You are right, it was quite obvious. First of all, I made an error in
my e-mail since I was actually using lon_0=10 and not 0. But my main
problem was that proj was returning a wrong y-coordinate. The error is
the same as I mentioned in my first e-mail. Using cs2cs instead gave
me the correct coordinate, and those coordinates makes my map correct
:-)

Thanks for all you help Ed!

Espen

On 22/01/2008, Ed McNierney <[EMAIL PROTECTED]> wrote:
> Espen -
>
> Those EXTENT values look odd.  Your projection's X-origin is still at 0 
> degrees longitude.  If your X coordinates are 10 degrees East, then the 
> projected coordinates (in meters) shouldn't be about 37,000.  That would mean 
> you're about 37 km east of the prime meridian, and you're not.  At 60 degrees 
> North latitude one degree of longitude is 55.8 km, so if you're 10 degrees 
> East you should have an X coordinate of about 558,000 meters (a little more, 
> since you're more than 10 degrees East).
>
> When I use PROJ or CS2CS to project your values I get exactly the same Y 
> values but very different X values:
>
> cs2cs +from +init=epsg:4326 +to +proj=eqc +lat_ts=60 +lon_0=0 +x_0=0 +y_0=0 
> +ellps=WGS84 +datum=WGS84 +no_defs +units=m
> 10.6661 59.9155
> 593672.41       6669762.95 0.00
> 10.6824 59.9213
> 594579.66       6670408.60 0.00
>
> These X values seem reasonable for your projection.  The command line you 
> give for PROJ shouldn't actually work as given - the quotes get in the way.  
> You should be using a command like the one above to CS2CS, or:
>
> proj +proj=eqc +lat_ts=60 +lon_0=0 +x_0=0 +y_0=0 +ellps=WGS84 +datum=WGS84 
> +no_defs +units=m
>
> for the conversion.  I'm a little puzzled as to how you're getting those 
> results.  Could you try both the PROJ and CS2CS command lines above to 
> double-check, and report the version of PROJ you're using?  There's probably 
> something obvious here but I'm not seeing it.
>
>      - Ed
>
> Ed McNierney
> Chief Mapmaker
> Demand Media / TopoZone.com
> 73 Princeton Street, Suite 305
> North Chelmsford, MA 01863
> [EMAIL PROTECTED]
> Phone: +1 (978) 251-4242
> Fax: +1 (978) 251-1396
>
> -----Original Message-----
> From: UMN MapServer Users List [mailto:[EMAIL PROTECTED] On Behalf Of Espen 
> Isaksen
> Sent: Tuesday, January 22, 2008 1:45 PM
> To: [email protected]
> Subject: [UMN_MAPSERVER-USERS] proj=eqc with change of latitute
>
> Hi!
>
> This question originates from this earlier thread:
>
> http://lists.umn.edu/cgi-bin/wa?A2=ind0801&L=mapserver-users&D=1&T=0&P=37923
>
> But since I learned a lot during that thread and because the thread is
> full of all kinds of questions now I would like to ask a very specific
> question again in a new thread.
>
> I have also checked out the archives at this list and the PROJ list. I
> have found one similar question on this list in 2005:
>
> http://lists.umn.edu/cgi-bin/wa?A2=ind0504&L=MAPSERVER-USERS&P=R75005
>
> Here is my question:
>
> I have set up a mapfile with projection parameters like this:
>
> EXTENT 10.6661 59.9155 10.6824 59.9213
> PROJECTION
>     "init:epsg:4326"
> END
>
> From this mapfile I get a map over an small area in Oslo, Norway(I
> cannot publish the data since the data is not free)
>
> Then I change my map file to this:
>
> EXTENT 37074.96 6669762.95 37982.21 6670408.60
> PROJECTION
>     "+proj=eqc +lat_ts=60 +lon_0=0 +x_0=0 +y_0=0 +ellps=WGS84
> +datum=WGS84 +no_defs +units=m"
> END
>
> The extent is calculated by using PROJ like this:
> proj "+proj=eqc +lat_ts=60 +lon_0=0 +x_0=0 +y_0=0 +ellps=WGS84
> +datum=WGS84 +no_defs +units=m" < test.txt
>
> where test.txt consists of the geographic coordinates in my first map file.
>
> This mapfile creates a map about 18000 meters north of my first
> mapfile. This is similar to the post from 2005. Longitude seems to be
> correct.
>
> I do not understand why Mapserver does not create approximatly the
> same map in both map files? Is it not possible to use PROJ to convert
> coordinates like this? If I set lat_ts=0 I do of course get the same
> map, but that is not the projection I would like to use.
>
> And if I cannot do it like this, how can I calculate the shift in the
> y-coordinate(about 18000 meters in this example)?
>
> Kind regards,
> Espen
>

Reply via email to