Accuracy of route planners
 

Hi there,

I'm comparing various route planners available on the Internet and
trying to find out the most accurate (time-wise) one.

I have tried these three:

Google Maps (reference)
Map24
Viamichelin

Given travel times vary quite much and sometimes I feel that none of
them is right - not even close.

For example, from Zürich to St. Moritz given times varies between 2:18
and 2:24 (without traffic jams). However, almost every single website
tells that it takes about 3 hours (websites of hotels, official city
website etc). Google Maps says that about 140 km is autobahn (and
about 1-2 km in the city before autobahn). So it should take about
1:20 (less than 110 km/h, speed limit 120). Rest is windy road where
the speed is quite low, I suppose. I don't mind if the difference
between "reality" and route planner is 10 minutes or so but wonder how
it could take 3 hours... :)

Makes you thinking about believability and sensibility of the
programs. Why an earth to put big sums for research if the end-product
is, well, garbage..

I have found out that Map24 (at least) takes tunnels and their lowered
speed limits into consideration (about 60 for Gotthard though limit is
80 but the traffic is quite heavy) but then it gives the fastest time
for Zurich-St. Moritz... :o

Any ideas why there are so substantial differences in certain routes?
Do you have some comparisons between reality and computer generated
stuff?

Cheers.


-kb-

Re: Accuracy of route planners
 

<[email protected]> wrote in message
news:[email protected] oups.com...
Hi there,

I'm comparing various route planners available on the Internet and
trying to find out the most accurate (time-wise) one.

I have tried these three:

Google Maps (reference)
Map24
Viamichelin

Given travel times vary quite much and sometimes I feel that none of
them is right - not even close.

For example, from ZÃŒrich to St. Moritz given times varies between 2:18
and 2:24 (without traffic jams). However, almost every single website
tells that it takes about 3 hours (websites of hotels, official city
website etc). Google Maps says that about 140 km is autobahn (and
about 1-2 km in the city before autobahn). So it should take about
1:20 (less than 110 km/h, speed limit 120). Rest is windy road where
the speed is quite low, I suppose. I don't mind if the difference
between "reality" and route planner is 10 minutes or so but wonder how
it could take 3 hours... :)

Makes you thinking about believability and sensibility of the
programs. Why an earth to put big sums for research if the end-product
is, well, garbage..

I have found out that Map24 (at least) takes tunnels and their lowered
speed limits into consideration (about 60 for Gotthard though limit is
80 but the traffic is quite heavy) but then it gives the fastest time
for Zurich-St. Moritz... :o

Any ideas why there are so substantial differences in certain routes?
Do you have some comparisons between reality and computer generated
stuff?
---------------------------------------------------------------------------------------------------------------

I looked at www.mappy.com. It says 2 hours 27 minutes for your example
route. But none of these mapping programs are precisely accurate. They just
give you a ballpark idea of the drive time. Deviations of 10 minutes or less
are nothing.

I planned a driving trip in northwest Scotland, using www.multimap.com,
www.theaa.com, and www.rac.co.uk. Among these 3 sites, there were wide
variations of driving time, like 2 hours vs. 4 hours. Made it hard to lay
out the trip.


Marianne

Re: Accuracy of route planners
 

On Wed, 25 Jul 2007 01:52:54 -0700, [email protected] wrote:

>Hi there,
>
>I'm comparing various route planners available on the Internet and
>trying to find out the most accurate (time-wise) one.

>For example, from Zürich to St. Moritz given times varies between 2:18
>and 2:24 (without traffic jams). However, almost every single website
>tells that it takes about 3 hours (websites of hotels,

Maybe the hotel overestimates it to discourage late arrivals. I've
found Viamichelin to be quite accurate, but I wouldn't quibble over 15
minutes.

--
Barbara Vaughan
My email address is my first initial followed by my surname at libero dot it
I answer travel questions only in the newsgroup

Re: Accuracy of route planners
 

[email protected] wrote:
> Any ideas why there are so substantial differences in certain
> routes?Do you have some comparisons between reality and computer
> generated stuff?

Reality includes traffic jams, road construction, landslides,
accidents and dawdling tourists in rental cars.

Jens

--
Listening 'All I Wanna Ever Do' by Martin Schmitt

Re: Accuracy of route planners
 

On 2007-07-25, [email protected] <[email protected]> wrote:
>
> Any ideas why there are so substantial differences in certain routes?

