Fill alternative TPC dedx according to dEdxClusterRejectionFlagMaskAlt, store full reference #14144
Conversation
…skAlt opt. The old TrackTPC::getdEdx() is unchanged, the new info can be accessed as TrackTPC::getdEdxAlt()
|
REQUEST FOR PRODUCTION RELEASES: This will add The following labels are available |
|
Hello @shahor02, I’ve replied in the other pull request regarding clipping: To clarify the normalization logic — which is still unclear to me — I suggest we have a quick Zoom meeting. Ideally, we should add a decode function (relative to the track table signal) to the streamer. That way, I can be 100% sure the normalization is applied consistently, and we avoid any risk of mixing tracks from two incompatible categories. Best regards, |
…rmalization When filling AOD extra track info, the TPCSignal and the number of clusters used for PID are overridden to those coming from the TrackTPC::dEdxAlt if the nominal dEdx.dEdxTotTPC==0. This is flagged by extraInfo.flags |= o2::aod::track::TPCdEdxAlt The ratio is stored with the same precision as the nominal dEdxAlt.dEdxTPCTot in the extraInfo.TPCSignal. +Extra fixes like clamping to uint8_t instead of casting float to it.
2 participants
This is an alternative to #14140
It stores in the AO2D directly
Also, the
TPCSignalin theextraInfotree is not overridden to dEdxInfoAlt.dEdxTotTPC ifdEdxInfo.dEdxTotTPC==0