two variables refering to the same object is a nightmare in programming,
sooner or later they will be dissynchronized.
Anyhow, it s been too much. I have expressed my opposition on the
"standard" convention and I ll shut up and follow any decisions made on
the analysis meeting.
Chi
On Thu, 3 Apr 2003, Tancredi Botto wrote:
>
> Hello: I think there are two valid points for a using a different phi
> definition: uniform steps in the fitter and possibly phi averaging.
> Discussing at lenght whether a cut is defined as (x>0 && x<1) instead
> of (x>0.5) is just too much... and is anyways not a valid reason (a cut
> on cos(phi) is just as good).
>
> I think we agreed there was no problem in having two variables defined:
> phi and varphi.
>
> _ Phi is 0-360, is what blastmc uses, is how theorethical calcs are made,
> is what you use in the q-reference system (phi_pq) and is what the rest
> of the world uses and expect.
>
> _ Varphi can be -90, 270 or whatever definition may become useful. I agree
> it is the most natural variable for tracks, See above.
>
>
> So I vote very much against dephasing Phi and go changing the montecarlo
> in the process. I feel it would only be an intellectual virtuosism. I
> have no problems in defining new variables: we do that all the times.
>
>
> Things must be obvious if you use a different name (varphi) for a different
> definition (-90, 270). And it is not true that with two definitions you
> double the work, because a lot of codes (e.g. MC) are already written for
> Phi so you save that part.
>
> If anybody thinks there are problems let's discuss this further.
>
>
>
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:29 EST