Terminating the ChessGUI project.

[ppetrov]

Terminating the ChessGUI project.

Matthias, may I suggest that instead of terminating, you consider releasing its sources under e.g. the GPL license on some good free SCM host like Bitbucket or GitHub? There are good chances someone else will take over maintaining it and fixing bugs. It would be a shame for an already excellent project to simply die.

[alpha123]

Terminating the ChessGUI project.

There's nothing as bad as a buggy GUI.
I have not been able to account for time forfeits under Windows 64-Bit. Furthermore, ChessGUI has crashed 3 times in just a few days in Martin Thoresen's tournament. If I cannot figure out the cause of the crashes, it makes no sense to continue with the ChessGUI project.
If the project must end here, I shall update ChessGUI one last time.

Cheers,
Matthias.

If you must stop working on it, at least release the source.

Do you see the CB people giving up? (very buggy GUI)
Or the Arena people? (time forfeits)

Peter

[Matthias Gemuh]

Well, let me throw in a few thoughts here.

I have been extremely happy with how all the features of your GUI turned out after asking for them.

Matthias, you are a very kind person and I am sad to hear that you feel that you need to cancel the project.

Hi Martin,

you have actually helped a lot by highlighting a problem that cannot be ignored.

The idea to shutdown the project only occured to me because
1) the problem cannot be ignored
2) I doubt that I can solve a possibly Win64 issue using Win32
3) I am unlikely to upgrade to Win64 only to chase a bug

However, somebody has offered to help me trace the problem. He is a Visual Studio guy, with no Borland C++ Builder 5 experience, but if he still can help, then I should be able to keep the project alive.
Milos also gave a hint that can help.

As for logs, it's my fault that a program detects that it is crashing, and still does not save more hints before dying. Anyway, the log you have posted is quite helpful.

Cheers,
Matthias.

[Graham Banks]

Well - I don't have any problems with it for my CCRL 40/40 testing, so I'll keep using it regardless.
I'd be extremely disappointed to see you give up on ChessGUI, but I'll always be grateful to you for your efforts, and for being so quick to implement requested features.

Cheers,
Graham.

It's OK to continue using ChessGUI even for the CCRL if there is no obvious problem.

The only thing I don't like with the more recent versions is not being able to set the engine parameters for an unused engine immediately.
However, I overcome that hurdle by copying the ChessGUI.uci file from each engine folder of an earlier ChessGUI into the respective engine folders in the new version, and then renaming the ChessGUI.uci files to ChessGUI.cfg. Then I can enter the engine parameters immediately.

Cheers,
Graham.

[Matthias Gemuh]

The only thing I don't like with the more recent versions is not being able to set the engine parameters for an unused engine immediately.
However, I overcome that hurdle by copying the ChessGUI.uci file from each engine folder of an earlier ChessGUI into the respective engine folders in the new version, and then renaming the ChessGUI.uci files to ChessGUI.cfg. Then I can enter the engine parameters immediately.

Cheers,
Graham.

That procedure is OK. It is much easier to simply let all engines play test games till move 15 (set limit in GUI) in a 1-round Swiss at TC "Game in 30 secs", using default engine parameters. That would create all needed files in a few minutes.

But are you really talking about ChessGUI 0.197d ?

[Graham Banks]

The only thing I don't like with the more recent versions is not being able to set the engine parameters for an unused engine immediately.
However, I overcome that hurdle by copying the ChessGUI.uci file from each engine folder of an earlier ChessGUI into the respective engine folders in the new version, and then renaming the ChessGUI.uci files to ChessGUI.cfg. Then I can enter the engine parameters immediately.

Cheers,
Graham.

That procedure is OK. It is much easier to simply let all engines play test games till move 15 (set limit in GUI) in a 1-round Swiss at TC "Game in 30 secs", using default engine parameters. That would create all needed files in a few minutes.

But are you really talking about ChessGUI 0.197d ?

Pretty sure I need to that with both 0.196 and 0.197d. I'll check later.

[Matthias Gemuh]

Terminating the ChessGUI project.

Matthias, may I suggest that instead of terminating, you consider releasing its sources under e.g. the GPL license on some good free SCM host like Bitbucket or GitHub? There are good chances someone else will take over maintaining it and fixing bugs. It would be a shame for an already excellent project to simply die.

Hi Peter,

if I want to release the source code of ChessGUI, I would have to rewrite it first .
I started the project to write a program that plays Swiss tournaments only, without intending to release it (with or without code).
Almost every feature was carelessly added later. I have not respected any programming rules. The thing is too embarassing to release.

7 hours ago, I sent the source code to the guy I mentioned in this thread.
Since then, he's not able to speak yet.

Cheers,
Matthias.

[hgm]

Almost every feature was carelessly added later. I have not respected any programming rules.

Sounds a lot like WinBoard!

And that is open source...

[mhurd]

Hi Martin

Perhaps I am stating the obvious but I notice that you overclock your processor. Have you tried a tournement without the overclock to see if that is the cause of the problems. Overclocking can do strange things to some programs.

Mike

[Martin Thoresen]

Hi Martin

Perhaps I am stating the obvious but I notice that you overclock your processor. Have you tried a tournement without the overclock to see if that is the cause of the problems. Overclocking can do strange things to some programs.

Mike

A good suggestion, but this is not the issue as all other programs works week after week like they should, including chess programs like Arena and CB GUI.

And not to mention the fact that previous builds (aka pre 198) never crashed.