765t Calculations

The current Garmin routing algorithm is based on general road classifications and speed limits, not on historical data for specific road segments and week day / time of day patterns. Unless and until Garmin/Navteq add this data to their routing, you can expect to see these kinds of inaccuracies. Certain TomTom models have this capability (called IQRoutes) and they get high ratings from users for routing quality.

TomTom i'm coming to take a look.
Update. Nice but too much money $400 to get what i have now plus IQroutes

I am trying to correspond with Garmin on the routing from my home to a Walmart. If I have "fastest" route set, it takes me a way that amounts to traveling on a road (parallel to my destination road) for a considerable distance more than necessary, making what amounts to a wide (.4 miles) "U Turn" to the left and then back 1.7 miles to my destination ON MY RIGHT. If I change to "shortest" routing, it become a wide (.5 miles) "U Turn" with the return distance being 1.3 miles to my destination ON MY RIGHT.

When I return home it takes me back to the road I would have - on my own from the parallel road I was on -turned LEFT on earlier thn Garmin would have, driven 1.4 miles, then turned RIGHT and driven .7 miles to my destination ON MY LEFT as there is a street light with a left turn signal in front of the Walmart. A slightly longer way I could choose would have been to turn left a little later on the first road, gone .5 miles, turned RIGHT and then make a U-Turn (done by thousands of people a day at that spot going to the Walmart) (and note that I had not said to "avoid" U-turns) and then traveled .5 miles to my destination ON THE RIGHT.

If I had set the routing to "Less fuel" Garmin's routing would have matched my preferred route (as indicated above) EXCEPT that Garmin would have taken me to the point where I would have made the "U-Turn" on my second preferred route and then brought me back to my destination ON MY RIGHT (so Garmin knows that the U-Turn at that point is possible (it's a traffic light).

The "fastest" route is 12 miles (don't think decimals are shown here) and the estimated time to destination was 17:41

The "shortest" route is 11 miles and the estimated time to destination was 16:56.

The "less fuel" route is 10 miles and the estimated time to destination was 16:11.

The "fastest" route takes the most time and is the longest in miles. The "shortest" route is both longer and takes more time than the "less fuel" routing. Go figure!!??

I am wondering if there is something about the arrive ON THE RIGHT that is impacting the routing.

It has been noted by others that the Garmin routing algorithm seems to prefer right turns and tries to avoid left turns that cross over traffic lanes when arriving at a destination. This is just speculation based on observation, as I don't think Garmin publishes the technical characteristics of their routing methods.

Maybe Garmin assumes all lights are green in its time calculation?

LOL i thinks the lights are just not on...Period!!
In this case i gotta hand it to TomTom, i was made aware of what TomTom calls IQRoutes. That's maybe not the complete answer since i don't have one of these baby's to try but certainly a start. Find my Car doesn't work yet either, So far.

It's true Or may be calculated with Linear Line between 2 points. It seems never consider about traffic lights & congestion situations or speed limits.

Especially in NYC here we have traffic lights almost every block & speed limit here is 30 MPH most places. I can't rely on Garmin ETA.

Re: ETA - at least for me, my 765t seems to be accurate for ETA if I drive a solid 10mph over the posted limit the whole time, regardless of zone. This is ok usually on the highway, but pushing it on backroads sometimes is an adventure.

Do the algorithm guys assume that you will be speeding when calculating the ETA?

When I run a route in GPS simulation mode, the speed showing on the screen is typically 3-6 MPH over the speed limit. Therefore the time estimate seems to assume no red lights and a lead foot as well??

With my 765T I have never really checked the beginning ETA against my watch & time at my destination. I'll have to start looking I guess.

Since ETA 'adjusts' itself while on the road, I just 'adjust' my right foot as I see the need

I did notice the other day that on a particular 100 mi trip I took, the 765 said 100 mi & 1hr 49 min, while Mapquest quotes 101.56 miles & 2 hrs 4 min for the identical route. The Mapquest mileage & time were closer to actual.

edit- after posting I checked Google for grins- same routing, 99.5 mi 2hrs 11 min, fyi.

It has been discussed and proven in another forum that many of newer Garmin models, including the 7x5 series, learn from your driving habits and adjust ETA and simulation speeds for given road classifications. If you do a hard reset, these learned values are reset back to the system defaults until it learns new values based on your driving habits.

You can read about it here:
http://www.gpsreview.net/train-your-nuvi-eta/
the discussion is here:
http://forums.gpsreview.net/viewtopic.php?t=6802

thanks for the links!

Excellent info. I now know why the ETA on my newer Garmin is worse than my older one. This now makes sense.

I've owned a TOM TOM in the past and I wouldn't give it too many high fives to the IQ routes. The routes and the recal's were worst off than the garmin most of the time. No device is perfect but the Garmin comes close to the route it picks. Most of it has to do with the maps used by the device. Navteq seems to do better in the US.

Good to know...Thanks

Claudius