[daip] IMAGR problem

[Christina Lacey]

Hello,

I am using 31DEC99  and am experiencing an unusual problem. I intend to
upgrade to the latest AIPS version,  but want to fix/understand  the
problem I am experiencing first.  I am running AIPS on a PC, dual 933 MHz,
512 Mb ram and
using red hat 6.2 Linux.  Previously I used a dual 533 MHz, 512 Mb ram PC
with red hat 6.1 and never had this problem with this data.

At times, Imagr outputs files with no values set for the pixels when I run
it with DOTV= -1.  With DOTV=1, I can somtimes get IMAGR to work, but not
always. When I look at the imae header,  the maximum and minimum are
labeled:
Minimum=            NAN      Maximum=            NAN JY/BEAM

The CC file associated with the image is normal.

 Here is the output from an imagr run that failed. Once
an imaging run fails, it fails everytime. I used all default values in
IMAGR, except that I set channels to make a continuum image and I set 
imsize, cellsize, and niter.

affirm> IMAGR1: Task IMAGR  (release of 31DEC99) begins
affirm> IMAGR1: FNDPOL: Stokes I computed assuming V=0
affirm> IMAGR1: FNDPOL: Stokes I computed assuming V=0
affirm> IMAGR1: FNDPOL: Stokes I computed assuming V=0
affirm> IMAGR1: Create 2403_L980727.IMAGR .   1 (UV)  on disk  4  cno    1
affirm> IMAGR1: Beginning channel    1 through    7 with  2 IFs
affirm> IMAGR1: FNDPOL: Stokes I computed assuming V=0
affirm> IMAGR1: IMACPY: Copied   453340 visibilities to be imaged
affirm> IMAGR1: **** WARNING data included out of inner portion of UV
plane ****
affirm> IMAGR1: **** Watch for high-frequency & other poor cleaning
effects ****
affirm> IMAGR1: UVWAIT: begin finding uniform weights
affirm> IMAGR1: UVWAIT: Average grid weight  6.047E+03
affirm> IMAGR1: UVWAIT: Adding temperance S  1.209E+03
affirm> IMAGR1: UVWAIT: begin applying uniform or other weights
affirm> IMAGR1: UVWAIT: Sum of weights in  1.290E+09 and out  2.921E+09
affirm> IMAGR1: UVWAIT: Noise is increased by a factor 1.157 due to
weighting
affirm> IMAGR1: UVWAIT: Average summed weight  6.021E+03 over    5085254
vis
affirm> IMAGR1: Create 2403_L980727.IBM001.   1 (MA)  on disk  4  cno    8
affirm> IMAGR1: Create 2403_L980727.ICL001.   1 (MA)  on disk  4  cno    9
affirm> IMAGR1: GRDFLT: X and Y convolution type = SPHEROIDAL
affirm> IMAGR1: GRDFLT: X and Y parms =  3.0000  1.0000
affirm> IMAGR1: GRDFLT: convolution function sampled every 1/100 of a cell
affirm> IMAGR1: GRDMEM: Ave   14 Channels; 1.455625E+09 to 1.444375E+09 Hz
affirm> IMAGR1: Field   1 Sum of gridding weights =  3.10507E+10
affirm> IMAGR1: Field   1 Beam min =  -86.7 MilliJy, max =    1.0      Jy
affirm> IMAGR1: Field   1 fit FWHM =    4.406 x    3.905      arcsec, PA=
43.2
affirm> IMAGR1: CLNUV : Sorting data to make them fit
affirm> IMAGR1: Field   1 min =    nan      Jy,max =    nan      Jy
affirm> IMAGR1: CDECID: Will load every  834 th. residual
affirm> IMAGR1: BGC Clean: using  103 cell beam + residuals >      nan
Jy
affirm> IMAGR1: WARNING: > MAXPIXELS LOADED TO AP!!
affirm> IMAGR1:    20019 Residual map points loaded
affirm> IMAGR1: Reached Iter. limit, Max resid=     nan      Jy iter=
1000
affirm> IMAGR1: Total Cleaned flux density    =      nan      Jy     1000
comps
affirm> IMAGR1: ALGSUB: Ipol gridded model subtraction, chans    1 through
14
affirm> IMAGR1: Field   1 min =    nan      Jy,max =    nan      Jy
affirm> IMAGR1: Total Clean components     1000 reaches limit    1000
affirm> IMAGR1: Restoring Clean components
affirm> IMAGR1: Checking image max/min
affirm> IMAGR1: Deleting UV work file:
affirm> IMAGR1: Destroyed  1 extension files of type AN
affirm> IMAGR1: Destroyed  1 extension files of type FQ
affirm> IMAGR1: Destroyed UV image file: catno=      1 disk= 4
affirm> IMAGR1: Appears to have ended successfully
affirm> IMAGR1: affirmed     31DEC99 TST: Cpu=    1060.8  Real=    1124

I have never encountered the MAXPIXELS warning before.  Do you have any
idea what is going on?  Is this a problem with my version of AIPS that
upgrading will fix? 

Here is the end of a DOTV =1 IMAGR with the same data and IMAGR
paramters as above. IMAGR seems happy until it makes the cleaned image.
I don't see the MAXPIXELS warning anywhere in this run of IMAGR.

affirm> IMAGR1: Field   1 min =   -1.3 MilliJy,max =   10.3 MilliJy
affirm> IMAGR1: Loading field   1 from -2.28E-03 to  2.20E-02 every  7
pixel
affirm> IMAGR1: You have    30 seconds to select a menu item by:
affirm> IMAGR1: Press buttons A, B, or C to choose an operation
affirm> IMAGR1: Press button D for on-line help
affirm> IMAGR1: Clean continuing
affirm> IMAGR1: BGC Clean: using 2001 cell beam + residuals >     1.27
MicroJy
affirm> IMAGR1:     4970 Residual map points loaded
affirm> IMAGR1: Reached Iter. limit, Max resid= 149.256 MicroJy iter=
1000
affirm> IMAGR1: Total Cleaned flux density    =   22.919 MilliJy     1000
comps
affirm> IMAGR1: ALGSUB: Ipol gridded model subtraction, chans    1 through
14
affirm> IMAGR1: Field   1 min =   -1.3 MilliJy,max =   10.3 MilliJy
affirm> IMAGR1: Total Clean components     1000 reaches limit    1000
affirm> IMAGR1: Loading field   1 from -2.28E-03 to  2.20E-02 every  7
pixel
affirm> IMAGR1: Restoring Clean components
affirm> IMAGR1: Checking image max/min
affirm> IMAGR1: Loading field   1 from       NAN to       NAN every  7
pixel
affirm> IMAGR1: Deleting UV work file:
affirm> IMAGR1: Destroyed  1 extension files of type AN
affirm> IMAGR1: Destroyed  1 extension files of type FQ
affirm> IMAGR1: Destroyed UV image file: catno=      1 disk= 4
affirm> IMAGR1: Appears to have ended successfully
affirm> IMAGR1: affirmed     31DEC99 TST: Cpu=     308.3  Real=     378

What is going on?  Let me know if I need to send you any other output or
info to diagnose the problem. 

Thanks, Christina

Dr. Christina Lacey		lacey at sc.edu
University of South Carolina	office: 803-777-6089
Dept. of Physics and Astronomy	fax: 803-777-3065
Columbia, SC 29208