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

20031021: 20031015: Problem with NAWIPS version K build on SGI IRIX6.5 (no wind barbs plotting with gridded data)



Bob,
Yes, I posted the source code patch, or recommended downloading the
Garp GEMPAK5.6.j binary from the patch directory (I could cross name it 
Garp binary for 5.6.K since there is no difference in the code).

Steve Chiswell


>From: "Creasey, Robert USA" <address@hidden>
>Organization: UCAR/Unidata
>Keywords: 200310212135.h9LLZROb011506

>Steve,
>Thanks for your time and help...the patch did it...I missed seeing that...FYI 
> the LINUX binary's for version K also had the problem.
>Bob
>
>-----Original Message-----
>From: Unidata Support [mailto:address@hidden]
>Sent: Sunday, October 19, 2003 2:16 PM
>To: Creasey, Robert USA
>Cc: address@hidden
>Subject: 20031015: Problem with NAWIPS version K build on SGI IRIX6.5
>(no wind barbs plotting with gridded data) 
>
>
>
>Bob,
>
>Though you didn't mention it, I'm assuming you mean that the plot is in Garp.
>The rest of the programs should be fine. I did find one porting need to upgrad
> e garp
>to work with the vectors due to the IJSKIP addition.
>
>There are no changes with Garp between 5.6.J and 5.6.K otherwise, so you can e
> ither:
>1) Copy your Garp executable from 5.6.J to the 5.6.K distribution and use that
> ,
>or
>2) Go to the patches download area of GEMPAK, eg:
>   http://my.unidata.ucar.edu/modules.php?name=Downloads&d_op=viewsdownload&si
> d=10
>and download the routine:
>   Garp pvgrid.f routine update for 5.6.K (pvgrid.f_5.6.k) 
>
>   Description: This patch fixes a bug introduced in Garp with version 5.6.K w
> here changes to the underlying GEMPAK library 
>   routines created display problems with gridded vectors. Replace $GARPHOME/g
> empak/pvgrid.f with this routine. NOTE: This 
>   patch should only be used with GEMPAK 5.6.K.
>
>After replacing the specified routine, you can build garp again with:
>cd $GARPHOME
>make all
>make install
>
>If you've looked at the individual grid programs like gdcntr or gdwind, etc, y
> ou'll see that
>IJSKIP is a new parameter for allowing the user to subsample large grids (like
>  the NDFD
>national 5km output). This changed the GEMPAK library and therefore the Garp c
> alling sequence
>in the specified routine pvgrid.f was broken.
>
>
>Steve Chiswell
>Unidata User Support
>
>
>
>
>
>>From: "Creasey, Robert USA" <address@hidden>
>>Organization: UCAR/Unidata
>>Keywords: 200310152314.h9FNEAOb028391
>
>>
>>I have built version K for our SGI Octanes (IRIX 6.5  n32) with no error list
> e
>> d during the build (several warning).  All seems to work ok with gridded sca
> l
>> ar functions but the vector fields come up with o o o o's instead of wind ba
> r
>> bs at every grid location  (Info in model plan view shows gem files).  Versi
> o
>> n J worked fine .... displaying wind barbs with same data... 
>>
>> I have rebuilt k  several times,  compared Gemenviron between j and k... and
>  
>> I am starting to chase my tail....  Any Ideas?  
>>
>> Do you have a built version for SGI?
>>
>>(Could send you log of build if need be or images of example graphic).
>>
>> Bob Creasey                         (  +*))   (+ * )_ 
>> Naval Postgraduate School          ((* - * )(++,, , )) 
>> Dept of Meteorology Rm 254 MR/CY (( ,- *-  ,, -,- ), ))__
>> 589 Dyer Rd.                    (_______________________))
>> Monterey Ca 93943-5114              \   \     \\\\\\\    
>> address@hidden               / \  /\_    \\\\\\   
>>                                     / / / \    \\\\\      
>> 
>
>>
>
>