to emphasize king attack for both sides and mobility for the opponent (so that program attacks and simultaneously avoids positions where opponent has attack potential)

My problem with this approach is that the reduction techniques which were sort of OK for a more balanced program don't seem to work any more. Is it a common experience?

regards from an avid reader of Thorsten's posts

Thank you for replying.
I'm glad to hear about your creative efforts.
Since MChess has been mentioned, I can remind us of my original post:
I clearly remember preferring certain programs at that time, among them MChess, because it played with this (apparent) personality, and despite the fact that it fell behind in the race for ratings. I understand that the thesis of the original post is difficult for a commercial programmer, and also for others, focused on the ratings.
Admiration, and a special appreciation, and gratitude, has been expressed in this thread toward MChess, and others (I don't want to leave anybody out. Since Cristophe Theron has posted, everyone would agree that I can add his programs to the list.)

That must be worth something.
(Briefly restating my thesis: These qualities, not necessarily rewarded by the rating system, are becoming more important as the top programs become inhumanly strong, almost superfluously strong. I claim that there is a market for someone... That is the thesis.)

PK: Did you read my posts on page one? And the other ones in the threads "New Paradigm Desired", and "Reasons of the Crisis"?
The original post has been sidestepped a little; I admit some provocative language, because I had that thread "Reasons of the Crisis" strongly in mind.
No need to repeat myself here.

Best wishes to all.

[PK]

You are right that I wrote just a comment to a comment. As far as the general issues are concerned:

1) yes, nowadays it's definately worthwile to go for style rather than for rating. even a marketing specialist should be able to notice it, judging from the popularity of user-defined ChessMaster, Fruit or even Glaurung settings.

2) no, full width is a non-option (unless a genius emerges who will design a replacement to alpha-beta)

3) ...but, returning to my previous post, there might be a serious problem with standard pruning techniques if a programmer goes for style by the means of doing something strange in the evaluation function. IMHO this is especially true if You want Your engine to find unusual moves. The price for it is that history-based move ordering, late move reductions and so on become ineffective in the very moment when the game becomes fun to watch.

4) so, whereas returning to full width is way too radical, there are cases when one has to rethink even the established pruning techniques

regards

Pawel Koziol