Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: SAN v Coordinate v Long SAN

Author: Dan Honeycutt

Date: 09:20:47 06/13/04

Go up one level in this thread


On June 13, 2004 at 10:38:31, Robert Hyatt wrote:

>On June 11, 2004 at 23:43:42, Dan Honeycutt wrote:
>
>>On June 11, 2004 at 09:58:51, Anthony Cozzie wrote:
>>
>>>On June 11, 2004 at 08:33:51, Fabien Letouzey wrote:
>>>
>>>>On June 10, 2004 at 21:01:00, Anthony Cozzie wrote:
>>>>
>>>>>What it comes down to is this: is there _any_ point where coordinate is better
>>>>>than Long SAN?
>>>>
>>>>>anthony
>>>>
>>>>Yes, it can be produced with fewer information ("which piece is moving?" and "is
>>>>it a capture?" are not needed).
>>>>
>>>>In my software I don't need to follow the moves on a board to generate
>>>>coordinate-notation output, but I would with long SAN.
>>>>
>>>>---
>>>>
>>>>I am not saying I am against long SAN but there is a disadvantage, at least in
>>>>my programs.
>>>>
>>>>Fabien.
>>>
>>>I admit I hadn't thought of that; in Zappa I have piecetype & captype as part of
>>>the move.
>>>
>>>However, the beauty of the scheme is that if you don't want to do san you simply
>>>insert a blank character.
>>
>>...or nothing if the piecetype is optional.  You can put the piecetype
>>everywhere in whatever language you desire, with piece moves but not pawn moves
>>(my personal favorite), or nowhere.  All perfectly compliant with the standard
>>which ignores anyway if it is present.  The additional effort to parse is
>>trivial and you can make it more readable if you desire.  ditto with the
>>separator, "-", "x" or nothing - all ignored by the translator.
>>
>>Dan H.
>>
>>
>
>I like SAN because of its "size".  IE one goal has to be to avoid storing stuff
>that is not needed.  IE "-" in the above (Ng1-f3).  If we don't stick with SAN,
>which is the shortest representation (who cares if it is a bit harder to parse,
>so are compressed EGTB files harder to read but the space savings is worth it)
>then simple coordinate would be my next choice (e2e4)...
>
>

Allowing the piece moving and separator was an effort to bring harmony between
the "readability" and "not meant to be read" camps, but it does add size.

Another issue that has only been touched on - anything other than SAN is going
to mandate that the converter have a move generator.  Maybe SAN is the way to
go.  I can't make up my mind, best I just shut up.

Dan H.

>
>
>
>>  All moves are Xa1-a2 Xe3-e4 etc.
>>>
>>>anthony



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.