When doing 2016 recon with a current jar, very few events are getting written. At first I thought it was from applying a beam position offset when making zero beamwidth events, or something specific to the iss154 branch.

However, it happens during readout, and also with a current master branch.

This is a comparison of readout from around the last official release (which I have been using for production) and a current trunk:

July 18 readout:

Trigger Processing Results
        Single-Cluster Cuts
                Total Clusters Processed     :: 4692
                Passed Seed Energy Cut       :: 4692
                Passed Hit Count Cut         :: 4295
                Passed Total Energy Cut      :: 4243

        Cluster Pair Cuts
                Total Pairs Processed        :: 1727
                Passed Energy Sum Cut        :: 1683
                Passed Energy Difference Cut :: 1679
                Passed Energy Slope Cut      :: 1622
                Passed Coplanarity Cut       :: 1278

        Trigger Count :: 1278

The same events with an updated master:

Trigger Processing Results
        Single-Cluster Cuts
                Total Clusters Processed     :: 4289
                Passed Seed Energy Cut       :: 4289
                Passed Hit Count Cut         :: 1021
                Passed Total Energy Cut      :: 1017

        Cluster Pair Cuts
                Total Pairs Processed        :: 107
                Passed Energy Sum Cut        :: 107
                Passed Energy Difference Cut :: 107
                Passed Energy Slope Cut      :: 106
                Passed Coplanarity Cut       :: 79

        Trigger Count :: 79

The only difference in the steering files is the absence of the ConditionsDriver in the updated one. Does anyone know what change could be causing this? It looks like more events are failing the cluster hit count cut. Is this expected behavior?

-Bradley