Hi-
Here's the result with the fix to the switchOffReading statements for
the split-off study.
Using anaQA, my result with the split-off fix is:
BRBR = 0.0212526 +- 0.0025703(stat) +- 0.000953176(MC stat)
with VubAnalysisApp:
BRBR = 0.0212526 +- 0.0025703(stat) +- 0.000953176(MC stat)
and my result running on the old p00 inputs /u/ec/ursl/root/anaQA-p00/ is:
BRBR = 0.0211058 +- 0.00257061(stat) +- 0.000956234(MC stat)
which you would expect to be the same as Urs' result.
I am running with the latest version of the VubFit using the -W option as
Alessio pointed out to use the old 1d weights. Urs, I believe ran with
the older version of the fit, which would account for the small difference
in our results.
So it seems the effect of the split-off study is small and VubAnalysisApp
is consistent with anaQA
Ed
On Mon, 16 Jun 2003, Urs Langenegger wrote:
>
> Hoi,
>
> you can (nearly) reproduce the PRL numbers with
>
> RecoilAnalysis ursl-061003
> VubAnalysis ale_030609
>
> by following the instructions in
>
> http://www.slac.stanford.edu/BFROOT/www/Physics/Analysis/AWG/InclusiveSL/islrecoil/HOWTO-PRL
>
> There are small differences in the logfiles (in particular, it seems
> that a K+ misid PidTable as lost 12 tracks in the normalization, which
> I completely do not understand).
>
> The values you should get by following this recipe is
>
> 0.0206755 +- 0.00250616
>
> The PRL is
>
> 0.0205788 +- 0.00250994
>
> If you think that this reproducibility is not good enough, feel free
> to start working on it.
>
> Cheers,
> --U.
>
>
|