20011018: IMGDISP problems.

>From: Giovanni Leoncini <address@hidden>
>Organization: SJSU
>Keywords: 200110181917.f9IJHA119842 McIDAS IMGDISP ADDE

Giovanni,
>
>I'm trying to visualize some satellite images, but sometimes mcidas,
>wouldn't help. Of course they're in area format, and I can see them if I
>use other software, eg NTL. I have 2 problems:
>
>a I'm able to visualize some images 
>(eg http://metsun1.met.sjsu.edu/~leoncini/images/VIS_010713_0025) but the
>command MAP gives me this message:
>
>MAP: Unable to find a NAV start point in image
>
>no matter what parameter or keywords I use, gives me the same result even
>though NTL displays lat lon lines and political borders correctly.

This problem is probably caused by there being a copy of a FrameN.M (e.g.
Frame1.0) file in your MCPATH that shouldn't be there.  Here is how
to determine if this is the case:

o from the McIDAS session in which you see the problem, use DMAP to
  list the location of all Frame* files:

  DMAP Frame

o If you see one or more Frame files in a directory _other_ than .../.mctmp/.../
  (i.e. a subdirectory of ~/.mctmp) than you need to delete that/those
  files _after_ exiting your McIDAS session.  Since this file may be owned
  by a different user, you may need to get 'root' to delete the file

>the 2nd problem is maybe worse:
>when I try to display 2 kind of images, not only one single image, mcidas
>never gives me a message and if I look at the processes I see one
>corrisponding to the command I typed, and another one "agetserv" that has
>the same time stamp, or 1 sec larger.

I do not understand this comment, but I betting that the problem you are
seeing is caused by the exact same thing as above.  The solution is the
same, EXIT all McIDAS sessions and delete FrameN.M files from all directories
in your MCPATH.

>you can find 2 examples of those images @
>http://metsun1.met.sjsu.edu/~leoncini/images/NCAL_010713_2130
>http://metsun1.met.sjsu.edu/~leoncini/images/VIS_010713_0025
>
>We are running MCIDAS 7.7 on a Sparc 2.7 machine.
>
>any clue?

Please try the suggestion above and let me know if that does not solve
your problem.

Tom Yoksas