Freeware program for analysis?

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Rebel, chrisw

James Constance
Posts: 358
Joined: Wed Mar 08, 2006 8:36 pm
Location: UK

Re: Freeware program for analysis?

Post by James Constance »

Why not just use Fritz 5.32, as it is freeware now.
kaissa
Posts: 131
Joined: Fri Mar 10, 2006 11:21 pm

Re: Freeware program for analysis?

Post by kaissa »

royb wrote:
kaissa wrote:
Dan Astrachan wrote:
kaissa wrote: I am training a young player and need a database software with analysis feature. My first choice was Scid. Free, nice program but does not write to PGN files for some reason. Komodo can annotate games but sometimes send an error message and stops analyzing.
Hi

Scid (or even better, its fork on steroids Scid-Vs-PC) is perfectly capable of doing what you describe i.e. batch-analyze games from a PGN file with any UCI engine. You must be doing something wrong.

I can whip up a quick tutorial if you're interested.
I would love to have that tutorial. The analysis goes fine when suddenly I get this:

"This database is read-only and can't be altered.
This database is read-only and can't be altered.
while executing
"sc_game flag T [sc_game number] 1"
(procedure "markExercise" line 81)
invoked from within
"markExercise $prevscore $score"
(procedure "addAnnotation" line 155)
invoked from within
"addAnnotation"
(procedure "autoplay" line 27)
invoked from within
"autoplay"
("after" script)"

I simply do not understand why Scid or Scid-Vs-PC cannot write to PGN. I exported one annotated game in Scid format to PGN within Scid but the output was garbage.
It appears to be a feature of SCID. This is from the SCID tutorial (link follows):

You cannot edit a PGN file in Scid, as they are opened read-only and file maintenance functions are only available for Scid-format database files. So you will want to create one or more Scid-format databases and import games from PGN files into them.

The link for the SCID tutorial is here: http://scid.sourceforge.net/tutorial/

If you create a SCID database from the .pgn file, THEN that database is writeable.
This seems to be the problem. Why would an open source program insist on using its own file format instead of an already existing standard format puzzles me.

I have not had the chance to play with Winboard yet but at the moment I will use the Scid to have the engine analyze the game with Informant symbols and then input the analysis in Arena as if typing.

Thanks for your help,
kaissa
Posts: 131
Joined: Fri Mar 10, 2006 11:21 pm

Re: Freeware program for analysis?

Post by kaissa »

James Constance wrote:Why not just use Fritz 5.32, as it is freeware now.
The engine is weaker than Komodo or Stockfish. Although the games will not be master level games, having a world class engine analyze the game should provide me and the student points to go over.
User avatar
hgm
Posts: 27809
Joined: Fri Mar 10, 2006 10:06 am
Location: Amsterdam
Full name: H G Muller

Re: Freeware program for analysis?

Post by hgm »

kaissa wrote:This seems to be the problem. Why would an open source program insist on using its own file format instead of an already existing standard format puzzles me.
There is a good reason for that: speed. SAN moves are very cumbersome to parse, as you need to know where pieces currently are, and how they move to decide which of them is meant. In addition PGN is bulky because it is text, and writing down a move takes 4-5 characters (= bytes) when you include the space (and 1.5 more when you include the redundant move numbers). In the SCID database format a move only takes a single byte, and is explicit in what piece it moves, so that it would not have to be derived from rule knowledge.

All this is pretty unimportant when you are dealing with a few hundred games, but SCID is designed to handle databases with millions of games.

WinBoard is not primarily a database program, and operates only on PGN. But to make fast searching of positions possible, it does create a compact format in memory. To load a PGN with 40,000 games therefore already takes 15 sec, while for the internal format searching through it takes only about a second.
phenri
Posts: 284
Joined: Tue Aug 13, 2013 9:44 am

Re: Freeware program for analysis?

Post by phenri »

kaissa wrote:
James Constance wrote:Why not just use Fritz 5.32, as it is freeware now.
The engine is weaker than Komodo or Stockfish. Although the games will not be master level games, having a world class engine analyze the game should provide me and the student points to go over.
Correct me if I'm wrong, I'll add that Fritz 5.32 does not support the UCI protocol, and it is only from Fritz 7 GUI plus an service pack that the UCI is supported.
James Constance
Posts: 358
Joined: Wed Mar 08, 2006 8:36 pm
Location: UK

Re: Freeware program for analysis?

Post by James Constance »

kaissa wrote:
James Constance wrote:Why not just use Fritz 5.32, as it is freeware now.
The engine is weaker than Komodo or Stockfish. Although the games will not be master level games, having a world class engine analyze the game should provide me and the student points to go over.
Fritz 5.32 will play at grandmaster level. For games between most players that will show the mistakes. I'd be surprised if having the latest engines would help teaching.

It's funny how when the latest engine comes out the earlier versions go down in our estimation! :)