Re: [Bug-gnubg] Re: FIBS rating and error rates

[Joern Thyssen]

On Wed, Aug 27, 2003 at 12:08:32AM +0200, address@hidden wrote
> Hi,
> 
> I recently completed another series of simulations in order to determine
> the relation  between error rates  and FIBS rating. The  new simulations
> resulted in a much improved formula,  using a bilinear fit to move error
> and cube error separately.

I've modifyed gnubg to use the new formula.

> The results  are summarized on http://www.cs.ubc.ca/~kvdoel/tmp/ratings,
> and I'd appreciate it if anyone  could have a critical look, making sure
> I didn't mess up anywhere.

Let me know when you're done -- I'll texiify the text and add it to the
manual.

> Bottom line is the following formula for the rating difference:
> 
> R = a2(N)*EPM+b(N)*EPC,
> where
> a2(N) = 8798 + 2526/N,

I guess you made a typo here: it should be 25526

Jørn

pgpWiBekmXgHi.pgp
Description: PGP signature