[daip] Re: RMSD question

[Leonia Kogan]

Mark,

>I tried the new version and indeed it does work on larger images.
>Obviously this routine is very computer time intensive and we
>may have problems running it on our 8192x8192x100 SiO masers
>image cubes.

Yes I could expect the time problem.

>Would it be possible to add XINC YINC parameters that would allow 
>the program to calculate the rms levels far less often and then 
>interpolate to estimate the pixel rms?

It is possible. It is the question of my time. Let me keep it in my mind.

On ythe other hand: are you sure that interpolation can give a plausible result?

Leonia

----- Begin Included Message -----

>From reid at cfa.harvard.edu Wed Feb  6 12:49 MST 2002
X-Authentication-Warning: mjr.harvard.edu: reid owned process doing -bs
Date: Wed, 6 Feb 2002 14:52:13 -0500 (EST)
From: Mark Reid <reid at cfa.harvard.edu>
X-X-Sender:  <reid at mjr.harvard.edu>
To: Leonia Kogan <lkogan at zia.aoc.NRAO.EDU>
Subject: Re: RMSD question
In-Reply-To: <200202061520.IAA13023 at bonito.aoc.nrao.edu>
MIME-Version: 1.0
X-Lines: 83
Status: RO


Leonia,

I tried the new version and indeed it does work on larger images.
Obviously this routine is very computer time intensive and we
may have problems running it on our 8192x8192x100 SiO masers
image cubes.  

Would it be possible to add XINC YINC parameters that would allow 
the program to calculate the rms levels far less often and then 
interpolate to estimate the pixel rms?

Mark

On Wed, 6 Feb 2002, Leonia Kogan wrote:

> Mark,
> 
> I think you uses the original version of RMSD wich has a limit of 1024
> pixels.  I changed the limit to 8192 soon after put the task at the
> system.
> 
> 
> Please, look at the fortran codes: $APGNOT/RMSD.FOR
> The fiteenth line should be:
> 
>       PARAMETER (IMSIZE = 8192)
> 
> If you see something less instead of 8192, that confirms you use the
> old version and something wrong is with MNJ
> 
> Leonia
> ----- Begin Included Message -----
> 
> >From reid at cfa.harvard.edu Wed Feb  6 08:10 MST 2002
> X-Authentication-Warning: mjr.harvard.edu: reid owned process doing -bs
> Date: Wed, 6 Feb 2002 10:12:31 -0500 (EST)
> From: Mark Reid <reid at cfa.harvard.edu>
> X-X-Sender:  <reid at mjr.harvard.edu>
> To: <lkogan at cv3.cv.nrao.edu>
> Subject: RMSD question
> MIME-Version: 1.0
> X-Lines: 24
> Status: RO
> 
> Leonia,
> 
> I tried RMSD and it looks like a very useful task
> for images that have dynamic range limitations near
> strong sources.   Thanks for making the task.
> 
> I had one problem running it.  I tried to feed it
> an sub-image that was 3000 pixels on a side, and
> it complained that the X-axis was too large.
> Can you look into this problem?  Ultimately, 
> we will need to analyze 8192 pixel images.
> 
> Thanks,
> Mark
> 
> 
> -------------------------------------------------------------
> Mark J. Reid                 Phone: 617-495-7470 
> Harvard-Smithsonian CfA      Fax  : 617-495-7345
> 60 Garden Street             Email: reid at cfa.harvard.edu
> Cambridge, MA 02138, USA     Web  : cfa-www.harvard.edu/~reid                  
> -------------------------------------------------------------
> 
> 
> 
> 
> ----- End Included Message -----
> 


-------------------------------------------------------------
Mark J. Reid                 Phone: 617-495-7470 
Harvard-Smithsonian CfA      Fax  : 617-495-7345
60 Garden Street             Email: reid at cfa.harvard.edu
Cambridge, MA 02138, USA     Web  : cfa-www.harvard.edu/~reid                  
-------------------------------------------------------------




----- End Included Message -----