[FWTools] Differences between Windows and Unix proj4?

Frank Warmerdam warmerdam at pobox.com
Fri Jun 1 10:15:49 EDT 2007

Kolberg Sjur A wrote:
> Dear list;
> I am struggling with the re-projection of some data from the Norwegian met office (met.no). Through the process we have come to compare some simple test, transforming a point using the proj command in a console window. The met office has tried, with all of these versions:
> Rel. 4.3.3, 23 Sept. 1995
> Rel. 4.4.9, 29 Oct 2004
> Rel. 4.5.0, 22 Oct 2006
> and gets the projected position:
> spiralis:~$ /usr/bin/proj +ellps=WGS84 +proj=stere +lat_0=90 +lat_ts=60 
> +lon_0=0 -f '%.5f'
> 10.202 63.696
> 493913.51155 -2744506.20400
> Whereas I get, with version  
> Rel. 4.5.0, 22 Oct 2006, installed as a part of fwtools 1.2.2 or 1.3.1:
> C:\Program Files\FWTools1.2.2>proj +ellps=WGS84 +proj=stere +lat_0=90 
> +lat_ts=60 +lon_0=0 -f '%.5f'
> 10.202 63.696
> '496554.53029' '-2759181.42981'
> Are there some hidden defaults here, that might be different?


There is a defaults file, which can be ignored by adding +no_defs to your
parameters.  However, while I was able to reproduce the same difference you
see adding +no_defs had no effect.  Furthermore, I tried adding -V to see
all the detailed information of processing and it failed on windows
complaining of some problem parsing input.

I'd appreciate it if you could file a bug against PROJ.4 on this issue as
I may not have time to pursue it in detail this morning.  But the results
are very surprising to me.

Best regards,
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | President OSGeo, http://osgeo.org

More information about the FWTools mailing list