[mkgmap-dev] Question on routing difference

Tue Jun 7 14:56:27 BST 2022

Garmin routing seems to be like it uses only major roads (based on route 
class) to figure out any longer routes and uses small roads only in the 
beginning and end to get to the larger routes.

This is a pain in bicycle routing because it can direct to large roads 
even when wanted. It seems if there is no valid route, garmin uses what 
is has. It can direct bicycle to a bicycle=no road like a major trunk or 
motorway. (That is at least how my cycling garmin devices definately 
work, even in cycle mode).

If you increase road class of smaller roads, they get taken into account 
too. But the route calculation and recalculation can be VERY long. When 
I tried elevating cycleways to similar class with motorways, a 17km 
route took on some earlier devices something like 20mins to 
calculate/recalculate the route.

I've never tried toll roads. Petrhaps I could misuse that in my style to 
make large roads that bicycle is not allowed to be "toll" and make 
garmin avoid them better. This message gave me an idea to try out. I've 
no idea how well that would work as there are practically no toll roads 
where I live.

On 5/29/22 5:17 PM, Gerd Petermann wrote:
> Hi jan,
> 
> maybe my routing profile for OFM bike is different?
> 
> Not sure what Minko recommends today. Mine says "Faster Time", Standard  Elevation Mode, only road type avoidance is for "Roll  Roads".
> When I remove the toll roads avoidance the route is different and follows the major road.
> 
> Gerd
> 
> 
> ________________________________________
> Von: mkgmap-dev <mkgmap-dev-bounces at lists.mkgmap.org.uk> im Auftrag von jan meisters <jan_m23 at gmx.net>
> Gesendet: Sonntag, 29. Mai 2022 16:07
> An: Development list for mkgmap
> Betreff: Re: [mkgmap-dev] Question on routing difference
> 
> Hi Gerd,
> 
> here OFM lite gives the same unwanted result as OFM full :-(
> 
> Jan
> 
>> Am 29.05.2022 um 14:54 schrieb Gerd Petermann <gpetermann_muenchen at hotmail.com>:
>>
>> Hi Jan,
>>
>> the artifical way would be a highway=residential, not path. Anyhow, I tried to reproduce the different routing results with the mentioned change in the OFM lite style
>> but found no difference, the wanted route is calculated for both versions.
>>
>> Gerd
>>
>> ________________________________________
>> Von: mkgmap-dev <mkgmap-dev-bounces at lists.mkgmap.org.uk> im Auftrag von Gerd Petermann <gpetermann_muenchen at hotmail.com>
>> Gesendet: Sonntag, 29. Mai 2022 14:10
>> An: Development list for mkgmap
>> Betreff: Re: [mkgmap-dev] Question on routing difference
>>
>> Hi Jan,
>>
>> not sure if you would find it with that id, since it would be an artificial way. Don't have time now, will look into this later.
>>
>> Gerd
>>
>> ________________________________________
>> Von: mkgmap-dev <mkgmap-dev-bounces at lists.mkgmap.org.uk> im Auftrag von jan meisters <jan_m23 at gmx.net>
>> Gesendet: Sonntag, 29. Mai 2022 14:07
>> An: Development list for mkgmap
>> Betreff: Re: [mkgmap-dev] Question on routing difference
>>
>> Hi Gerd,
>>
>> do you mean another routable line?
>> All (routable) highways are echotagged in my style atm, but I can´t find 27463238 twice.
>>
>> Jan
>>
>>
>>> Am 29.05.2022 um 09:16 schrieb Gerd Petermann <gpetermann_muenchen at hotmail.com>:
>>>
>>> Hi Jan,
>>>
>>> might be the oneway:bicycle=no on way 27463238 which can create an additional path in the opposite direction.
>>>
>>> Gerd
>>>
>>> ________________________________________
>>> Von: mkgmap-dev <mkgmap-dev-bounces at lists.mkgmap.org.uk> im Auftrag von jan meisters <jan_m23 at gmx.net>
>>> Gesendet: Samstag, 28. Mai 2022 20:15
>>> An: Development list for mkgmap
>>> Betreff: [mkgmap-dev] Question on routing difference
>>>
>>> Hi all,
>>>
>>> I´m using an altered copy of the OFM style and therefore sometimes compare the results.
>>> One routing difference I found I was able to lead back, but the cause I don´t understand at all.
>>>
>>> My test-route should prefer the small residential „Altengabengäßchen“ over the primary „Viktoriastrasse“.
>>> Latest OFM does, my version not since I removed {add bicycle=yes} from this line:
>>> highway=path & surface ~ '(paved|asphalt|sett|concrete|paving_stones|paving_stones:30)' & access!=no & access!=private & vehicle!=no { set highway=pedestrian; add bicycle=yes; add motorcar=yes; }
>>>
>>> But unfortunately there is no path or pedestrian in the test-route, nor is it an option to use one.
>>> Anyone has an idea how this path>pedestrian rule could affect routing on residential/primary?
>>> Same happens when I replay the change with the original OFM.
>>>
>>> Up-to-date osm.pbf, route from BC and screenshots are here: https://files.mkgmap.org.uk/download/556/test_route.zip
>>>
>>> Thanks
>>> Jan
>>> _______________________________________________
>>> mkgmap-dev mailing list
>>> mkgmap-dev at lists.mkgmap.org.uk
>>> https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
>>
> 
> _______________________________________________
> mkgmap-dev mailing list
> mkgmap-dev at lists.mkgmap.org.uk
> https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
> _______________________________________________
> mkgmap-dev mailing list
> mkgmap-dev at lists.mkgmap.org.uk
> https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
> 