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

[IDV #JIJ-912647]: IDV - Possibly Collaboration bug, file sharing, RTMA GRIB data display question,

Hi Dan-

> Bryan has installed IDV 2.0RC3 on our AMD 64-X2 machine. The parameters
> from single RTMA grids are showing up perfectly--also, the problem of
> subsetting the derived projects I mentioned earlier today has also been
> fixed. It also seems like memory usage has gone done considerably when
> trying to view the RTMA data.

Glad to hear it.

> However, I can no longer view image sequences of the RTMA grid data. When
> you originally mentioned concating the two files together, I tried using
> the Linux command "filea fileb > filec"... which worked properly in RC2.

Did you mean "cat filea fileb > filec"?  I downloaded the file you
pointed to (which I assume is the result of the above) and didn't 
have this problem with RC3.

> RC3 craps out when it tries to create a display from a file like filec.
> The exact error I get is attached to this email. When I close the error
> warning, IDV plots some RTMA data, but the timestamp is definately wrong
> (and it only plots one time). Anyways, I've put three files up on our ftp
> server, at ftp://ftp.met.utah.edu/pub/jhorel/support if you'd like some
> more samples. The first two files that begin with gsi are the original
> grid files, and the rtmatest.grb2.ndfd is a file I created using the cat
> command from those two files.

Did you concatenate different files to the same test file name and then
try to open that in the RC3.  We create an index of the file for quick access
to the GRIB records and it might be that the index file is still pointing
to offsets for the old files.  Try running the cat command again to a
different file name and then open that in 2.0RC3.  Otherwise, I'm not
sure what it going on.  

> I also noticed that you have completely removed collaboration from RC3. Is
> there any way that you can put it back into RC4, and just let users know
> that it's not supported yet? Anyways, this is just a request, I know it's
> still buggy.

We were trying to reduce the number of support questions about this. ;-)
Our plan is to have the collaboration menu available through a plug-in
(File->Plugins) which people could download if they want to use it.
But ponder your request because if you want to do collaboration with
someone else, they both have to download the plugin.

> Finally, in one of the emails you wrote John Horel, you mentioned time
> binning. Could you explain to me how this feature is suppose to work? John
> wasn't too sure of it either. Ideally, we'd like to seperate obs made at
> certain times from the other obs. Our ideal situation would be where we
> could just see obs made +-15 or +-30 minutes from the hour. Is this
> possible with the time binning tool? I tried playing with it today, but it
> didn't seem to work properly (not display the right obs the way I
> described above).

The binning works to group all observations in a bin.  So, in your
example, if you set your time binning to be 45 minutes past the hour
and a bin size of 30 minutes, you should get what you want.  However,
it looks like there is a bug so when I try this I always get a time
of on the hour and 30 minutes past.  Is that what you are seeing?
BTW, this is a new feature, so thanks for the input on how you want
it to work.  For the next (2.1) release, we will be working on spatial
and temporal subsetting of point data and more efficient access to it
locally and remotely.  If you have more input on what you'd like to
see in terms of functionality, send us a note (please start a separate
thread though).

Don Murray
> Thanks for all your help; have a good evening.
> 
> Dan
> 
> 
> >
> >>
> >> Ticket Details
> >> ===================
> >> Ticket ID: JIJ-912647
> >> Department: Support IDV
> >> Priority: Normal
> >> Status: Open
> >>
> >>
> >
> >
> >
> >
> 
> 
> 
> 


Ticket Details
===================
Ticket ID: JIJ-912647
Department: Support IDV
Priority: Normal
Status: Open