meeting 2023 11 20 - JacobPilawa/TriaxSchwarzschild_wiki_5 GitHub Wiki
-
Cleaning up / finalizing plots we wanted to see from last time -- here are the intrinsic 3D sigma profiles with bands representing all profiles within a delta chi2 of 10 from the minimum, to go with the corresponding beta profiles we looked at last time.
-
I've plotted the intrinisc sigma below as bands with all models less than chi2 of 10, but I've also added the mean and standard deviations of the profiles, too. In this way, the recovered sigmas are astonishingly close to the simulated sigmas.
-
I've also left in the other diagnostic plots we've been looking at for completeness.
-
Most Problematic Beta Profiles: G1_r3 & G1_r5; G2_r4;, G5_r3 & G5_r5
-
G1, G4, G5 consistently overpredict the halo.
Diagnostics for G1
Diagnostics for G2
Diagnostics for G3
Diagnostics for G4
Diagnostics for G5
As we've discussed, one of the limiting features on our
In early tests, the runtime associated with the modified orblib.f90 is not substantially different from the original version, but this hasn't been tested carefully. Additionally, the orbit library with the additional intrinsic moments grid is roughly 8% larger than the previous version. With our archiving routine, we've reduced the file size by ~90%, so even with the new grid the overall filesize is substantially smaller than just a few months ago.