Re: [Bug-gnubg] is this a bug?

[Joern Thyssen]

On Sun, Aug 31, 2003 at 10:31:30AM -0700, Hugh Sconyers wrote

> position id = tm0AwGKNty0CAA
> match id = cAkAAAAAAAAA
> 
> Cube analysis
> 3-ply cubeless equity  +1.133
>   0.783 0.574 0.031 - 0.217 0.037 0.001
> Cubeful equities:
> 1. Double, pass         +1.000
> 2. Double, take         +2.122  ( +1.122)
> 3. No double            +0.892  ( -0.108)
> Proper cube action: Double, pass
> 
> ==========================
> 
> If the cubeless equity is > 1 and you own the cube this should be "too
> good to double"

The cube is centered according to the match ID and with Jacoby rule it
can never be too good.

Note that gnubg doesn't use the cubeless equity when determining whether
it too good; it uses the "no double" equity which in this case is well
below 1.

Jørn

pgpqgc5q9R04W.pgp
Description: PGP signature