[ gpsdrive ] v2.09 versus v2.10: coords off!

[Ross Scanlon]

On Fri, 30 Oct 2009 10:00:30 -0500
Rick Richardson <rickrich at gmail.com> wrote:

> Check out the coords.  V2.09 is right within 1 meters.  v2.10 is WAY OFF,
> especially image 2!
> 
> http://www.rkkda.com/tmp/gpsdrive-209-210-1.png
> http://www.rkkda.com/tmp/gpsdrive-209-210-2.png
> 
> Mapquest aerials showing v2.09 is right...
> 
> http://atlas.mapquest.com/maps/map.adp?zoom=14&formtype=latlong&latlongtype=decimal&latitude=45.054050&longitude=-93.427150&dtype=h
> 
> http://atlas.mapquest.com/maps/map.adp?zoom=14&formtype=latlong&latlongtype=decimal&latitude=45.052666&longitude=-93.423083&dtype=h

Hi Rick

Are these screen shots with downloaded T at H maps or with local rendered mapnik maps?

If they are downloaded T at H then the problem is probably with the scaling (for want of a better word) index in map_koord.txt and someone here who is more familiar with that part of the source should have a look at it or maybe you should log a bug on the sourceforge page.

I'd suggest you try it with a local mapnik installation and see if they are still off, I have no issues with this method of display from gpsdrive but all methods of raster maps show some irregularities when overlayed with poi's.  All cases are because of the scaling index in map_koord.txt

Every raster map will show slightly differently, I'd also suggest that it's very hard to compare the two when you are using two different maps.

Does 2.09 show the same offset when loaded with the maps you have in 2.10?

Does 2.10 show the same offset when at the same scale as 2.09 or vv?

At the moment you are comparing apples with oranges.

-- 
Cheers
Ross