6 minutes is hardly a substantial difference in a 2 hour estimate.

Re: Accuracy of route planners
 

On 25 heinä, 23:50, Jesper Lauridsen <[email protected]>
wrote:
> On 2007-07-25, [email protected] <[email protected]> wrote:
>
>
>
> > Any ideas why there are so substantial differences in certain routes?
>
> 6 minutes is hardly a substantial difference in a 2 hour estimate.

Estimate is 2:24 and "reality" is 3 hours. That's like 36 minutes to
me. Or 20 %.

I guess Viamichelin and Map24 are quite right and hotels and other
stuff are bit "exaggerated".

Re: Accuracy of route planners
 

On Wed, 25 Jul 2007 14:46:47 -0700, [email protected] wrote:

>On 25 heinä, 23:50, Jesper Lauridsen <[email protected]>
>wrote:
>> On 2007-07-25, [email protected] <[email protected]> wrote:
>>
>>
>>
>> > Any ideas why there are so substantial differences in certain routes?
>>
>> 6 minutes is hardly a substantial difference in a 2 hour estimate.
>
>Estimate is 2:24 and "reality" is 3 hours. That's like 36 minutes to
>me. Or 20 %.
>
>I guess Viamichelin and Map24 are quite right and hotels and other
>stuff are bit "exaggerated".

Map24 has a thing about Belgrade - it insists it is in Belgium.
--

Martin

Re: Accuracy of route planners
 

On 26 heinä, 01:04, Martin <[email protected]> wrote:
> On Wed, 25 Jul 2007 14:46:47 -0700, [email protected] wrote:
> >On 25 heinä, 23:50, Jesper Lauridsen <[email protected]>
> >wrote:
> >> On 2007-07-25, [email protected] <[email protected]> wrote:
>
> >> > Any ideas why there are so substantial differences in certain routes?
>
> >> 6 minutes is hardly a substantial difference in a 2 hour estimate.
>
> >Estimate is 2:24 and "reality" is 3 hours. That's like 36 minutes to
> >me. Or 20 %.
>
> >I guess Viamichelin and Map24 are quite right and hotels and other
> >stuff are bit "exaggerated".
>
> Map24 has a thing about Belgrade - it insists it is in Belgium.
> --
>
> Martin

Which country did you select and which language? Does it have any
Serbian cities or streets? Did you write "Belgrade, Serbia" or just
Belgrade?

According to Wikipedia, Belgrade is located in:

- Serbia
- Belgium
- Maine
- Minnesota
- Missouri
- North Carolina
- Nebraska
- Texas

There are also 17 Paris' in USA...

Re: Accuracy of route planners
 

On Thu, 26 Jul 2007 06:41:57 -0700, [email protected] wrote:

>On 26 heinä, 01:04, Martin <[email protected]> wrote:
>> On Wed, 25 Jul 2007 14:46:47 -0700, [email protected] wrote:
>> >On 25 heinä, 23:50, Jesper Lauridsen <[email protected]>
>> >wrote:
>> >> On 2007-07-25, [email protected] <[email protected]> wrote:
>>
>> >> > Any ideas why there are so substantial differences in certain routes?
>>
>> >> 6 minutes is hardly a substantial difference in a 2 hour estimate.
>>
>> >Estimate is 2:24 and "reality" is 3 hours. That's like 36 minutes to
>> >me. Or 20 %.
>>
>> >I guess Viamichelin and Map24 are quite right and hotels and other
>> >stuff are bit "exaggerated".
>>
>> Map24 has a thing about Belgrade - it insists it is in Belgium.
>> --
>>
>> Martin
>
>Which country did you select and which language? Does it have any
>Serbian cities or streets? Did you write "Belgrade, Serbia" or just
>Belgrade?

