The plots below compare 5pt data and 12pt data. Both sets of data are unscaled in the plots.

Note: These trees use the "old" mass definition; files are located in pagebs/diJetAnalysis/jetFinderAnalysis/personalDijetTreeAnalysis/trees/akt06005_v3 and akt06012_v3 and use the "old" StRoot saved in ezarndt/dijets/StRootOld.

First comparing eta and phi:
 
Dijet mass for all, barrel-barrel dijets, barrel-endcap dijets, and endcap-endcap dijets:

Jet pt for high and low jets:

The same high/low jet pt plots from above but now broken into barrel and endcap:

Neutral fraction:

Neutral fraction for high/low jets broken into barrel and endcap: The 5pt data still looks relatively good in the endcap, the 12pt data definitely show a trigger bias for high neutral fraction in the endcap.

Basically these plots confirm that 5pt data is preferred.