delphes 3 have included fake candidates ?

[Gavin]

Hello,

In delphes2, there are no fake candidates entering electron muon and photon collections. For delphes 3, I think it have included fake candidates in default delphes_card_ATLAS.tcl. I don't sure whether I am right ?

Cheers,
Gavin

[Gavin]

In delphes2, electrons and photons may create a candidate in the jet collection. We must use "EhadOverEem" to clean lepton collections from jets. But in delphes3, I don't need to do that. Do you agree with me?

[Michele Selvaggi]

Hi,

the UniqueObjectFinder module already takes care of cleaning leptons from jets (as long leptons as leptons pass the isolation criteria)

Michele

[Gavin]

Hi,

The paper (​http://indico.cern.ch/getFile.py/access?contribId=0&resId=0&materialId=slides&confId=266825) has used delphes 2 and a constant jet-to-photon fake rate of 10(-3) to simulate zphotonjet production. I have simulated zphotonjet process with delphes 3. But my result with delphes 3 is larger one order than that paper. (I have checked that paper using delphes 2).

Maybe I don't express clearly. I just want to know about the jet-to-photon fake situation in delphes 3.

Thanks,
Gavin

[Michele Selvaggi]

Hi Gavin,

If I understand correctly you are observing more photons than you are expecting. The isolation criteria may have changed between Delphes2 and Delphes3. Could you try to tighten the photon isolation criteria in your detector card? Or check that the isolation is compatible with that of Delphes 2.

I am not sure, however, how you could set some "jet-to-photon fake rate". To my knowledge this was not possible in Delphes2. Did you apply this directly to the Delphes output? As a side comment, please be aware that it is now possible to do this with Delphes3. You can create a module that converts reconstructed jets into photons, according to some parametrised probability. Let us know if you need this.

Michele

[Pavel Demin]

Just for information.

There was no photon isolation criteria in Delphes 2.

[Gavin]

Replying to mselvaggi:
Hi Michele,

If I understand correctly you are observing more photons than you are expecting. The isolation criteria may have changed between Delphes2 and Delphes3. Could you try to tighten the photon isolation criteria in your detector card? Or check that the isolation is compatible with that of Delphes 2.

There was no photon isolation criteria, so I didn't consider photon isolation in delphes 2.

I am not sure, however, how you could set some "jet-to-photon fake rate". To my knowledge this was not possible in Delphes2. Did you apply this directly to the Delphes output?

I just scaled the total events number with "jet-to-photon fake rate" when I dealt with delphes2 output.

As a side comment, please be aware that it is now possible to do this with Delphes3. You can create a module that converts reconstructed jets into photons, according to some parametrised probability. Let us know if you need this.

Yes, I need jet-fake-to-photon events. But I don't know how to do in delphes3.

Cheers,
Gavin

[Michele Selvaggi]

Hi Gavin,

just to be sure, could you precise the following: do you have too many photons (or too few) in Delphes 3?

You can tune the number of photons by relaxing (if you need more photons) or tighten (if you need less photons) the isolation criteria.
Moreover if want to reproduce exactly the Delphes2 results you can relax totally the isolation criteria.

To do so you can set the parameter to a large value:
set PTRatioMax 1000

Michele

[Gavin]

Hi Michele,

I use the default delphes_atlas_card in delphes2 and delphes3. But I relaxed totally the isolation criteria with PTRatioMax at 1000 in delphes3. I analysed the both delphes output files with same cuts. The results is below,

delphes2 27866 (I didn't use the jet-fake-photon rate)
delphes3 3124

Cheers,
Gavin

[Michele Selvaggi]

Hi Gavin,

thanks for the testing. Could you tell whether you are using the eflow algorithm in Delphes2?
In the detector card, the corresponding flag JET_Eflow should be to set to false (i.e 0)

If I remember correctly the photon collection was bugged in Delphes 2.
Could you also provide some small event sample (100 events) so that we can do some cross-check?

Thanks

Michele