Trying to pin down what happened to the theta/phi reconstruction.
I am going to label the Transform Functions used as follows:
TF 1 - Standard TF prior to increasing the angular range. Allowed theta < 54.5
degrees.
TF 2 - Included by Brant but commented out. Allowed theta < 63 degrees.
TF 3 - Written by me, based on TF 2. Allowed theta < 60 degrees.
Timeline:
November 6, 2006 (simulated chiquita data)
Used Method 1, LDF 4, TF 1 - Looks good.
November 16, 2006 Added LDF 6 and TDF 5 to code. Did not change Transform Function.
November 20, 2006 Changed to TF2. There is a note that one shower reconstructs beyond the supposed limit of theta... Other notes on Nov. 27 & 30 say unable to reproduce this result. Never solved.
December 7, 2006 Chiquita data - still ok. On Dec. 20 it is noted that v2.34 was used.
January 19, 2007 Simulated Chiquita showers - looks ok.
January 23, 2007 A bug in CTShower code is fixed (related to averaging of CIT 1000).
February 2, 2007 Revised chiquita data set to include 18k showers (up from 16k). Now the odd theta/phi relationship appears.
February 15, 2007 (real chiquita data)
Probably used Method 1, LDF 4, TF 1.
New libraries 2.41 and 2.40 were installed Feb. 16, 2007.
=> Problem occurs around the time I fixed the Chiquita averaging bug and put in LDF 6 and TDF 5. The LDF/TDF are CHICOS-only, so I don't think I would have used them on the CHIQUITA data set.
CTShower Jan 23 Before fixing the Chiquita averaging bug.
CTShower Feb7 After fixing the Chiquita averaging bug.
Ran Diff on the two libraries
The Transform function was changed. Oddly the Jan 23 code uses the extended
TF (63 degrees) and the Feb 7 code uses the standard TF (54.7 degrees).
The autoFit4 program looks the same.