[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

20030128: Some more advice needed ... and mods to MSG nav/cal



>From: "=?ISO-8859-1?Q?Marianne=20K=F6nig?=" <address@hidden>
>Organization:  EUMETSAT
>Keywords:  200301231030.h0NAUbx07842 McIDAS ADDE

Marianne,

>Thank you Tom,,
>that was very helpful. I'll try this "central ADDE server" approach. Reading
>in sampled data is not a solution, because we would always want to have full
>resolution.

I thought that was probably the case, but I felt the need to mention
all possibilities.  The good thing about McIDAS, however, is that the
pixels returned in a blow-down are the original pixels from the image.
You may not get all of them (you get every Nth pixel for a -N
blow-down) but what you do get are original values.  This does not, of
course, help in studies of the pixel to pixel variation of images.

>Reading in sections of the image might be an alternative. 

OK.  This is more complicated to manage in some respects, but it is
a common technique.

>I have tested my data accessing routine (not the actual MSG server, but the
>program which we use as an interface to any other software which needs
>satellite) with (a) raeding in a full image at first call, and (b) reading
>it in line bu line with successive calls to mcaget, and the time difference
>is roughly a factor of 100, so there is a considerable difference. I'll try
>to solve this with my Unix support.

Faster machines would help also (reference your earlier message asking
about the various platforms that we use as test beds here at Unidata).

>Also, just as info to you: I have looked into the current versions of the MSG 
>nav and cal routines (nvxmsg.dlm and kbxmsg.dlm) and made some changes to the
>m to really account for the true navigation/calibration. I have mentioned that
>to Dee, telling her that I'd like to contact the responsible MUG programmer
>to get to some common understanding concerning the MSG calibration and
>navigation - but unfortunately I never got an answer from her. That is a
>little frustrating. 

I understand your frustration.  The folks at SSEC have been very busy
lately planning the next McIDAS upgrade, so the time is ripe for
getting MSG changes into the system.  It might be wise for you to send
an email to the mug (address@hidden) support crew so they can open
an inquiry for you.  This at least gets the stated need for
modifications into the inquiry tracking system where someone will
follow-up.

>Anyways, thank you agian for your help,

You are welcome.  

>and have a great week,

You too.

Tom

From: "=?ISO-8859-1?Q?Marianne=20K=F6nig?=" <address@hidden>
Date: Tue, 28 Jan 2003 16:24:03 +0100
Subject: MUG and MSG

Hi Tom,

As you suggested, I wrote a little mail concerning this MSG stuff.

At least our satellite images look really nice. Our ground segment
still fails t o properly navigate the images (but they are only 10
degrees off in both lat and lon, so who cares ...), but they are
seriously working on this, because a proper navigation is an absolute
MUST for all in-house post-processing.

I got our Unix experts on this mcaget/mcalin/server problem, so I hope
we can co me up with a solution. I will let you know - this might be of
interest to more p eople.

Marianne