Hi Alessio,
>
> I cant' see the fix in the committed code:
> can you point me to the bug?
>
in fitInit.icc
>
> Okay...
> I need to clarify better this:
> the plot you're referring to
> (http://www.slac.stanford.edu/~asarti/recoil/mxq2/TestRicgen.eps)
> IS NOT made using weights for HY MC but JUST using Fermi Motion weights
I know Alessio. You did not get the point.
If there is a difference of 20% in the weights calculated by you and by
Riccardo, this has to come from somewhere.
One option is the difference in the non-res shape.
But this is not the case, since I see only small differences (<10-5%) and
we are talking about something around 20% in the weights...
Then, assuming that you both implemented correctly the weights
calculation, I think that the difference can be due to a different
breco efficiency on the RESONANT MC. And this can affect the final weights
in particular the very first ones (low mx).
Otherwise there should be a bug, probably in the relative normalization
resonant vs nonresonant.
Daniele
|