Winboard Forum

When trying to compile it "UCHAR_MAX" from limits.h wasn't found here (Ubuntu). After inserting

#include <limits.h>

in util.cpp the problem was solved.

vp

Fixed in 1.4w24
I was informed about it but i waited with a new version to fix some more things.
Like a racing condition ,reported by Leo Dijksman.

http://www.geenvis.net/pg.html
Best
Fonzy

small update:
fixed: Illegal moves resulted in polyglot waiting for xboard to declare a timeloss
Polyglot now forfeits with the comment {Polyglot: Illegal engine move [White|Black] } , that should make it to the pgn file.
If Resign is used and the number of ResignMoves is reached,
Polyglot resigns immediate, instead of sending the last engine move together with a resignation.
(untested:) ics: a drawing move(reaching 50 moves rule,3fold,or drawing material) from the engine will be preceded with "offer draw"
previously just the move with the result(1/2-1/2) was send.

http://www.geenvis.net/pg.html
Best
Fonzy

If the makefile.gcc file is what should be used to compile with Cygwin, any chance of maybe with a future release getting the makefile to compile without requiring the dll file?

I suppose compiling with this in the commandline is suitable to acquire a functioning exe file?

g++ -O3 -mno-cygwin -o polyglot *.cpp

Thanks for the updates to app.

Charles Browne wrote:If the makefile.gcc file is what should be used to compile with Cygwin, any chance of maybe with a future release getting the makefile to compile without requiring the dll file?

I suppose compiling with this in the commandline is suitable to acquire a functioning exe file?

g++ -O3 -mno-cygwin -o polyglot *.cpp

Thanks for the updates to app.

Doesn't the generic makefile work for gcc?
I only kept in the gcc one for historic reasons(read: too lazy to remove it ).

There is a makefile in your sources that has no extension. If that is the one you mean by generic, I have put the gcc extension on the end of that file and it too compiles requiring a dll file.

Hey lazy is good, no real big rush, need for it here. But it would be helpful to have in a future release.

In the mean time it would be extremely helpful to know if this is suitable to compile your source files with.

g++ -O3 -mno-cygwin -o polyglot *.cpp

Charles Browne wrote:There is a makefile in your sources that has no extension. If that is the one you mean by generic, I have put the gcc extension on the end of that file and it too compiles requiring a dll file.

Hey lazy is good, no real big rush, need for it here. But it would be helpful to have in a future release.

In the mean time it would be extremely helpful to know if this is suitable to compile your source files with.

g++ -O3 -mno-cygwin -o polyglot *.cpp

As far as i know the makefile.gcc version is from the times polyglot (on runtime) needed the cygwin1.dll.
I never compiled polyglot with cygwin myself.
What dll file does cygwin require to compile polyglot?
Note that as polyglot does not need to run cygwin1.dll there is no real need to use cygwin to compile polyglot..

No biggy guy.

In the makefile of Winboard, past and present versions, which compiles with Cygwin is an option that forces Cygwin to compile Winboard with the mingw compiler so that a dll file is not required to be used with Winboard. That's all I was asking for, I can whip a makefile up for myself some day to use with Polyglot. Cheers.

Take it easy man.

Charles Browne wrote:No biggy guy.

In the makefile of Winboard, past and present versions, which compiles with Cygwin is an option that forces Cygwin to compile Winboard with the mingw compiler so that a dll file is not required to be used with Winboard. That's all I was asking for, I can whip a makefile up for myself some day to use with Polyglot. Cheers.

Take it easy man.

I am taking it easy,don't worry
Just curious about the problem.
Best
Fonzy

will be preceded with "offer draw"
previously just the move with the result(1/2-1/2) was send.

There is curious problem with forwarding draw claims to FICS. In a computer tournament games succeed each other very rapidly. Since FICS also adjudicates draws
the game will likely be finished by the time the draw claim arrives at FICS. In that case the draw claim from the engine may actually be interpreted as a draw offer in the next game!

My FICS bot icsdroneng does not forward draw claims anymore.

Michel wrote:

will be preceded with "offer draw"
previously just the move with the result(1/2-1/2) was send.

There is curious problem with forwarding draw claims to FICS. In a computer tournament games succeed each other very rapidly. Since FICS also adjudicates draws
the game will likely be finished by the time the draw claim arrives at FICS. In that case the draw claim from the engine may actually be interpreted as a draw offer in the next game!

My FICS bot icsdroneng does not forward draw claims anymore.

This is getting ugly,one should now seperatly cater for fics and ics?
note that it will only send the "offer draw" if winboard had send "ics" to polyglot in the first place.
would that happen with fics too?

fixed: Illegal moves resulted in polyglot waiting for xboard to declare a timeloss

In my version an illegal move from the engine caused PG to exit (cleanly).

I have now also changed it so that it will forfeit instead. This seems indeed like a good idea.

BTW. As I read the WB protocol I think the message from PG should contain "resign" since the game result is not coming from a fact on the board.

Why not simply relay the illegal move? WinBoard would then forfeit the engine, giving the proper reason in the PGN? By substituting "resign" you would hide the true reason for the user.

It would be even better if Polyglot would also play the illegal move on its internal board, so that when the GUI and opponent do not reject it, it could simply play on. That would make it, for instance, more Shatranj friendly. If the GUI can play with legality-testing off, Polyglot should not try to frustrate that. It is only meant as an adapter. Not to outsmart (= obstruct...) the GUI.

By substituting "resign" you would hide the true reason for the user.

No. The comment indicates why PG resigns.

Why not simply relay the illegal move?

PG does a lot of legality checking internally. This is what makes it such a stable program.
So simply relaying an illegal move would not work.

Michel wrote:PG does a lot of legality checking internally. This is what makes it such a stable program.
So simply relaying an illegal move would not work.

As a user I would consider that a _very_ bad feature. It means it outsteps the bounds of its mandate (translating the protcols), and tries to interfere with what is being transmitted. A capital offense for translators...

As a user I would consider that a _very_ bad feature. It means it outsteps the bounds of its mandate (translating the protcols), and tries to interfere with what is being transmitted. A capital offense for translators...

If you start critizising PG again then I'll stop working on it.

You can do the adaptions yourself.

H.G.Muller wrote:Why not simply relay the illegal move? WinBoard would then forfeit the engine, giving the proper reason in the PGN? By substituting "resign" you would hide the true reason for the user.

Well one could do it that way,or forfeit with {Polyglot: Illegal engine move [White|Black] }.
To me there seems to be no big difference.

It would be even better if Polyglot would also play the illegal move on its internal board, so that when the GUI and opponent do not reject it, it could simply play on. That would make it, for instance, more Shatranj friendly. If the GUI can play with legality-testing off, Polyglot should not try to frustrate that. It is only meant as an adapter. Not to outsmart (= obstruct...) the GUI.

Clearly,polyglot is restricted to chess now.I don't have any plans for Shatranj
I wonder if there will be many complaints about polyglot frustrating gui legality-testing features

Michel wrote:If you start critizising PG again then I'll stop working on it.

You can do the adaptions yourself.

You seem to have a severe attitude problem.

Very well, then! I will start my own fork. I can't work with people that feel they are above criticism, and I will certainly not be silenced by blackmail or threats...

You seem to have a severe attitude problem

Its obviously you that has an attitude problem.

If you feel that every discussion has to be sprinkled with derogatory terms and insults then I cannot work with you.