[mkgmap-dev] R: Progressing patches

Mon Aug 17 13:36:57 BST 2009

--- Lun 17/8/09, Mark Burton <markb at ordern.com> ha scritto:

> Da: Mark Burton <markb at ordern.com>
> Oggetto: Re: [mkgmap-dev] R:  Progressing patches
> A: mkgmap-dev at lists.mkgmap.org.uk
> Data: Lunedì 17 agosto 2009, 11:28
> 
> Hi Marco,
> 
> > Hi Mark, I did not follow very well last discussions.
> I agree in enabling remove-short-arcs by default, but with
> no min length (i.e. min lenght=0).
> > 
> > Could you please explain why arcs shorter than 5m
> should be collapsed?
> 
> Arc lengths appear to be encoded in units of 16 feet. That
> is equal to
> approx 4.8m. i.e. an arc that is less than 4.8m is encoded
> as length 0. So if we want all arcs to have a non-zero
> length, they
> must be at least 4.8 metres, rounding up to an integer
> gives 5m.

Very strange about the 16 feet units. As far as I kew garmin utilizes MKSA

And even if the 16feet units is true, maybe during the encoding a rouding is appied:
- an arc <2.4m is encoded as zero 16feet units
- between 2.5m and 7.6m is one 16feet units.
- etc.

At last, in my experience with mkgmap never had problems with --remove-short-arcs without min arc length (i.e. remove only zero length arcs)

Ciao, Marco.

> 
> Personally, I am not convinced that having zero length arcs
> really
> makes a difference. However, some people report that the
> minimum length
> needs to be greater than 0 otherwise routing is broken in
> some places.
> 
> Cheers,
> 
> Mark
> _______________________________________________
> mkgmap-dev mailing list
> mkgmap-dev at lists.mkgmap.org.uk
> http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
>