Previously, I was looking into the difference between unfolding data using a Pythia8 generated response and a Pythia6 (embedding) generated response:

https://drupal.star.bnl.gov/STAR/blog/dmawxc/update-07102019-run-9-pp-comparing-data-unfolded-pythia8-priors-vs-pythia6-priors

There seems to be something different between unfolding with the Pythia8 and Pythia6 responses, and previously I noticed that the ratio between the particle and (matched) detector level spectra were different between Pythia8 and Pythia6.  This lead me to suspect that there was something up with the jet-matching algorithm in the Pythia8 case.  One thing that came to mind was the "trigger logic" I was applying to the Pythia8 case...

In the Pythia6 jet-matching, I apply trigger cuts at the particle-level only.  However, in the Pythia8 jet-matching, I apply trigger cuts at both the particle- and detector-level.  The plots below check if it makes an impact if I apply trigger cuts at the particle-level, detector-level, or both.

As you can see, it makes no impact.  This is to be expected considering the way the Pythia8 events are generated: I save only events with a pi0 or prompt photon that satisfy certain pT and eta criteria, and then I apply the parameterized detector response to each saved event.  So the Pythia8 particle-level and detector-level triggers are the same.  Hence, this is not the reason for the difference between the Pythia6 and Pythia8 corrections.