logo separator

[mkgmap-dev] crude routing of my gpsmap 276Cx

From Gerd Petermann gpetermann_muenchen at hotmail.com on Thu Sep 26 10:09:34 BST 2019

Hi Ticker,

hm, yes, it seems impossible to reach the road "Rödderweg" without using a road tagged with access=destination or a highway=path or a highway=track.  So, Garmin could be right to use the shorter route.

My problem is that I never see the long detour shown in Manfreds pictures using Garmin software. I can only reproduce it using brouter in car mode:
http://brouter.de/brouter-web/#map=14/50.6935/6.4034/osm-mapnik-german_style&lonlats=6.425371,50.693678;6.43351,50.715885&profile=car-eco

Gerd

________________________________________
Von: mkgmap-dev <mkgmap-dev-bounces at lists.mkgmap.org.uk> im Auftrag von Ticker Berkin <rwb-mkgmap at jagit.co.uk>
Gesendet: Donnerstag, 26. September 2019 10:37
An: mkgmap development
Betreff: Re: [mkgmap-dev] crude routing of my gpsmap 276Cx

Hi

I've found that my eTrex 30x, asking for a shortest distance
car/motorcycle route, will use a road that has mkgmap:car=no as an
intermediate part of of the route rather than a much longer route using
just valid roads. I've only noticed this where the non-car road is very
near the start/end of the route.

For the fastest time routing, I haven't seen this happen.

I'm not sure what the cut-offs are, but maybe its relate to the Garmin
logic that finds the nearest road to the start/end point and uses that,
regardless of the transport mode flags, to get to/from the full road
network.

