[Bug-gnubg] RE: Monte Carlo

[Albert Silver]

You're refering to the Double at the end of game 5. I went and took a
look and I'm even more mystified than you are. You see, not only can I
not reproduce my result, I can't even reproduce YOUR results. At all.
I'm using build 030804.

Here's what I'm getting (*sigh*):

In the file -->

Cube analysis
2-ply cubeless equity  +0.657 (Money:  +0.653)
   74.0%  18.9%   1.0% -  26.0%   2.6%   0.0%
Cubeful equities:
1. Double, pass         +1.000
2. Double, take         +1.033  ( +0.033)
3. No double            +0.847  ( -0.153)
Proper cube action: Double, pass
--------------------------------------------

Here I think there may be some bug, possibly some residue from the
cache. I find that switching back and forth between METs, sometimes
running a 3-ply, and then switching, etc, leads to discrepancies in the
results reported at 2-ply. 

I have seen some really big changes, but am not sure the procedure I
used to produce them. 

                                                Albert



> -----Original Message-----
> From: Joseph Heled [mailto:address@hidden
> Sent: Tuesday, August 12, 2003 11:48 PM
> To: Albert Silver
> Cc: Joern Thyssen; gnubg
> Subject: Re: Monte Carlo
> 
> Still, there is a problem.
> Can someone else tell me if he gets the same results as me or not?
> 
> 
> (cLfBATA2NycGAA / MAEgA0AACAAA)
> 
> (in the sgf file)
> 
> Cube analysis
> 2-ply cubeless equity  +0.657 (Money:  +0.653)
>     74.0%  18.9%   1.0% -  26.0%   2.6%   0.0%
> Cubeful equities:
> 1. No double            +1.040
> 2. Double, take         +1.234  ( +0.194)
> 3. Double, pass         +1.000  ( -0.040)
> Proper cube action: Too good to double, pass (17.0%)
> 
> With
>   set matchequitytable "met/zadeh.xml"
> 
> Cube analysis
> 2-ply cubeless equity  +0.658 (Money:  +0.653)
>     74.0%  18.9%   1.1% -  26.0%   2.6%   0.0%
> Cubeful equities:
> 1. Double, pass         +1.000
> 2. Double, take         +1.112  ( +0.112)
> 3. No double            +0.919  ( -0.081)
> Proper cube action: Double, pass
> 
> set matchequitytable "met/snowie.xml"
> 
> Cube analysis
> 2-ply cubeless equity  +0.617 (Money:  +0.615)
>     72.5%  18.7%   1.0% -  27.5%   3.1%   0.1%
> Cubeful equities:
> 1. Double, take         +0.932
> 2. Double, pass         +1.000  ( +0.068)
> 3. No double            +0.805  ( -0.127)
> Proper cube action: Double, take
> 
> set matchequitytable "met/jacobs.xml"
> 
> 
> Cube analysis
> 2-ply cubeless equity  +0.616 (Money:  +0.613)
>     72.3%  18.8%   1.1% -  27.7%   3.2%   0.1%
> Cubeful equities:
> 1. Double, take         +0.980
> 2. Double, pass         +1.000  ( +0.020)
> 3. No double            +0.849  ( -0.132)
> Proper cube action: Double, take
> 
> 
> Albert Silver wrote:
> > Snowie or Jacobs, I'm not 100% sure anymore.
> >
> >                             Albert
> >
> > On Tue, 2003-08-12 at 00:32, Joseph Heled wrote:
> >
> >>Joern thought maybe we are using different equity tables. I am using
> >>zadeh. What did you use for the WC?
> >>
> >>-Joseph
>