16. Routing Engine Performance

All the bells and whistles aside, what really matters on any GPS is the quality of the maps, and how intelligently the GPS calculates the path from point A to B..

The most common complaint I hear about GPS navigation is "the GPS took me on a crazy route", followed closely by: "that definitely was not the best/shortest way."

The two most important components of any GPS navigation system are the routing engine and mapping accuracy. The routing engine is the software that lets the GPS decide which route it should choose to get from point A to point B. Many people falsely assume that simply having the same maps installed on two different GPS devices will result in the same routing choices. Not so, as there is a great deal of math, personal preferences, and economics involved when your GPS begins calculating the "best" route.

Not surprisingly, this also one of the most difficult and complex aspects for GPS manufacturers to get right, and is the aspect of GPS navigation that is least well understood by consumers. Nevertheless, routing engine quality should be at the top of any GPS buyer's checklist.

The two largest mapping data providers are NAVTEQ and Tele Atlas . NAVTEQ used to have the advantage in mapping quality in North America. However, Tele Atlas has recently improved its US database, and both companies are now roughly equal in quality. The mapping provider shouldn't be a determining factor in routing quality, as both NAVTEQ and Tele Atlas provide excellent maps. However, it IS important to consider how often a GPS manufacturer releases map updates. Garmin and TomTom release yearly map updates, while Magellan, LG, Navigon, and others release mapping updates on a much less regular basis. Virtually all GPS manufacturers charge a map update fee.

However, there is a lot more to a routing engine than just the mapping data provider. How a given GPS will determine the "best" route is not only the result of the mapping data, but is also the result of a given manufacturers decision on how many additional data points, or attributes, to factor into the calculation. For example, does the GPS know if certain roads closed after hours or on the weekend? That information is available to GPS manufacturers, but at an additional cost. Similarly, does the GPS know what the average rate of speed is for a given road? That information will also cost a manufacturer more to include.

Statistically speaking, more accidents happen when making left turns as opposed to right turns -- is the GPS unit factoring that into the route? Each company takes all this data and compiles it in different ways, making unique decisions about what data is more important than others. The number of attributes used varies depending on how much money the manufacturer is willing to spend on their mapping data. For example, a full NAVTEQ map database that includes all available attributes can cost the manufacturer twice as much as a basic set of attributes.

All Magellan and Garmin GPS units use NAVTEQ mapping data, while TomTom uses Tele Atlas.

Test #1: Actual Road Test

To test the RoadMate 1430's routing engine, I road tested the GPS on an 800+ mile road trip from New York City to Raleigh, N.C.

I also installed a nuvi 660 in the car, and compared the routing decisions, ETA accuracy, traffic alerts, missed turn re-routing, and GPS signal reception quality on both units.

Overall I was very pleased with the RoadMate's routing decisions, and on several occasions made better routing decisions than the Garmin nuvi. For example, when leaving New York City, the RoadMate 1430 preferred smaller freeways that don't allow commercial vehicle traffic, while the nuvi wanted me to drive along the larger major highways. Following the Magellan RoadMate's planned route easily saved me an hour by avoiding lots of commercial vehicle traffic.

From a pure routing perspective, the RoadMate did a better job choosing the best path. However, the RoadMate took longer to acquire a GPS signal, and exhibited odd behavior when I missed a turn. When I would miss a turn, the Garmin would almost instantly re-calculate, and would efficiently plot a course to get me back on track. The RoadMate, on the other hand, had a tendency to demand U-turns, or lengthy circular paths. In other words, instead of re-calculating based on the new path, the RoadMate would insist on doubling back to the original course. During these moments, I was glad I had the nuvi as a backup.

The RoadMate also had occasional difficulty resuming the route if I turned off the GPS to stop for food or gas. When powered back on, the RoadMate hung for several minutes until it re-calculated my position.

The RoadMate tended to over estimate the ETA. The Garmin was much more accurate in its estimated arrival times. I suspect this is largely due to the fact that Garmin's GPS units "learn" your driving behavior, and, over time, become more accurate in their ETA calculations.

