bug-gnubg
[Top][All Lists]
Advanced

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

Re: [Bug-gnubg] GNU notes


From: Jim Segrave
Subject: Re: [Bug-gnubg] GNU notes
Date: Mon, 16 Sep 2002 19:58:32 +0200
User-agent: Mutt/1.4i

On Mon 16 Sep 2002 (10:33 -0300), Albert Silver wrote:
> 
> 
> > I see Joern's already replied, just adding my comments.
> > 
> > > 1)                   So I'll start with the bug. I set up a position
> and
> > > saved it, yet every single time I load it, it shows up with 17 (!)
> > > pieces with an extra one on the bar and another taken off. If I
> enter
> > > Edit and click on either piece they disappear, but it still shows
> when I
> > > load it. I can zip this, but should I send it to this mailing list
> or
> > > someone specific?
> > 
> > I'd say zip it and send it to the mailing list. I'd certainly like to
> > see it.
> 
> Ok, I included it here. It is called Bug.zip. It is one of the positions
> taken from Kit Woolsey's article "When God is Wrong". In the article,
> Woolsey shared 21 positions where Snowie 3-ply (its highest setting)
> told him he was badly mistaken, but where a rollout vindicated his
> decision. If you have the article or know it, be warned that the 13th
> position is wrong and Snowie did not get it wrong (nor GNU). FYI, GNU
> got some 8-9 correct. 

I take it it is the first position in this article? In which case,
there's a minor fault here - you have a Bottom piece on the 13 point,
it should be on the 15 point. Which doesn't alter the fact that if you
set up the position and match ID:
Position ID: HGccCEzZtgGCIQ
Match ID: QQnqAAAAAAAA

    GNU Backgammon  Position ID: HGccCEzZtgGCIQ
                    Match ID   : QQnqAAAAAAAA
    +24-23-22-21-20-19------18-17-16-15-14-13-+  O: Top (Cube: 2)
    | X     O     X  O | O |    O     X  O    |  0 points
    |       O     X  O |   |    O        O    |  
    |       O        O |   |             O    |  
    |                  |   |                  |  
    |                  |   |                  |  
    |                  |BAR|                  |v 7 point match
    |                  |   |                  |  
    |                  |   |                  |  
    |                  |   |                  |  
    |    O  X  X  X  X |   | X                |  Rolled 42
    | X  O  X  X  X  X |   | X  O             |  0 points
    +-1--2--3--4--5--6-------7--8--9-10-11-12-+  X: Bottom

(Bottom) 


Then save the match out, when you try to reload, two extra pieces
appear, one bottom piece off, one extra top piece on the bar. The
position and match IDs shown are:

Position ID: HGccCMyybQMEQw
Match ID: QQnrAAAAAAAA

Then (in graphics mode), saving and loading the save file gives:

    GNU Backgammon  Position ID: HGccCMyybQMEQw
                    Match ID   : QQnrAAAAAAAA
    +24-23-22-21-20-19------18-17-16-15-14-13-+  O: Top (Cube: 2)
    | X     O     X  O | O |    O     X  O    |  0 points
    |       O     X  O | O |    O        O    |  
    |       O        O |   |             O    |  
    |                  |   |                  |  
    |                  |   |                  |  
    |                  |BAR|                  |v 7 point match
    |                  |   |                  |  
    |                  |   |                  |  
    |                  |   |                  |  
    |    O  X  X  X  X |   | X                |  Rolled 42
    | X  O  X  X  X  X |   | X  O             |  0 points
    +-1--2--3--4--5--6-------7--8--9-10-11-12-+  X: Bottom

(Bottom) 

With Bottom having one chequer off.

If I try this in text mode, loading the same file gives the above
without showing a chequer off.


Trying to set that position Id directly flags an illegal postion and
core dumps in text mode, in graphics mode, the position ID is ignored.
Actually, I can't get repeatable core dumps - sometimss it just says
it's an illegal position, sometimes it core dumps with a backtrace
which makes no sense.

-- 
Jim Segrave           address@hidden




reply via email to

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