This ART job compares the output between Serial Athena and AthenaMP digitization jobs. The output agrees already apart from the GenParticle masses. The TP conversion for these variables is lossy, but reproducible, so the fix is to ensure that the files being compared have gone through the same number of TP conversion cycles. Currently the AthenaMP jobs have an extra cycle due to the RDOMerge step to combine the outputs of the various workers. Adding a dummy RDOMerge step for the Serial Athena job should fix the differences.
NB This MR only affects an ART script, so will have no effect on the CI pipeline output.