More information on meedium deep u band stacks

There is a problem in making the medium deep u band megacam stacks - it worked fine when Gene did this in 05/2011, but now it doesn't (why). I've done some investigations, it appears to be the recipes:

Ingredients:

• make a stack of all the uband warps with this label: u.MD01.20110602
• use different combinations of code or recipes to figure out what's wrong

First, I checked out the trunk at r31601 (which is the r that gene's processing claims to be, except on the eam branch). As it was, it didn't have the correct recipes for TEST1 for megacam, so I copied over the ippconfig from the current trunk (r33784). I did various iterations of different ippconfigs (from back then to now) and trunks (from then to now). Here's a summary:

branch/trunk? what?	ippconfig?	result
trunk r31601	trunk r33784	bad stacks
eam branch 20110505 -r31601	trunk r33784	bad stacks
eam branch 20110505 -r31601	eam branch 201105 (r31657)	good stacks
trunk r33784	trunk r33784	bad stacks
trunk r33784	eam branch 201105 (r31657)	good stacks

So, from this I conclude that there's something in the recipe. I need to investigate this: the log files indicate no (or little) difference in the recipes, but basically only include ppStack's recipe. It turns out when stacking it also uses ppSub and other configurations, and I need to be more careful and investigate the mdc files to determine what's changed...

However, I present some pictures (as proof) of the good/bad stacks:

Gene's original processing (r31657 eam_branches/20110505)
r33784 (trunk and config) - bad stacks
trunk = r33784 and config = r31657 from gene's branch - good stacks