Ticker

On Thu, 2019-09-26 at 07:52 +0000, Gerd Petermann wrote:
> Hi Manfred,
>
> I don't know the actual signs and I don't care if the tagging is
> right or wrong. At least the combination of bicycle=yes and foot=no
> on https://www.openstreetmap.org/way/25285589 looks very wrong:
> bicycle=yes
> fixme=check access rules
> foot=no
> highway=tertiary
> maxspeed=30
> maxweight=7.5
> name=Brandenberger Straße
> ref=K 30
>
> Anyway, it seems that mkgmap sets the correct flags for the road, the
> Garmin algo seems to simply ignore the "no-through-route" flag when
> the detour is
> too long. I've tried this with a sample file where the shortest route
> goes through a road with access=destination + highway=residential and
> the longer route
> just has ways with highway=residential. As long as the detour is
> rather small Mapsource takes the longer (allowed) route when using
> the car profile.
> All other profiles seem to always use the shorter route. So, either
> something is wrong in the map data (problem in mkgmap) or something
> is wrong in the Garmin algo.
>
> Please try if you can reproduce the long route using a map created
> with mkgmap r4292 and the default style coming with that version.
>
> Gerd
>
> ________________________________________
> Von: mkgmap-dev <mkgmap-dev-bounces at lists.mkgmap.org.uk> im Auftrag
> von manfred.haiduk <manfred.haiduk at freenet.de>
> Gesendet: Donnerstag, 26. September 2019 09:14
> An: Development list for mkgmap; dd8kq at gmx.de
> Betreff: Re: [mkgmap-dev] crude routing of my gpsmap 276Cx
>
>
>
> For your additional information, i think the tag access=destination
> is to be seen with the additional tag for weight limit, as the street
> is not allowed to be used for cars heavier than 7.5 tons
>
> Von meinem Samsung Gerät gesendet.
>
>
> -------- Ursprüngliche Nachricht --------
> Von: Gerd Petermann <gpetermann_muenchen at hotmail.com>
> Datum: 26.09.2019 08:55 (GMT+01:00)
> An: Development list for mkgmap <mkgmap-dev at lists.mkgmap.org.uk>,
> dd8kq at gmx.de
> Betreff: Re: [mkgmap-dev] crude routing of my gpsmap 276Cx
>
> Hi Manfred,
>
> I tried to reproduce your results with BC but I get a different
> route.
> at the moment I wonder why Mapsource shows the shorter route when
> using car/motorcycle. The ways
> https://www.openstreetmap.org/way/25285586 and
> https://www.openstreetmap.org/way/25285589
> both have the tag access=destination, so my understanding is that
> cars should not be routed through them, but this seems to be ignored.
>   I'll have to find out why.
>
> Gerd
>
> ________________________________________
> Von: DD8KQ <dd8kq at gmx.de>
> Gesendet: Mittwoch, 25. September 2019 14:53
> An: Development list for mkgmap; Gerd Petermann
> Betreff: Re: [mkgmap-dev] crude routing of my gpsmap 276Cx
>
> Hi Gerd
>
> it's really weird. Trying this with MapSource you are right, there is
> no
> long detour. But, then, i tried the same with BaseCamp and you know
> what, BaseCamp produced on my Laptop this detour, see attached
> picture
> Test2.
>
> And just for testing, i tried the same area on a Smartfone running
> the
> Navigator App with OSM maps, and this progromm is routing the same
> detour, see picture screenshot.
>
> So for me there must be something in the OSM Source regarding the
> street
> K30 which will cause the strange routing. But my experience is too
> small
> to figure it out
>
> Am 25.09.2019 um 11:12 schrieb Gerd Petermann:
> > Hi Manfred,
> >
> > No idea. I cannot reproduce the result with the default style in
> > Mapsource. I tried with vehicle car as well as with cycling or
> > pedestrian, I never get the long detour.
> > Can you reproduce it with Mapsource? If yes, please provide more
> > details about the routing settings.
> >
> > ________________________________________
> > Von: mkgmap-dev <mkgmap-dev-bounces at lists.mkgmap.org.uk> im Auftrag
> > von DD8KQ <dd8kq at gmx.de>
> > Gesendet: Dienstag, 24. September 2019 21:58
> > An: Development list for mkgmap
> > Betreff: [mkgmap-dev] crude routing of my gpsmap 276Cx
> >
> > Hi
> >
> > may be that somebody has an idea why the routing of my gpsmap 276Cx
> > is
> > so crude. I'm using almost the standard style to build my own
> > garmin
> > maps. And as the example is mostly in the vicinity of my hometown,i
> > had
> > not experienced this error before, as i need not a gps in my home
> > area.
> > If you look at picture 20190924_173131.jpg this is a routing i
> > would
> > expect to the destination, short and fast. If you look at
> > 20190924_173229.jpg, the destination is only a few meters away from
> > the
> > first one, the routing is nearly three times more than before with
> > the
> > same settings on my device. I would appreciate any advice what to
> > change
> > or any hint where i can look for why this could happen.
> >
> > Btw, if i change the setting to use bicycle routing, the device
> > will use
> > for both destinations the expected way
> >
> > --
> >
> > #####################################################
> >
> >        Viele Grüße und 73 de Manfred Haiduk, DD8KQ
> >        e-mail mhaiduk at t-online.de dd8kq at gmx.de
> >
> > #####################################################
> >
> > _______________________________________________
> > mkgmap-dev mailing list
> > mkgmap-dev at lists.mkgmap.org.uk
> > http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
> >
> --
>
> #####################################################
>
>       Viele Grüße und 73 de Manfred Haiduk, DD8KQ
>       e-mail mhaiduk at t-online.de dd8kq at gmx.de
>
> #####################################################
>
> _______________________________________________
> mkgmap-dev mailing list
> mkgmap-dev at lists.mkgmap.org.uk
> http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
> _______________________________________________
> mkgmap-dev mailing list
> mkgmap-dev at lists.mkgmap.org.uk
> http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
_______________________________________________
mkgmap-dev mailing list
mkgmap-dev at lists.mkgmap.org.uk
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev


More information about the mkgmap-dev mailing list