[Mapserver-users] Can't seem to get the projection correct

Tue Feb 18 13:56:12 PST 2003

Dan -

Use merc, not tmerc.  Tmerc is the Transverse Mercator, which (although
it's more popular) is not what you want.

	- Ed

Ed McNierney
President and Chief Mapmaker
TopoZone.com / Maps a la carte, Inc.
73 Princeton Street, Suite 305
North Chelmsford, MA  01863
Phone: (978) 251-4242  Fax: (978) 251-1396
ed at topozone.com

-----Original Message-----
From: Martin, Daniel [mailto:DMartin at erac.com] 
Sent: Tuesday, February 18, 2003 4:43 PM
To: mapserver-users at lists.gis.umn.edu
Subject: RE: [Mapserver-users] Can't seem to get the projection correct

Thanks Ed, I'm tinkering with Mercator now (or at least I think I am).
Not
having much luck actually - but I think you are right.  I'm just not
very
experienced with projections, so working with them is a matter of trial
and
error for me.  I'm assuming you mean Mercator as in proj=tmerc not
proj=utm,
though I wouldn't know better either way.  I've basically been trying
options from the PROJ list, http://mapserver.gis.umn.edu/doc36/proj.html

Most of the time I end up with an error, a blank map, or a map that
looks
like a rolled newspaper.  I'll get it right eventually.

Thanks again,
Dan

> -----Original Message-----
> From: Ed McNierney [mailto:ed at topozone.com]
> Sent: Tuesday, February 18, 2003 2:49 PM
> To: Martin, Daniel; mapserver-users at lists.gis.umn.edu
> Subject: RE: [Mapserver-users] Can't seem to get the 
> projection correct
> 
> 
> Daniel -
> 
> As I mentioned before, you are probably accustomed to seeing 
> maps in the
> Mercator projection.  Rather than fiddling with the image size, you
> might want to look into having MapServer generate Mercator output
> images.
> 
> The Mercator projection stretches the Y-dimension as compared to the
> geographic projection.  The distance between parallels of latitude
> increases as you move away from the equator.  At a 
> Missouri-ish latitude
> of 40 degrees North, one degree of latitude is 1.414 times taller in a
> Mercator projection than in a geographic projection.  This would mean
> that if your horizontal (longitude) scale were, say, 7 pixels per
> degree, you would expect your vertical (latitude) scale to be 
> 7 * 1.414
> = 9.9 pixels per degree.
> 
> Since this happens to be exactly what we see in the MapInfo image you
> sent (which is roughly centered on Missouri), I think it's 
> pretty likely
> that both your memory and your MapInfo maps are in the Mercator
> projection.  If you set up MapServer to create Mercator 
> output images I
> suspect you'll match your MapInfo results without having to fudge the
> image size (which isn't really correct anyway because the 
> vertical scale
> distortion is nonlinear and varies with the latitude).
> 
> 	- Ed
> 
> Ed McNierney
> President and Chief Mapmaker
> TopoZone.com / Maps a la carte, Inc.
> 73 Princeton Street, Suite 305
> North Chelmsford, MA  01863
> Phone: (978) 251-4242  Fax: (978) 251-1396
> ed at topozone.com
> 
> 
> -----Original Message-----
> From: Martin, Daniel [mailto:DMartin at erac.com] 
> Sent: Tuesday, February 18, 2003 3:17 PM
> To: mapserver-users at lists.gis.umn.edu
> Subject: RE: [Mapserver-users] Can't seem to get the 
> projection correct
> 
> I want to thank everyone for all the responses.
> 
> Daniel Morissette hit the nail on the head.  MapInfo and Mapquest are
> indeed
> doing some matter of x-scale adjustment to make maps "look better".  
> 
> I know this isn't scientific, but my maps in MapInfo look 
> like I expect
> them
> too.  For instance, my home state of Missouri that I've seen on maps
> thousands of times "looks" like I know it to look in MapInfo.  In
> MapServer
> it "looks squished".  I don't know how to express it any 
> different than
> that.  One way I've adjusted for it is to force a new aspect ratio in
> the
> <img > tag on the HTML template to "squish" the image back.  
> But I hate
> that
> solution.  It's not a good solution in my book.
> 
> I'm going to take a look into some things mentioned and see 
> if I can get
> anywhere.  I'll be sure to let you know how they turn out.
> 
> Thanks again,
> Dan Martin
> 
> 
> 
> > -----Original Message-----
> > From: Daniel Morissette [mailto:morissette at dmsolutions.ca]
> > Sent: Tuesday, February 18, 2003 12:10 PM
> > To: Martin, Daniel
> > Cc: mapserver-users at lists.gis.umn.edu
> > Subject: Re: [Mapserver-users] Can't seem to get the 
> > projection correct
> > 
> > 
> > "Martin, Daniel" wrote:
> > > 
> > > Sorry to correct you, but that's not true - at least not 
> > since I've worked
> > > with MapInfo (version 5 through 7). That may be an optional 
> > configuration,
> > > but by default MapInfo doesn't change the aspect ratio of 
> > the data with the
> > > window.  By default, the data's horizontal extent stays the 
> > same as the
> > > window is resized, but the vertical changes to maintain the 
> > projection.
> > > 
> > 
> > You're correct, I was wrong on that, I should have checked first.  I
> > found the option that's causing this effect... open your 
> lat/lon map,
> > then select Map/Options, the "Distance/Area Using:" radio box 
> > is set to
> > Spherical.  Change the projection to "Non-Earth", and the
> > "Distance/Area" will switch to cartesian, and then your map 
> will look
> > exactly as it does in MapServer.  
> > 
> > I'm not sure what this spherical distance option does exactly but it
> > seems that MapInfo (and MapQuest) use this to adjust the 
> > x-scale of the
> > map to make it look better, that's why the map looks 
> > different, this is
> > not a projection effect, it's a map scaling effect.  
> Unfortunately at
> > this point MapServer supports only maps with x-scale == y-scale (or
> > square pixels if you like) so you won't be able to reproduce 
> > that unless
> > you can define a projection that produces a similar effect.
> > 
> > -- 
> > ------------------------------------------------------------
> >  Daniel Morissette               morissette at dmsolutions.ca
> >  DM Solutions Group              http://www.dmsolutions.ca/
> > ------------------------------------------------------------
> > 
> _______________________________________________
> Mapserver-users mailing list
> Mapserver-users at lists.gis.umn.edu
> http://lists.gis.umn.edu/mailman/listinfo/mapserver-users
> 
_______________________________________________
Mapserver-users mailing list
Mapserver-users at lists.gis.umn.edu
http://lists.gis.umn.edu/mailman/listinfo/mapserver-users