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

[IDV #ZSD-139205]: IDV - RHI Does Not Plot

Hi Greg-

The latest nightly build (7-28) addresses the problems you found:

- no more out of memory error for the probes (either CDM or old
way).
- It looks like the new IDV is displaying the data correctly.
For the BMX files, the NCDC and old IDV code (based on the same
reader) is showing 367 radials for each elevation and there are
really only 360 in this file.  Since the GEMPAK displays are
identical to the new IDV displays and the code to read that
was developed independently, we're inclined to think the new
IDV displays are "truth".

If you have time to check this out and still have problems,
let me know.

Don Murray

> -----Original Message-----
> From: Unidata IDV Support [mailto:address@hidden]
> Sent: Thursday, July 27, 2006 9:46 AM
> To: Salottolo Greg
> Cc: address@hidden
> Subject: [IDV #ZSD-139205]: IDV - RHI Does Not Plot
> 
> 
> Hi Greg-
> 
> > Yes, I was too quick.
> >
> > Got the latest. However, the probe out of memory error still occurs with 
> > CDM checked. Ok with previous reader (CDM not checked).
> 
> Are you probing a single level or the entire volume? I tried both
> the old and the new and for a single level and memory useage was
> the same.  For probing the volume, the non-cdm way (old way)
> taxes my system and almost runs out of memory, but obviously the
> new way is using just a little more memory which throws it over
> the threshold.  I got about the same memory results with 5 files,
> but on 7, it pushes the system over the edge?
> 
> >>> Was probing volume. Single level OK.
> 
> > I am also seeing the image shift with the new reader. Old reader consistent 
> > with NCDC Viewer display (BMX data with aircraft track). New viewer image 
> > displayed about 2 km northeast (real rough estimate) when compared with 
> > NCDC Viewer. Of course does the NCDC Viewer image represent the "truth."
> 
> Yes, that was the issue I mentioned yesterday.  I've compared
> the NCDC viewer, the IRAS viewer (which the NCDC viewer and old
> IDV code was based on) and GEMPAK.  The new IDV and GEMPAK displays
> are consistent with no gaps, while the NCDC and IRAS viewers have
> the gaps (in the 14:03 image).  We'll be tracking this down today
> and hopefully figure out which is "truth".
> 
> Let me know if you see other issues and thanks again for your
> input and testing.
> 
> > Greg ...
> >
> > -----Original Message-----
> > From: Unidata IDV Support [mailto:address@hidden]
> > Sent: Thursday, July 27, 2006 8:36 AM
> > To: Salottolo Greg
> > Cc: address@hidden
> > Subject: [IDV #ZSD-139205]: IDV - RHI Does Not Plot
> >
> >
> > Hi Greg-
> >
> > > Installed 7-26 nightly. Same error regarding probe. Did not see reader 
> > > option. Maybe build did not run on 7-26 PM.
> >
> > The early bird gets the old build? ;-)
> >
> > The "nightly" build didn't finish until after 6 am.  Try the again
> > with the 7-27 nightly build.
> >
> > Maybe we should change nightly to daily. ;-)
> >
> > Don
> >
> >
> > Ticket Details
> > ===================
> > Ticket ID: ZSD-139205
> > Department: Support IDV
> > Priority: Critical
> > Status: Open
> >
> >
> >
> 
> 
> Ticket Details
> ===================
> Ticket ID: ZSD-139205
> Department: Support IDV
> Priority: Critical
> Status: Open
> 
> 
> 


Ticket Details
===================
Ticket ID: ZSD-139205
Department: Support IDV
Priority: Critical
Status: Open