[Rivet] LHCb minbias analysis in Rivet?

[Hendrik Hoeth]

Hi Andrea,

Thus spake Andrea contu (andrea.contu at cern.ch):

> as I said I managed to clean up the code and to fix the problems with
> p/pi and k/pi ratios. [...] I updated the files on lxplus at
> ~acontu/public/RivetPlugin.

I've looked at the code for quite a while now, cleaned it up a bit more
(whitespace, unused variables, manual computation of pT, manual
computation of abs(), variable declarations faaaar away from the
definitions, lots of dead code, ...) and I still have a couple of
questions/comments:

- In your lifetime calculation you have these lines:
    // Correct Pythia6 PIDs for f0(980), f0(1370) mesons
    if (pid == 10331) pid = 30221;
    if (pid == 10221) pid = 9010221;
  Why do you need them? Firstly, I can't believe Pythia screws up
  the f0 particle ids. Secondly, you introduce generator dependent code
  in a Rivet analysis -- shouldn't do that. Thirdly, those lifetimes are
  all so small that they don't play any role in your calculation.

- About your lifetime cut:
    // Lifetime cut: ctau sum of all particle ancestors < 10^-9 m
    // according to the paper (see eq. 5)
    const double MAX_CTAU = 1.0e-9; // [m]
    double ancestor_lftsum = getMotherLifeTimeSum(p);
    if ( (ancestor_lftsum < 0.0) || (ancestor_lftsum > MAX_CTAU) ) continue;
  I can't find equation 5 in the paper. In fact, the only reference to a
  lifetime cut on the level of 10^-9 meters (boy, how do you measure
  that?) is in footnote 2, where you ask for three stable particles in
  Monte Carlo to mimic your primary vertex requirement. From what I see,
  that whole lifetime calculation shouldn't be in the Rivet analysis at
  all.

Can you please comment?

Cheers,

    Hendrik

-- 
If your dreams don't scare you, then you are not dreaming big enough.