bug-gnubg
[Top][All Lists]
Advanced

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

Re: [Bug-gnubg] backgammon position evaluation bug?


From: Christian Anthon
Subject: Re: [Bug-gnubg] backgammon position evaluation bug?
Date: Sat, 21 Oct 2006 08:49:29 +0200

The evaluation function operates on the position prior to the roll
where red wins 34/36 games or 94.4%. The hint function is what you
where looking for.

Christian.

On 10/20/06, Karl Berry <address@hidden> wrote:
We received this report at address@hidden  I guess it's for you ...

Best,
Karl



From:   Freddy Galliers-Pratt <address@hidden>

Dear Gnu

Can you please advise me on something I have just noticed.

For game position:

CQAAcAAAAAAAAA

Match ID( UQnvAAAAAAAA)

Black has rolled 6&3 and bears 2/off [2]

Red then rolls 2&1 to go to game position:

AgAAJAAAAAAAAA

Match ID( EQHlAAAAAAAA)

And mathematically cannot win, but bears 1/off.

However if game equities are now checked, he is massive favourite to
win, .944

Position ID: AgAAJAAAAAAAAA
Match ID: EQHlAAAAAAAA
Evaluator: BEAROFF2
Player Opponent
Position 9 2
Cubeless equity : +0.8889
Owned cube : +0.8889
Centered cube : +0.8889
Opponent owns cube : +0.8889
Player Opponent
Position 9 2
Bearing off Bearing at least one chequer off
Rolls Player Opponent Player Opponent
0 0.000 0.000 100.000 100.000
1 94.444 100.000 0.000 0.000
2 5.556 0.000 0.000 0.000
Average rolls
Bearing off Saving gammon
Player Opponent Player Opponent
Mean 1.056 1.000 0.000 0.000
Std dev 0.229 0.000 0.000 0.000
Effective pip count:
Player Opponent
EPC 8.620 8.167
Wastage 4.620 6.167
EPC = 8.167 * Average rolls
Wastage = EPC - pips

Win W(g) W(bg) L(g) L(bg) MWC Cubeful
static: 0.944 0.000 0.000 0.000 0.000 60.76% 60.76%


Yet with only a 2 left and a chequer on the 3, it is impossible.

I am reading your equities wrong, or is this a bug?

Really appreciate your thoughts.

Best regards

Freddy Galliers


_______________________________________________
Bug-gnubg mailing list
address@hidden
http://lists.gnu.org/mailman/listinfo/bug-gnubg





reply via email to

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