Update dd4hep references after 2907
remove symlink dd4hep->detdesc references to the ones from https://lhcb-nightlies.web.cern.ch/nightly/lhcb-master/2352/ Some issues observed when comparing dd4hep and detdesc references (some may have already been discussed in other places, but putting them also here for completeness)
- VP cluster monitoring: residuals look significantly larger in dd4hep than in detdesc
- allen_gaudi_velo_ut_with_mcchecking performance significantly worse in dd4hep than in detdesc (eff 81% vs 90% and ghost rate 18% vs 9%)
- hlt2_fast_reco_with_mcchecking performance significantly different between dd4hep and detdesc:
- Downstream efficiency lower in “fromBD” categories in dd4hep
- BestLong efficiency significantly worse in dd4hep (3% vs 90%)
- Best efficiency looks compatible
- LongGhostFiltered significantly worse in dd4hep
- hlt2_light_reco_calo_efficiency: dd4hep has much lower number of produced tracks (but calo efficiency looks compatible) - it’s weird because in hlt2_light_reco_pr_kf_with_mcchecking the number of tracks between dd4hep and detdesc looks similar
- hlt2_light_reco_pr_kf_without_UT_with_monitoring_2022:
- Way more tracks in dd4hep than in detdesc
- hlt2_light_reco_with_mcchecking:
- Downstream efficiency and ghost rate significantly better in dd4hep than in detdesc
- BestLong efficiency and ghost rate significantly worse in dd4hep than in detdesc
- hlt2_pr_kf_longmuon:
- Lower number of matched tracks in dd4hep
- More BadQuality tracks in dd4hep
- Hlt2_protoparticles_baseline:
- CaloFutureMergedPi0: different counters for CorrectedEnergy and Inner etc
- The counters look much more similar between dd4hep/detdesc for the Hlt2_protoparticles_fastest test
- hlt2_reco_baseline, hlt2_reco_baseline_with_mcchecking_mpp_with_bfield: significantly lower number of reco long tracks and long tracking efficiency in dd4hep than in detdesc
- hlt2_reco_calo_resolution_gamma: tracking counters different as in other tests, photon counters look compatible
- hlt2_reco_pattern_reco_hit_masking: lower number of long tracks in dd4hep, similarly to other tests