[mkgmap-dev] Memory problems since 1245
From Clinton Gladstone clinton.gladstone at googlemail.com on Tue Oct 6 11:08:18 BST 2009
On Tue, Oct 6, 2009 at 9:29 AM, Du Plessis, Bennie <Bennie.DuPlessis at sappi.com> wrote: > > Like Apollinaris says: --generate-sea takes very long. > And it doesn't work well with tiles, and depends on --ignore-osm-bounds > which buggers intertile routing. > So I compiled --generate-sea with a style with nothing but the > coastline, which ran for 1.5 hr and gave me a lovely map with the land & > sea seperated but nothing else. The plan is to use this in a mapset if I > want to have a blue sea. Almost like an overlay, but maybe an underlay? This is the approach I am experimenting with: I create a map with a lower draw priority which includes only coastline, and use the --generate-sea option. Since the coastline does not change very much, this map does not need to be generated very often. Also, since the map is not relevant for routing, and is much smaller than normal maps, it does not necessarily need to be split. Of course, the other maps need to be transparent for this to work. I've tried this out with a map of Ontario Canada, and it seems to work quite well, with the exception of the usual generate-sea problems (some flooded islands, etc.). I've tested it on a Nuvi and an eTrex. Cheers.
- Previous message: [mkgmap-dev] Memory problems since 1245
- Next message: [mkgmap-dev] Memory problems since 1245
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the mkgmap-dev mailing list