Arguably though, what matters most is accurate maps and timely ETA's. Who meets that challenge the best? I decided to try and find out. I have, at my disposal, a Garmin Nuvi 3490LMT, a TomTom 740 Live, an iPad 3G and a Samsung Galaxy S2 Android smartphone. Here is who I tested: Google Navigation, Telenav (for Android), Navigon* (for iPad), Mapquest (for Android), Waze (for Android), MotionX Drive, TomTom (TeleAtlas), Garmin (Navteq). All products used were the current versions of the app, or current version of an available downloadable map.
* Note: this test was done prior to Navigon's new app with the optional map update subscription, however, the test did include the most recent map version Navigon made available at the time. I could not test ALK CoPilot because they can't get their new software to run on some (my) US variant of the Samsung Galaxy S2.
There is a particularly challenging route in the Midwest that really separates the men from the boys when it comes to routing algorithms and map accuracy. It is the route from Des Moines, IA to St Louis, MO. Here is what that route looks like:
I picked this route because it provides so many challenges to overcome when determining the best route. The highlighted route is, in my opinion, the best route to take. Why?
- 99% of this route is a 4-lane divided US or State highway at a speed limit of 65 mph, with some limited 55 mph zones near towns.
- There is only 1 through-town section where your speed has to reduce to slower speeds. But it is also a good, and necessary, gas stop along the way.
- It is one of the shortest options when total miles are considered.
- About 60% of this route is known as the Avenue of the Saints. This is a federally designated highway system connecting St Louis, MO to St Paul, MN.
- The Avenue of the Saints starts in St Louis and the route above follows it completely until the route turns directly west halfway between Keokuk, IA and Mt Pleasant, IA.
- Note: The Avenue of the Saints continues directly north from where this route diverts to the west, through Iowa City, IA and eventually up to St. Paul, MN.
- To create this Avenue of the Saints, extensive construction was required over about a decade, to make this route 4 lane.
- As part of this extensive construction, a significant number of highway bypasses were constructed around small towns. What used to be a 2 lane highway directly through many small towns is now a 4 lane highway bypass around nearly all of those towns.
- This route uses almost no freeways until the last 25 miles in the metro St. Louis area.
- There are many tempting alternatives.
- Des Moines to Kansas City to St Louis - it's all freeway.
- Des Moines to Davenport to Peoria to St Louis - it's all freeway.
- Des Moines to Iowa City then south to join this route - it's more freeway at least.
This route, with an actual speed that averaged no more than 5 mph over the posted speed limit took 5 hours and 35 minutes to drive with 2 refreshment stops that included a gas fill-up along the way.
Obviously, I could not use all 8 of these solutions while driving. Instead, while parked at my starting location, I calculated a route to the same final address on all 8 products and immediately recorded the resulting ETA. I did this on all 8 devices within 5 minutes of each other. I calculated the route at approximately 10:00pm CST in hopes of minimizing any bias from traffic reports.
Here is how they did:
- Google Navigation: 6 hours 37 minutes (off by +1 hour 2 minutes)
- Telenav: 6 hours 28 minutes (off by + 53 minutes)
- Navigon: 7 hours 0 minutes (off by +1 hour 25 minutes)
- Mapquest: 6 hours 16 minutes (off by + 41 minutes)
- Waze: 6 hours 34 minutes (off by +59 minutes)
- MotionXDrive: 7 hours 11 minutes (worst result, off by +1 hour 36 minutes)
- TomTom: 5 hours 53 minutes (off by +18 minutes)
- Garmin: 5 hours 44 minutes (off by +9 minutes)
This is a single route. This is not a comprehensive test of different routes. It is not a scientific test. It is simply one real world example that applies to me personally. For this route, on this day, there are some interesting conclusions:
- The so-called advantage of "online" maps isn't an advantage at all.
- Online maps should be more accurate because the latest maps is always available.
- In practice, for this route, every online map provider fails to live up to that claim and has squandered the leverage they could have.
- Online maps were actually some of the worst performers in the test.
- There is a curious difference between Navigon and Garmin - both Navteq customers.
- There could be differences between Navigon and Garmin's routing algorithm that accounts for the some of the difference in time. In my experience, they are pretty close generally, on other routes.
- More likely, we are seeing a real life example of what Navteq gets accomplished in sequential map updates. As I stated above, this test was done before Navigon changed to their new quarterly subscription update option. Therefore, the Navteq maps in their app were older than the Navteq maps in the Garmin Nuvi.
- The crowd-sourced map approach doesn't yield best in class results. Both Google Navigation and Waze are in the middle of the pack when it comes to the route time.
- While Google did select my opinion of the ideal route, their ETA is way too conservative - a common complaint about Google Navigation.
- Crowd-sourced "real world" speeds seem to be having an impact. TomTom IQRoutes, once considered the king of "real world" driving conditions seems to be continuing to hold its own, at least in this limited test.
- Garmin's ETAs were always pretty good. It appears that with their map data now incorporating trafficTrends functionality, that their ETAs might be getting even better. They clearly were spot on in this test.
