Re: [Bug-gnubg] Errors in evaluation of races with backgammons

[Joseph Heled]

Not sure how to set up gnubg to evaluate higher plies. This is after
manually editing the position

Position ID:    /wgAWBz/AAAAAA
Match ID:       cAkgAAAAAAAE

Evaluator:      Race


        Win     W(g)    W(bg)   L(g)    L(bg)   MWC       Cubeful
static: 1.000   1.000   0.678   0.000   0.000   100.00%   100.00%
 1 ply: 1.000   1.000   0.678   0.000   0.000   100.00%   100.00%
 2 ply: 1.000   1.000   0.799   0.000   0.000   100.00%   100.00%

On 27 February 2012 19:33, Joseph Heled <address@hidden> wrote:
> On 27 February 2012 19:09, Philippe Michel <address@hidden> wrote:
>> On Mon, 27 Feb 2012, Joseph Heled wrote:
>>
>>> clearing checkers from the back is not the same as bearoff - you *can*
>>> move any checker you want.
>>
>>
>> Embarrassing. You're right of course and in my example 3s and 4s can
>> certainly be played in a different way than in a bearoff.
>
> An easy mistake to make. Still ....
>
> Using some python trickery it is possible to get the true backgammon
> figure. "true" here means that you always make the optimal move which
> minimises the probability of backgammon.  The number is 40615026 /
> (36^5) = 0.6716982730973429
>
> Now, nngnu gives those numbers for plies 0 to 5
>
>  0ply  0.6775242686271667
>  1      0.6775177717208862
>  2      0.6775177717208862
>  3      0.6775176525115967
>  4      0.6717002391815186
>  5      0.6716980934143066
>
> which I am happy about, but it seems that *there is* some problem in
> gnubg, which gives different numbers for different plies.
>
> I think someone should take a good look at this!
>
> -Joseph