Based on actual real-world driving, I found the Magellan RoadMate better at route calculation than the Garmin nuvi 660. However, the RoadMate's routing performance was tempered by some disconcerting performance and reliability issues (poor re-routing, GPS fix time, and occasional hangs).

Test #2: Simulated Routes

I also performed a bench-test, entering three different trips of varying distance in three different regions of the United States, and compared the RoadMate's results to a Garmin nuvi 760, and a TomTom GO 730. All three test units had the latest available maps installed, and the most current firmware as of this writing.

Our first test took us along the Northeast, starting in Southbury, CT and ending in New York City:

Magellan's RoadMate 1430 and Garmin's nuvi 760 choose essentially identical routes. The Garmin nuvi is slightly more aggressive in its estimated drive time, but both GPS units do a good job routing, announce which side of the street our destination is on, and ensure that we arrive with the destination on our right.

TomTom's GO 730 has us arriving with the destination address on the opposite (left) side of the street. We would then either have to make an illegal U-turn or go around the block to arrive at 135 Central Park West on our right. TomTom also reports distances in yards instead of feet, which is less commonly used here in the United States.

Yahoo! Maps (not shown in the table above) chooses a route identical to the RoadMate and Garmin, but estimates the trip will take 3 minutes longer than Garmin estimates (1 hour 22 minutes). Google Maps also chooses the same route as the Garmin nüvi and Magellan RoadMate.

Winner of Test #1: It's a tie between the RoadMate 1430 and nuvi 760.

The first test was approximately 80 miles in length and took us from Connecticut to New York City.

For the second test, I started out at Fry's Electronics in San Jose, California and entered a destination address at 300 Grove Street in San Francisco (about 46 miles in distance):

Again the Magellan RoadMate and Garmin nuvi choose virtually identical routes, and the TomTom GO chooses the least effective route. While Magellan and Garmin both route us along the 101 highway from San Jose to San Francisco, TomTom's GO 730 instead routes us along highway 880 North to 80 West -- an odd choice that adds around 4 miles (~6 minutes) to the journey, assuming there's no traffic 880 (which there usually is). Even worse, TomTom again has us arriving on the wrong side of the street, with our destination address on the left.

Both the Magellan RoadMate 1430 and the Garmin nüvi 760 arrive with our destination address on the right side of the street, and both units announce what side of the street our destination is on.

Winner of Test #2: It's a tie between the RoadMate 1430 and nuvi 760.

For the third test, I chose a short, 3-mile trip from Beacon Street in Boston to Cambridge, Massachusetts:

The results of test #3 are consistent with the previous two tests: TomTom's GO 730 chooses a route that's about 30% longer than Magellan's Garmin's. Garmin chooses a route that's slightly longer in distance than the RoadMate's.

Magellan's RoadMate just slightly edged out Garmin's nuvi 760 in this test, and TomTom's GO 730 again comes in third.

TomTom's GO 730 unit performed least well in all three routing engine tests, producing the longest routes in all three tests, and putting us on the wrong side of the street at our destination address.

I've received many questions asking why it's important to arrive on the right side of the street. In more rural areas, it likely won't make much difference if you arrive with your destination on the right or left. However, when driving in cities along wide avenues, such as in New York City, it's very important to arrive on the same side of the street as your destination. Otherwise you'll need to make an illegal U-turn or drive around the block.

Regarding the estimated drive times, some readers have commented that these tests unfairly penalize TomTom for more conservatively estimating the drive time, and reward Garmin for unrealistically aggressive (short) drive times. I disagree with these assertions, as I have actually driven many routes with all three devices connected and compared the routing choices made by each.

Only the Garmin nüvi "learns" from your own specific driving behavior, and adjusts estimated arrival times accordingly. That means that the longer you drive with a Garmin nüvi GPS, the more accurate the arrival times will become. TomTom and Magellan's devices, however, will always product the same estimated trip times.

All three tests show Magellan's RoadMate 1430 has a solid routing engine, performing as well or slightly better than a similarly priced Garmin model. TomTom's GO 970 (with TomTom's newest maps installed) came in 3rd place in all three routing tests.