Re: Magellan RoadMate and Garmin 2610 compared

Curtis wrote:

> JAC <[EMAIL PROTECTED]> wrote in message news:<[EMAIL PROTECTED]>...
>
> > On 1 Dec 2003 16:47:29 -0800, [EMAIL PROTECTED] (Curtis) wrote:
>
>
> > > > From further testing it appears Garmin has bowed to commercial
> >
> > > pressure. IF you have two routes. Route one being the direct route and
> > > the other being through the long way business district it will always
> > > choose the business district. I ran a couple of test driving and it
> > > always picks the business district.
> >
> > Utter nonsense.  I've used the SPIII since the week it has come out,
> > then the 2610 since it came out.  This just simply isn't true.
>
>
> To futher pound the point home and put it to bed.
>
> http://www.creeperscobra.com/ny.jpg    routed through business
> district when it should have continued straight.
>
> http://www.creeperscobra.com/nj.jpg     routed on DIFFERENT road
> through business district. When it should have continued straight on
> SAME ROAD.

Seems to me you are reaching very generalized conclusions based on a small 
data sample.  Deciding that it will "always choose the business district" 
requires far more than a few test cases, as does your previous conclusion 
that it isn't able to do overpasses correctly.  Since JAC and others see 
plenty of cases where the routing around business districts and overpasses 
is done correctly it's clear that your conclusions need some qualifiers.  I 
don't think anyone would disagree with the conclusion that the routing 
sometimes has errors.  But concluding that it *always* routes incorrectly 
will bring disagreement from those who have seen it work correctly in most 
cases.

When I first got MetroGuide 4.01 for my eMap I checked the locations of a 
couple nearby restaurants.  Both were off by over a mile from their actual 
locations.  I could have concluded that the whole POI database was 
worthless, but I continued checking and found that most of it is quite 
good.  In fact in actual use it has correctly led me to nearly a hundred 
restaurants on trips with only a few errors and those were because there 
had been a recent change of ownership. My initial impression based on my 
local area was not at all representative of the quality of the data in 
other parts of the country.
> OBVIOUSLY,  GARMINS programming does this as navtech doesn't do the
> routing.

Not at all obvious since the route finding depends on the map data.  In 
your first example I note that the highway '8' symbol is located on the 
business route you'd prefer to avoid and also that the route you'd prefer 
is rather curved and may therefore be slightly longer than the 'Main St.' 
route.  Both of these characteristics are part of the NavTech map data and 
may well lead the routing algorithm to pick the business district routing 
even though it's slower in reality.  BTW, MG4.01 using TeleAtlas data 
routes the way you prefer and also shows the main highway to be straighter 
and therefore clearly shorter than using Main St.  I'm not sure which map 
company has the more accurate depiction of the actual road path.

In the second example the chosen route uses the road that's drawn with the 
thicker blue line - indicating a more major street.  This again would be 
based on the NavTech map data and would indicate to the routing algorithm 
that such a road is generally to be preferred over minor streets which are 
drawn with thinner lines.  I've driven on Mt. Airy Rd. and know that in 
this particular case it's a more major street than how it's shown on the 
map - but that's a choice made by NavTech based on their data sources, not 
Garmin.