First I didn;t specify a country. It chose Belgium, then I specified the right
country and it drew a map from France.
--

Martin

Re: Accuracy of route planners
 

On 26 heinä, 16:45, Martin <[email protected]> wrote:
> On Thu, 26 Jul 2007 06:41:57 -0700, [email protected] wrote:
> >On 26 heinä, 01:04, Martin <[email protected]> wrote:
> >> On Wed, 25 Jul 2007 14:46:47 -0700, [email protected] wrote:
> >> >On 25 heinä, 23:50, Jesper Lauridsen <[email protected]>
> >> >wrote:
> >> >> On 2007-07-25, [email protected] <[email protected]> wrote:
>
> >> >> > Any ideas why there are so substantial differences in certain routes?
>
> >> >> 6 minutes is hardly a substantial difference in a 2 hour estimate.
>
> >> >Estimate is 2:24 and "reality" is 3 hours. That's like 36 minutes to
> >> >me. Or 20 %.
>
> >> >I guess Viamichelin and Map24 are quite right and hotels and other
> >> >stuff are bit "exaggerated".
>
> >> Map24 has a thing about Belgrade - it insists it is in Belgium.
> >> --
>
> >> Martin
>
> >Which country did you select and which language? Does it have any
> >Serbian cities or streets? Did you write "Belgrade, Serbia" or just
> >Belgrade?
>
> First I didn;t specify a country. It chose Belgium, then I specified the right
> country and it drew a map from France.
> --
>
> Martin

Got it.

You'll find Belgrade if you type Beograd. That's the Serbian name for
Belgrade. Just browse the map with your mouse and you'll notice it.

It's same with Nice. I usually say Nizza but it's Nice in French. If
you type Nizza, it finds something from Italy.

You have to be precise.

Cheers.

Re: Accuracy of route planners
 

On Thu, 26 Jul 2007 07:01:32 -0700, [email protected] wrote:

>On 26 heinä, 16:45, Martin <[email protected]> wrote:
>> On Thu, 26 Jul 2007 06:41:57 -0700, [email protected] wrote:
>> >On 26 heinä, 01:04, Martin <[email protected]> wrote:
>> >> On Wed, 25 Jul 2007 14:46:47 -0700, [email protected] wrote:
>> >> >On 25 heinä, 23:50, Jesper Lauridsen <[email protected]>
>> >> >wrote:
>> >> >> On 2007-07-25, [email protected] <[email protected]> wrote:
>>
>> >> >> > Any ideas why there are so substantial differences in certain routes?
>>
>> >> >> 6 minutes is hardly a substantial difference in a 2 hour estimate.
>>
>> >> >Estimate is 2:24 and "reality" is 3 hours. That's like 36 minutes to
>> >> >me. Or 20 %.
>>
>> >> >I guess Viamichelin and Map24 are quite right and hotels and other
>> >> >stuff are bit "exaggerated".
>>
>> >> Map24 has a thing about Belgrade - it insists it is in Belgium.
>> >> --
>>
>> >> Martin
>>
>> >Which country did you select and which language? Does it have any
>> >Serbian cities or streets? Did you write "Belgrade, Serbia" or just
>> >Belgrade?
>>
>> First I didn;t specify a country. It chose Belgium, then I specified the right
>> country and it drew a map from France.
>> --
>>
>> Martin
>
>Got it.
>
>You'll find Belgrade if you type Beograd. That's the Serbian name for
>Belgrade. Just browse the map with your mouse and you'll notice it.
>
>It's same with Nice. I usually say Nizza but it's Nice in French. If
>you type Nizza, it finds something from Italy.
>
>You have to be precise.

OK! :-)
--

Martin

Re: Accuracy of route planners
 

The biggest problem in time I have found with the on line routing is
that they use the speed limits of the countries they are based in.
So if you use the American google maps you get the American speed
limits on the roads, even then they do place a max limit at 130km.
You get better results using a mapping site that web sites in each of
hte countries and starting from the web site of the country you want
the distance for.

Re: Accuracy of route planners
 

[Sent around 04:30 of July 25 apparently lost in cyberspace.]

[email protected] wrote:
> various route planners . . .trying to find out the most accurate (time-wise) one.

The formula I use, that I find consistently accurate, is to divide the
distance by 70kph.

--
__________________________________________________ _________________
Un San Francisqueño en San Francisco.
http://geocities.com/dancefest/ -<->- http://geocities.com/iconoc/
TouringSFO: http://geocities.com/touringsfo/ <-> IClast @ Gmail.com

Re: Accuracy of route planners
 

On Wed, 25 Jul 2007 01:52:54 -0700, [email protected] wrote:

We took a trip to Cortona on Thursday, and I used Via Michelin to
suggest the route. We basically knew all the roads as far as Perugia,
so we just wanted some route advice from there to Cortona.

