DetOffsets input in MARX: Need to use DY_AVG and DZ_AVG?

[hamogu]

The thread http://cxc.harvard.edu/ciao/threads/marx_sim/ claims that the DY_AVG and DZ_AVG header keywords are supposed to be used in the computation of DetOffset inputs to MARX.

I need to investigate why the thread's author thought that that's necessary. I think that MARX should to that itself by accessing the telD1999-07-23aimptsN0002.fits file (it's part of the MARX distribution), so this step should not be necessary. If it is, it's a bug in MARX that I need to investigate. Note that for future observations improved thermal modelling will most likely allow mission planning to aim at the nominal aimpoint again and correct for the drift in the observation planning already. There is a memo about that, but I cannot find it right now.