Author: Vasik Rajlich
Date: 03:52:07 02/23/05
Go up one level in this thread
On February 22, 2005 at 09:36:38, Peter Berger wrote: >On February 22, 2005 at 04:47:26, Vasik Rajlich wrote: > >>The part I don't get here is the "takes the strengths and weaknesses of the >>engine into account". Maybe if some amateur engine has a bug, sure - although in >>this case - fix the bug. >> >>I just don't buy this stuff about choosing lines which suit the engine. > >I used to mostly agree with this point of view, but my tests suggest different. >Maybe you just put too much content in this statement, because you compare with >human players' styles. > >As an example: Compared to top professionals Crafty has severe problems with >tactics and kingsafety, especially with kings castled in opposite directions. It >plays a fine endgame and there is a certain type of late middlegame where it can >shine. That's the kind of statement I am thinking about - there are some obvious >conclusions for an opening repertoire, no rocket science at all. > >So, if this is the starting point, you wouldn't choose the Sicilian Dragon as >main weapon against 1. e4, would you? An opening that might be fine for other >programs. > >Of course a future version of Crafty might rule in tactics and understand these >attacks just fine. Then you will make other choices. > >This effect is quite measurable. > Actually, this might just have to do with being weaker. All of the amateur engines can evaluate simple stuff like general piece centralization & mobility, keeping the pawn structure clean, etc. Evaluating king attacks is much tougher. Also, if you're getting outsearched by even the slightest bit, then a king attack position is a good way to get killed. It may well be that a weaker engine should avoid playing certain types of positions against a stronger engine. A funny opposite example came from the last WCCC, in Crafty-Shredder. Your 3. Bb5+ was not a bad idea, since it's hard for black to unbalance the position, but what was really funny was when Shredder played .. Qg4, basically getting on two knees for a draw. This is a rare example, but a book that understands the opponent's level may give some boost. (say - 10 ELO) >>Anyway - this too could be tested. For example, Junior has a reputation as being >>strong in dynamic positions. IMHO this is nonsense - Junior is just more >>aggressive in those positions, which is a completely different thing. Can we >>show that Junior really performs better than Shredder, for example in a match >>from a preselected set of dynamic positions? >> >>I highly doubt it. > >Same here - reason is that all top programs are very balanced and also similar >when it is about positions they like and dislike. I doubt that there is any >strong effect if you say just switch books between Shredder and Fritz ( the >Junior book has a few problems, so it might not be a good example). > >Most amateur programs are very similar too - usually strong in tactics, somewhat >weaker in positional play, without clear further characteristics. They will all >tend to mistreat the French, Pirc or the Ben-Oni and do better with the Sicilian >or Nimzo-Indian. They will also all do better in playing riskier openings >against stronger competitors and less risky lines against weaker ones. Only more >mature engines seem to develop a visible style which would mean some strength >that separates them from others - and then could profit from a book that tries >to bring these situations about. > Yes, but here two there are two further points: 1) Style != level. For example, Gothmog is aggressive, while Crafty is positional, so they'll treat some positions completely differently, but it doesn't follow from this that Gothmog will score better in sharper positions. If only playing sharp positions was as simple as playing actively and aggressively :) 2) Surely there is some difference in level, but how much? I have yet to see even one statistic that suggests such a difference for any two equally-matched engines. There are plenty of experiments that could show this, so it's at least suspicious that it hasn't happened yet. > >> >>IMHO - an amateur engine shouldn't bother with book until let's say four years >>go by. At that point, some professional chess player should be hired who will >>start from an automatically generated book and spend some month or so making >>some adjustments, preferably in the range of moves 5-15. In addition to this, it >>may make sense to keep running the engine and "pre-computing" some results, >>which can be spot-checked as appropriate. This last step will especially help at >>fast time controls. >> >>Vas >> > >I don't think this approach is optimal , but it will lead to a very reasonable >book. So what is the optimal approach? (Aside from full-time professional chess player spending 60 hours a week on the book? :)) Vas > >Peter
This page took 0.01 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.