The attached txt files provide direct comparison statistics of matched sources between the LSST and HSC RC dataset runs. The *_sum.txt show the per visit/tract results, whereas the *_shortSum.txt show just the averages over the visit/tract entries.

All of the diagnostics shown have accompanying figures. Several example plots are also attached.

While most of the global statistics look quite good, closer scrutiny of these figures and those in DM-6818 have revealed some regressions in the LSST stack. Of particular note, the CModel vs. psf plot on DM-6818 (plot-t0-HSC-I-mag_modelfit_CModel-psfMagHist.png) reveals a large density of objects that have likely been misclassified as galaxies (namely the strip of red objects just below the threshold for the star/galaxy classification) and a "bump" at bright magnitudes in the "stars". The latter seems to be rooted in the psf modeling at the single frame stage, as indicated in the direct size comparison plot shown here:

An investigation into the causes of these issues will be conducted as part of the DM-8299 epic.

Filed DM-8532 & DM-8533 to investigate the issues described above.

Given that this comparison has been performed and follow-up issues filed, I think we can regard this as done.

Lauren MacArthur, if there are further issues resulting from this comparison, please file them as part of DM-8299.

Chatted to Lauren; agreed that we don't expect anything more on this ticket.