Winboard Forum

hello all,

sorry I need some help.
Michael Sherwin's Romi 2a/b does not work properly for me in Arena 1.1
(and pls dont send me to the Arena Forum...).
The Version 1 worked fine and was a very good newcomer, really nice.
Version 2 does start ok, but whenever it comes to making the first move
I get an exception.
Anybody else having troubles with it? Or is it a problem located on my system (or in front of the screen )?

Peter

Peter Eizenhammer wrote:hello all,

sorry I need some help.
Michael Sherwin's Romi 2a/b does not work properly for me in Arena 1.1
(and pls dont send me to the Arena Forum...).
The Version 1 worked fine and was a very good newcomer, really nice.
Version 2 does start ok, but whenever it comes to making the first move
I get an exception.
Anybody else having troubles with it? Or is it a problem located on my system (or in front of the screen )?

Peter

Hello Peter,

This engine sends a lot of output after the move. I wonder if that is the source of the problem.

The engine does not work in Winboard which tells me that it has an issue somewhere. Wait for the updated engine.

The problem is not with you.



And why not the Arena Forum?



Ignore that last.

Later.

Roger Brown wrote:

Peter Eizenhammer wrote:hello all,

sorry I need some help.
Michael Sherwin's Romi 2a/b does not work properly for me in Arena 1.1
(and pls dont send me to the Arena Forum...).
The Version 1 worked fine and was a very good newcomer, really nice.
Version 2 does start ok, but whenever it comes to making the first move
I get an exception.
Anybody else having troubles with it? Or is it a problem located on my system (or in front of the screen )?

Peter

Hello Peter,

This engine sends a lot of output after the move. I wonder if that is the source of the problem.

The engine does not work in Winboard which tells me that it has an issue somewhere. Wait for the updated engine.

The problem is not with you.



And why not the Arena Forum?



Ignore that last.

Later.

RomiChess Prot 2a have a "loop" or so, but Proto 2b runs here without any problem !

Ofcourse, both of you told what your OS is, CPU type and posted a part of the log/debug file so it is easy to help to find the problem.

Best wishes,
Leo.

Thx Roger and Leo for your answers.

I should of course have given more info, so:
AthlonXP 2000+ (1,67GHz), WinXP Pro SP2, Arena 1.1.

I tried to add xboard, but it seems to change nothing.

Here the output according to Arena Debug window (in German "Fehlersuche"):

739750>1:?
739765>1:force
739765>1:quit
740359*1*-------------------------Starte Motor 1 RomiProto2b-------------------------
740359*1*Configured Engine 1 Type: WB2
740500*1*Engine 1 dir: K:\tempie\romichess_P2b
740562*1*Engine 1 commandline: K:\tempie\romichess_P2b\ProtoRomi2b.exe
740890>1:xboard
740890>1:protover 2
741656<1:This is RomiChess by Michael J Sherwin
741656<1:Version Proto Romi 2b -- 6/19/2005
741656<1:Copyright 2005 Michael J Sherwin
741656<1:feature analyze=1
741656>1:accepted analyze
741656<1:feature done=1
741656>1:accepted done
742093>1:new
742093>1:random
742093>1:level 0 1 1
742187<1:Error (unknown command): accepted
742187<1:Error (unknown command): accepted
742187>1:post
742187>1:hard
742187>1:easy
742218<1:Error (unknown command): random
753890**-------------NewGame!!!-------------
753890>1:level 0 1 1
753890>1:new
753890>1:level 0 1 1
753921<1:Error (unknown command): post
753953<1:Error (unknown command): hard
753984<1:Error (unknown command): easy
761031>1:computer
761031*1*Start calc, move no: 0
761234>1:level 0 1 1
761234>1:new
761234>1:level 0 1 1
761312<1:Error (unknown command): computer
761468>1:time 6000
761468>1:otim 6000
761468>1:white
761468>1:go

Well, and here there is the nice "unknow software exception" window, allways a pleasure to have...

For someone with more knowledge about the expected/allowed commands
it should be possible to figure out whats going on, I hope.

Many thanks for your help,

regards
Peter

Leo Dijksman wrote:
RomiChess Prot 2a have a "loop" or so, but Proto 2b runs here without any problem !

Ofcourse, both of you told what your OS is, CPU type and posted a part of the log/debug file so it is easy to help to find the problem.

Best wishes,
Leo.

Hello Leo,

Guilty as charged.



I am not at home so I cannot send the debug but the other data is Windows XP, Pentium III, 800+ mhz and 512 mb of ram. Winboard 4.27.

The program exits with an error message.

Posting about a bug report without details is annoying, a waste of time and not helpful.

I apologise for doing it.

Later.

Hello all!

I have found a bug in RomiChess - it doesn't work properly in time control "time per game", works ok in "time for x moves".

Here's the reason:

Code: Select all

File io.c
Line 204:
      sscanf(data, "level %d %d %d", &numMoves, &totTime, &timeInc);

Line 307:
      maxTime = myTime / (numMoves >> 1);


In "time per game" mode first arument of level command is 0, and there is division by 0.

Hello Piotr,

Piotr Cichy wrote:Hello all!

I have found a bug in RomiChess - it doesn't work properly in time control "time per game", works ok in "time for x moves".
Here's the reason:

Code: Select all

File io.c
Line 204:
      sscanf(data, "level %d %d %d", &numMoves, &totTime, &timeInc);

Line 307:
      maxTime = myTime / (numMoves >> 1);


In "time per game" mode first arument of level command is 0, and there is division by 0.

you are the man!
(but silly me and Roger, we really should have seen this at once, tsts... )
Thanks a lot.

Peter

Peter Eizenhammer wrote:Hello Piotr,

Piotr Cichy wrote:Hello all!

I have found a bug in RomiChess - it doesn't work properly in time control "time per game", works ok in "time for x moves".
Here's the reason:

Code: Select all

File io.c
Line 204:
      sscanf(data, "level %d %d %d", &numMoves, &totTime, &timeInc);

Line 307:
      maxTime = myTime / (numMoves >> 1);


In "time per game" mode first arument of level command is 0, and there is division by 0.

you are the man!
(but silly me and Roger, we really should have seen this at once, tsts... )
Thanks a lot.

Peter

Hello Peter,


Me?


See or understand code?


Uh.......





That is the sum total of my intellectual capacities in this regard.


Later.

Just to let you know, it seems as Michael have found the problem so an 2c version will be available soon!

Thanks for the bugreports,

Leo

Leo Dijksman wrote:Just to let you know, it seems as Michael have found the problem so an 2c version will be available soon!

Thanks for the bugreports,

Leo

When there is still a problem please send a log/debug file to the author!
His email is in the readme.

Best wishes,
Leo.