Re: Garmin GPSmap 76C Review

From: Dale DePriest (Dale_at_gpsinformation.net)
Date: 07/28/04 

Date: Wed, 28 Jul 2004 08:37:23 -0700

Hans-Georg Michna wrote:

> On Wed, 28 Jul 2004 10:12:36 +0200, "Magne W. Mathisen"
> <ms.mathisen@c2iSP.AMnet> wrote:
>
>
>>I am doing some testing on this subject now (My 60 CS is driving along
>>on its own right now sitting on my desk). The first preliminary results
>>indicate that your procedure works. I am running the 3.43 beta firmware
>>with Euro City Select v6 (same as you I think). Routing set for
>>"Car/Motorbike" and "Avoid Highways".
>>
>>I think you have a small typo on your website (and in some of your
>>postings). Jack is testing with US City *Navigator* v5 *not* US City
>>*Select* v5. There might be a difference there... So if someone could
>>test it with US City *Select* and / or Euro City *Navigator*...
>
>
> Magne,
>
> first of all, thanks for testing! I have already corrected the
> mistake. Your name is now in
> http://www.michna.com/gps.htm#GARMIN_GPSMAP_76C as well. If you
> don't like that, please let me know and I'll take you out again.
>
>
>>During my testing I also found a few other things:
>>
>>First: Mapsource (runing v 6.3) works the same way, i.e. keeps you of
>>highways for 10 to 20 km.
>
>
> Interesting. I didn't try that. They all seem to use the same
> routing algorithm.
>
>
>>Second: Routing for "Truck" does not avoid highways at all (at least in
>>my preliminary tests). This is true for both the 60CS (ver. 3.43 beta)
>>and Mapsource 6.3.
>
>
> Also interesting. Good to know.
>
>
>>Third: Routing for "Bicycle" aviods highways all the time (tested on a
>>106 km route).
>
>
> Try a much longer one, like 600 km. You may find that you're
> re-routed onto a freeway after a couple of hundred kilometers.
>
> Somebody mentioned here that this may only be a limitation of
> the routing algorithm that can only handle up to 50 turns and
> therefore has to make a simpler route on very long or complex
> ones.
>
> All my tests have been done with the following setting, which is
> apparently not the default:
>
> MENU, MENU, Setup, ENTER, Routing, ENTER, Follow Road Options,
> ENTER, Calculation Method, ENTER, Best Route, ENTER.
>
> I just tried to force it by asking for a route from one remote
> part of Munich to a remote part of Berlin (some 600 km along the
> freeway) by bicycle, avoiding highways, and found that the
> device routes me partly over freeways, using 42 turns. When I
> calculate the same route on MapSource 6.3, the route is very
> long, much longer than the shortest route, bends back and
> involves long detours, using 124 turns. When I change to
> Car/Motorcycle, I also get 124 turns (probably coincidence), but
> a much more straight route. When I loosen the conditions and
> allow routing over highways, I get a good route with some 35
> turns for Car/Motorcycle.
>
> In short, the routing algorithm is quirky and defective. We can
> only hope that it works reasonably well in most common
> situations. What a pity that Garmin can't have a correct routing
> algorithm!

For routing there is no correct algorithm for anybody!. Even the human
mind which is much more complex than a GPS cannot always produce the
best results. There are many refinements and if you have a main frame
processor you could probably determine an optimum route under most
conditions but with current mapping data even this would be flawed by
the incorrect or misleading data. To even think that there is a
"correct" algorithm that Garmin should have produced is flawed thinking
in the current state of the art. I am sure that 20 years form now we
will all look back at the primitive stuff of today. It is not an easy
task to determine the best route under all conditions and get an answer
in a reasonable waiting period.

The best algorithm I have seen, by the way, is a cost driven approach
using pc board routing software which was adapted to road use. It was
developed by intergraph and ran on their unix box for fire department
and police department dispatching.

Dale

Dale
>
> By the way, I also found that Car/Motorcycle avoids the smallest
> residential streets, while Bicycle does not. So if you want the
> really really shortest route, set your vehicle type to Bicycle.
> Not sure if that sometimes routes via bicycle-only routes
> though.
>
>
>>I use "Truck" routing sometimes to give me a route that *prefers*
>>highways to avoid a shorter, and maybe even a bit faster route, on small
>>and bendy roads.
>
>
> Good to know how all this works.
>
> Hans-Georg
> 

-- 
     _      _     Dale DePriest
    /`) _  //     http://users.cwnet.com/dalede
  o/_/ (_(_X_(`   For GPS and GPS/PDAs

Relevant Pages

  • Re: Valid Routing Query
    ... The route you describe for a journey from Woking to Slough route ... Exeter, Bristol or Swindon would be much higher than the cost of your ... however based on the online routing ... other maps to allow a complete tracing. ...
    (uk.railway)
  • RE: Fax routing
    ... I understand you could not route your ... Open Server Management ... E-mail incoming routing method" ...
    (microsoft.public.windows.server.sbs)
  • Re: What is the default precedence: local-route, static-route, OSPF-route?
    ... Routing Example #1 ... - EIGRP summary route ... the one with the lowest administrative distance value is ... This administrative distance of a static route can be ...
    (comp.dcom.sys.cisco)
  • Routing in the network :-)
    ... Itojun and I had played off and on ... routing information. ... So AT&T gives me the default route to IP-A1 ... up to FreeBSD.net and AT&T's network went down.. ...
    (freebsd-arch)
  • Re: Routing in the network :-)
    ... Itojun and I had played off and on ... routing information. ... So AT&T gives me the default route to IP-A1 ... up to FreeBSD.net and AT&T's network went down.. ...
    (freebsd-arch)