bug-gnubg
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Bug-gnubg] Problems with Gnubg 04-Mar-2009 release


From: Joaquín Koifman
Subject: Re: [Bug-gnubg] Problems with Gnubg 04-Mar-2009 release
Date: Wed, 18 Mar 2009 20:55:20 -0300

Efe,

Correct, that only happens when the cube has already been turned. The tutor doesn't inform about the mistake nor mark the move with yellow.

Joaquin


2009/3/18 <address@hidden>
Dear Christian, Joaquín, Max,
 
I just stumbled again upon a faulty "Too Good To Double" tutor error during one of my matches against Gnubg.
 
Attached is the match file for your observation - I occured during Game 5. I knew it was too good to double anyway, but tried it just to see whether the tutor was going to stop me against committing this mistake. According to the anaysis pane, it was a blunder of 0.764 points magnitude marked with a big yellow!!! Obviously, Gnubg dropped it...
 
I guess you can open the same file with your own Gnubg and then continue Game 5 from the Redouble move onwards. Please inform me how the match continues for you? Does the tutor warn you against the move or does it just go ahead with your decision just to let Gnubg pass the cube?
 
I am now wondering with suspect, whether the tutor is not functioning correctly for "Too Good To Double" when the cube is already turned??? Anyone wanna comment on this??
 
Kind Regards,
 
Efe
 
 
 


--- On Sat, 14/3/09, address@hidden <address@hidden> wrote:

From: address@hidden <address@hiddenco.uk>

Subject: Re: [Bug-gnubg] Problems with Gnubg 04-Mar-2009 release
To: "Christian Anthon" <address@hidden>, "Joaquín Koifman" <address@hidden>, "Massimiliano Maini" <address@hidden>
Cc: address@hidden
Date: Saturday, 14 March, 2009, 11:36 AM

Dear Christian, Joaquín and all,
 
I have some further diagnosis and observations since yesterday that I would like to share with you upon Joaquín's request:
 
8) I realize now that my older version (Sep-08) is analyzing the match "on-the-go" [player0, analysis and evaluation settings in both versions all set as checker play="supremo", cube play= "worldclass", tutor decision="same as evaluation"]. What I mean with this is that when I for example make a very lucky, very unlucky, doubtful, bad, very bad move etc, I immediately see its effects (bold, italics, colored etc.) in the "Game Record" panel after I confirm my move by clicking on the dice.
 
On the other hand, the newer version (Mar-09) does not do this "on-the-go" analysis right after the move. All comments (bold, italics, bad, very bad, lucky, unlucky etc.) only appear in the "Game Record" panel, if and only if you perform a full analysis of the match afterwards. In addition, the missing "delta equity" for each move with respect to the average dice roll in the parenthesis also appears surprisingly after the analysis!
 
This is basically the reason why it is taking so short for the older version to analyze the match (since it is just analyzing gnubg's moves - half the work!), whereas the newer version starts everything from scratch. I double-checked the settings for everything and can assure you that they are the same. Does this mean anything to you? Do you encounter the same problem? Is it in anyway related to an already addressed error?
 
A) After Joaquín's request, I tried to artificially create some "Too Good To Double" positions by using the edit mode. He is right - I could not see the problem at least in these examples anymore with Mar-09 version - maybe this problem also depended on the match scores when I encountered them in the past?
 
Therefore, I assume for now that the Tutor works fine with the newer version regarding "Too Good To Double" decisions. If I ever encounter a similar error in the future during any of my matches, I will same the game and forward it as a concrete evidence for your inspection.
 
C) As I wrote in my original mail, those cases were definitely wins or losses for Gnubg and its equity was already probably -1.000 or +1.000. So Joaquín's argument makes sense. Since all possible moves are having the same equity value at that instant, Gnubg cannot differentiate between them and probably picks "any random" one amongst them. That's the only reasonable explanation for it.
 
Kind Regards & Nice Weekend to all!
 
Efe


--- On Fri, 13/3/09, Joaquín Koifman <address@hidden> wrote:

From: Joaquín Koifman <address@hidden>

Subject: Re: [Bug-gnubg] Problems with Gnubg 04-Mar-2009 release
To: "Christian Anthon" <address@hidden>
Cc: bug-gnubg-bounces+massimiliano.maini=amadeus.com@gnu.org, "Massimiliano Maini" <address@hidden>, address@hidden
Date: Friday, 13 March, 2009, 2:24 PM

Hi all,

A couple of comments about some issues:

1) It's still not solved in the 20080313 version.

8) I think that what happened was that the match was already analyzed. 45 seconds is normal for an analysis while 5 seconds is very, very fast.. Could you try to reproduce it again, Efe?

A) I couldn't reproduce neither in the 20080313 version nor in the 20080304 one.

C) Probably gnubg had already won the match or there was no posibility that it could save the gammon, in the examples you gave. Thus, the program made any move because the equities would be all the same. Could you check this also?

Joaquin




reply via email to

[Prev in Thread] Current Thread [Next in Thread]