Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Go programming

Author: Roy Eassa

Date: 09:15:41 02/13/04

Go up one level in this thread


On February 13, 2004 at 09:56:55, Janosch Zwerensky wrote:

>
>>That's the problem.  Each is "Go"ing it alone.  Nobody has tried to create a
>>single program with ALL the GUI features, but with a plug-in architecture for
>>playing engines.  There are maybe 5 top programs, each with a few dozen (or
>>fewer) features and no plug-in engine interface.  Compare that to ChessBase,
>>which has hundreds & hundreds of features.
>
>Analysis features for Go programs won't sell as long as there are no programs
>which actually have a decent understanding of the game. As far as keeping
>records of one's own games is concerned, a tool like CGoban2 will suffice
>completely, and database applications will probably be less useful to go-players
>than to chess players, because rote memorization of tactical sequences (tesujis)
>and strategic exchange sequences (joseki) or opening lines (fuseki) has a
>significantly lower pay-off in go than in chess even if you are not high-rated.
>
>Regards,
>Janosch


The features that would really sell might not be IDENTICAL to those of a good
chess tool like ChessBase, but I'd bet there are OTHER features that would still
be considered invaluable.  We might not know what they are yet, but a peek 20
years into the future might make us say, "Duh, why didn't I think of that?".
I.e., somebody will grab the lion's share of the market at some point by having
the most and best features; just because we don't currently know what those
features are doens't mean it won't happen.  (Just my humble opinion.)

Have you looked at SmartGo?  I am a beta tester for that program.  It might be
in the lead to someday be the ChessBase of the Go world.  Its playing engine is
quite slow and weak, but it has the most annotation and analysis features
available today, I think.  But they're still a tiny drop in the ocean compared
to what ChessBase offers.

I think the author of SmartGo is crippling himself by tying the Go "brains" of
the program so closely to the GUI, and by working on both himself.  I think he
should decouple those parts as much as possible and convince the authors of
Go++, Many Faces, etc., to adapt their engines to his GUI (just a Chess Tiger
and Shredder became available for the ChessBase GUI, which benefits both their
authors and ChessBase financially).  I think he should focus on the features and
GUI and create an atmosphere where thousands of people worldwide would be
encouraged to create playing engines and compete against each other -- perhaps
annual rewards for the strongest engine?  The goal would be to take himself OUT
of the engine-writing role (where others have already proven themselves more
talented) and open that up to the world, with incentive, and to focus on
building a near-monopoly for the ENVIRONMENT like Microsoft or ChessBase.

  -Roy.



This page took 0 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.