[daip] IMAGR

[Jim Ulvestad]

I can't find any evidence that IMAGRPRM(11) actually works,
when you are doing multi-resolution cleaning.  

The evidence--we set NGAUSS to 4, WGAUSS to 0,2,4,8.
We also set IMAGRPRM(11)=0.5.  According to my reading of
the help file, the selection of which field to clean
should then depend on the quantity

Peak flux/(bmin*bmax)**0.5

The initial images are made, with field 1 having a
peak of about 80 mJy/beam in a 1 arcsec beam, and
field 4 having a peak of about 150 mJy/beam in an
8 arcsec beam.  According to the formula above,
the weight of field 1 then should be "80,"
the weight of field 4 should be 150/8= "20".
So IMAGR should pick its first components out of
field 1.  Instead, it goes merrily off and finds
its first components in field 4.

So it doesn't look to me as though IMAGRPRM(11) is
being used.  However, I could easily be missing 
some other input that is used to invoke this, or
I could be misunderstanding the EXPLAIN file.

Data set available on Stacy Teng's machine, on
request ...

jim