The Via Michelin map told us to exit the E45 for the RA6 in the
vicinity of Perugia. Unfortunately, nothing in the vicinity of Perugia
is labelled RA6 and I, the navigator, missed the turn. "RA" is a route
indicator I've never seen in Italy, but I assumed it stands for
raccordo (a connecting road), and when I saw a sign for Raccordo
Bettole, I thought it must be that. However, the sign said to go
straight ahead and the raccordo Bettole was never again mentioned. It
turned out that we should have turned at the indication for "A1
Roma-Firenze". (The A1 was a small circle, which usually means, not a
junction, but a road going towards a junction.)

This is a fault I have noticed before with Via Michelin: they don't
use the terminology found on the ground.

After I got home, I decided to do a review of mapping software to see
if anyone does it better. Here are my results:

the AA.com gave the exact signposting at the exit (although my memory
isn't good enough to confirm the accuracy), and referred specifically
to Raccordo Bettolle-Perugia. Quick site, minimum of stupid questions,
but also no route options, just a check box for "avoid tolls or
congestion charges". Without even entering "Italy", it chose the
correct towns immediately. The route is chose is similar to Via
Michelin's "fastest" route.

The RAC route planner site is very slow and a bit fussy. It chose more
or less the same route as the AA and referred to the exit from the E45
as "Raccordo Perugia-A1", without specific signposting. I'm not sure
this would have been sufficient. This site is If you decide you want
a different type of route, you have to go all the way back to the
beginning, but at least you don't have to re-enter all your
information. I was asked whether I wanted Corinaldo the town, or a
bunch of other stupid options including streets named Corinaldo in
various cities of Italy. Same with Cortona; one of the proposed
choices was a town called "Ruda" in Udine.

Mappy picked a very strange route, that didn't even pass near Perugia.
I went back to the beginning to select "shortest" instead of
"quickest" and had to re-enter everything. It gave the same route
again, which would neither be the shortest nor the quickest. Also,
after entering the towns and "Italy", I was presented a list of
options. "Corinaldo" was the first choice for the departure, but
"Cortona" was the third choice for the arrival. For some strange
reason, the first choice was "Cortina d'Ampezzo".

Map24 was quick, and chose the right towns immediately. It has options
for "quickest" and "shortest", but chose the same route for both,
which might be the shortest, but since it follows mostly two-lane
roads often through hilly country, I doubt that it's the quickest.
(They gave an estimate of 2 hours 10 minutes, which is less than any
of the others, but I don't think it's realistic.) This route was
similar to the Mappy route, and didn't pass near Perugia. I like the
fact that everything is in frames on one page, so you don't have to
backtrack much to change your route specifications.

ViaMichelin is fast and offers lots of choices: quickest, shortest,
most economical, and most scenic, as well as "Michelin's choice",
which was very roundabout and with a longer time estimate than either
the fastest or the more scenic. You can pair these choices with other
preferences, such as avoid tolls, avoid ferries and other off-road
itineraries. If you don't like the route, you can change these options
without going all the way back to the beginning. In fact, I like Via
Michelin best, except for the fact that the route indicators seem to
be theoretical route names, not the signage you will see along the
road.

--
Barbara Vaughan
My email address is my first initial followed by my surname at libero dot it
I answer travel questions only in the newsgroup

Re: Accuracy of route planners
 

On Sat, 28 Jul 2007 17:26:32 +0200, B Vaughan<[email protected]> wrote:

>On Wed, 25 Jul 2007 01:52:54 -0700, [email protected] wrote:
>
>We took a trip to Cortona on Thursday, and I used Via Michelin to
>suggest the route. We basically knew all the roads as far as Perugia,
>so we just wanted some route advice from there to Cortona.
>
>The Via Michelin map told us to exit the E45 for the RA6 in the
>vicinity of Perugia. Unfortunately, nothing in the vicinity of Perugia
>is labelled RA6 and I, the navigator, missed the turn. "RA" is a route
>indicator I've never seen in Italy, but I assumed it stands for
>raccordo (a connecting road), and when I saw a sign for Raccordo
>Bettole, I thought it must be that. However, the sign said to go
>straight ahead and the raccordo Bettole was never again mentioned. It
>turned out that we should have turned at the indication for "A1
>Roma-Firenze". (The A1 was a small circle, which usually means, not a
>junction, but a road going towards a junction.)

Via Michelin provides a map and a graphic of the signs at each turning.
You can see that the road is correctly labeled RA6 in several places
--

Martin