[mkgmap-dev] Routing issues
From GerdP gpetermann_muenchen at hotmail.com on Fri Feb 8 19:31:47 GMT 2013
Hi Fabian, yes, --ignore-maxspeed is useful if you want to use the maps for biking Gerd Fabian S. wrote > Thanks guys. Getting it from overpass now as a whole. Still, as long as > the base map is there, it will sometimes use the streets from the base > map. But it's okay for me to delete the base map. Sad that there isn't a > nicer solution. > > Another problem that I have is that the routing is not giving enough > "attention" to the highway, i.e. it uses streets that go parallel to the > highway sometimes. Sometimes the route even goes on and off the highway. > Could that be because I'm using the --ignore-maxspeeds flag? I've got > that one from the guy at garmin.openstreetmap.nl who uses it for some > reason. > > > On 09/02/13 06:11, Apollinaris Schöll wrote: >> In general it is not possible to download data this way and combine >> it. There is a high chance that the boundary nodes are missing and >> that is where routing brakes. To some extend Garmin devices will fix >> missing connections and fall back to the basemap routing to find a >> rout between tiles. >> It is highly recommended to download an extract for the whole area or >> cut it out from a planet or a big enough extract. Then use splitter to >> split it into 2 part. >> If the island is really small you could download the whole from XAPI >> or Overpass >> >> >> >> >> >> On Fri, Feb 8, 2013 at 10:37 AM, < > mkgmap-dev at .mailrange > > <mailto: > mkgmap-dev at .mailrange > >> wrote: >> >> I think it makes a lot of sense. Maybe I should clarify it further: >> - If I calculate a route that is completely within the top part of >> the island, everything's fine. >> - If I calculate a route that is completely within the bottom part >> of the island, everything's fine. >> - If I calculate a route that goes across the two parts of the >> island, the base map streets seem to be the only ones that provide >> a way across the "border". With gmapbmap.img that results in our >> crappy route, without gmapbmap.img that results in no route at all. >> >> >> >> On 09/02/13 05:30, Jaakko Helleranta.com wrote: >>> Hmm . That sounds interesting/odd (to me, a non developer). Are >>> you implying that you loose routing capability to _OSM_ data's POIs? >>> >>> I wonder what information there is in that gmapbmap.img if that's >>> the case... But more so: I've always understood that the unit >>> routes purely with the OSM-derived .img. .. But as noted, I'm no >>> dev person and my technical understanding of this is limited. >>> >>> -Jaakko >>> -- >>> > jaakko@ > <mailto: > jaakko@ > > * Skype: >>> jhelleranta * Mobile: +509-37-269154 <tel:%2B509-37-269154> * >>> http://go.hel.cc/about.me >>> >>> >>> On Fri, Feb 8, 2013 at 12:44 PM, < > mkgmap-dev at .mailrange > >> <mailto: > mkgmap-dev at .mailrange > >> wrote: >>> >>> Thanks Jaakko, I'm pretty sure that the problem is indeed >>> related to gmapbmap.img (the "wrong" street has the same >>> shape as one of the streets that show up when I disable my >>> gmapsupp.img). >>> >>> However, when I delete the gmapbmap.img, I loose routing >>> capabilities to many of the POIs and addresses! Is there >>> another way? Maybe edit the gmapbmap.img to delete the >>> streets (it's only a few)? >>> >>> There is another theory I have: As you can see in our wget >>> commands, we are downloading the island in two parts. It >>> seems that the routing has issues only when going between the >>> two parts. Maybe the overlap isn't big enough? Does anybody >>> know the minimum required overlap? >>> >>> >>> >>> >>> On 09/02/13 04:08, Jaakko Helleranta.com wrote: >>>> Checked the video: Looks _exactly_ like my basemap problem. >>>> -Jaakko >>>> >>>> On Fri, Feb 8, 2013 at 12:00 PM, Jaakko Helleranta.com >>>> < > jaakko@ > <mailto: > jaakko@ > >> wrote: >>>> >>>> I didn't look at the video but what you describe sounds >>>> like a pesky problem I have experienced before. >>>> My solution was to delete the global basemap(?) from the >>>> device memory. It's an .img file of about 50MB. Solved >>>> the problem. >>>> Cheers, >>>> -Jaakko >>>> -- >>>> > jaakko@ > <mailto: > jaakko@ > > * >>>> Skype: jhelleranta * Mobile: +509-37-269154 >>>> <tel:%2B509-37-269154> * http://go.hel.cc/about.me >>>> >>>> On Fri, Feb 8, 2013 at 11:36 AM, Fabian S. >>>> < > mkgmap-dev at .mailrange > >>> <mailto: > mkgmap-dev at .mailrange > >> wrote: >>>> >>>> >>>> we are having a problem with the maps: In some >>>> cases, the routing uses streets that don't exist! >>>> >>>> >>>> >>>> >>>> _______________________________________________ >>>> mkgmap-dev mailing list >>>> > mkgmap-dev at .org > <mailto: > mkgmap-dev at .org > > >>>> http://lists.mkgmap.org.uk/mailman/listinfo/mkgmap-dev >>> >>> >>> _______________________________________________ >>> mkgmap-dev mailing list >>> > mkgmap-dev at .org >>> <mailto: > mkgmap-dev at .org > > >>> http://lists.mkgmap.org.uk/mailman/listinfo/mkgmap-dev >>> >>> >>> >>> >>> _______________________________________________ >>> mkgmap-dev mailing list >>> > mkgmap-dev at .org > <mailto: > mkgmap-dev at .org > > >>> http://lists.mkgmap.org.uk/mailman/listinfo/mkgmap-dev >> >> >> _______________________________________________ >> mkgmap-dev mailing list >> > mkgmap-dev at .org > <mailto: > mkgmap-dev at .org > > >> http://lists.mkgmap.org.uk/mailman/listinfo/mkgmap-dev >> >> >> >> >> _______________________________________________ >> mkgmap-dev mailing list >> > mkgmap-dev at .org >> http://lists.mkgmap.org.uk/mailman/listinfo/mkgmap-dev > > > _______________________________________________ > mkgmap-dev mailing list > mkgmap-dev at .org > http://lists.mkgmap.org.uk/mailman/listinfo/mkgmap-dev -- View this message in context: http://gis.19327.n5.nabble.com/Routing-issues-tp5748615p5748640.html Sent from the Mkgmap Development mailing list archive at Nabble.com.
- Previous message: [mkgmap-dev] Routing issues
- Next message: [mkgmap-dev] Routing issues
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the mkgmap-dev mailing list