[Rivet] CDF_1988_S1865951

[Anton Karneyeu]

Hi Hendrik,
thanks again for fast response!

Really, this is only a question of units. I thought the HepMC 
installation at cern uses GeV as default momentum unit, but MeV is 
default. As soon as I specify GEV unit in HepMC::GenEvent() constructor 
explicitly, the problem disappear.

And, finally, there is no problem with pythia8 main32.cc example as this 
question is described in the comments inside the example.

Cheers,
Anton


17.11.2010 19:12, Hendrik Hoeth:
> Hi Anton,
>
> Thus spake Anton Karneyeu (Anton.Karneyeu at cern.ch):
>
>> thank you for investigation. Running AGILE with HepMC 2.06.03 fix
>> this problem. But reveals another one with hepmc file generated by
>> pythia8 hepmc interface.
>
> If I run
>    sed -i "s/MEV/GEV/" generator.hepmc
> before running Rivet, everything works fine. You compiled HepMC with
> units of MEV and MM. Now the numbers stored in the hepmc file are in
> GeV, but the file says MeV. Rivet believed that HepMC got the units
> right, assumes that the center of mass energy is a factor 1000 different
> from what you generated, can't match that with any of the two allowed
> energies of 630 and 1800 GeV, respectively (for CDF_2002), doesn't book
> a histogram, but then tries to fill the histos. Woooops, a segfault.
>
> I agree that Rivet shouldn't go south. But then, you've got a unit
> problem somewhere between Pythia8 and HepMC. If you find it, please send
> a patch to Torbjörn, too, because his main32.cc example is also
> affected.
>
> Cheers,
>
>     Hendrik
>