Travel Technology - Accuracy of Garmin GPS units?

(Lots of Garmin threads, but I think this question is a bit different...)

Does anybody here know how accurate the latest version of the small Garmin units are? I'm thinking mainly about the entry-level wear-on-your-wrist consumer product that costs about $150.

I've heard from some folks that GPS is very accurate, but a friend of mine has one and doesn't trust it at all. He claims he's tested it over a well-known and repeatedly-certified 5000 meters and his unit is always off by several hundred meters. (He wore it for a marathon once and it told him he had gone 27.4!)

Are these little gadgets worth anything - yet - or should I wait a couple of years until they get a lot better? Not that I need military-level accuracy: I'd just like to know that if the thing says I've run 10 miles, then I've really gone about 10 miles give or take a few meters - not give or take a mile.

Can't speak for the Garmin units specifically since I own two Magellans, but make sure you look for one that supports WAAS. When my unit is able to lock on to a WAAS signal it is noticably more accurate.

GPS isn't a very good tool for tracking exactly how far you have travelled. It generally has accuracy of just a few feet, but it has a strong tendency to occasionally (and often very briefly) get a weird reading that is off by as much as a few hundred feet. When you look at the accumulated distance travelled figure, you will almost always see results that are higher than the actual distance travelled. WAAS does not eliminate this problem and it isn't a problem for the more traditional uses because the erroneous readings are usually so brief that they don't cause any real concern.

People who assure you of their accuracy are correct, but don't use them the same way as your friend. I don't think any units will provide dead-on accuracy for accumulated distance travelled, particularly if any of that distance was under tree cover. 200 meters over 5000 is not suprising to me.

People who assure you of their accuracy are correct, but don't use them the same way as your friend. I don't think any units will provide dead-on accuracy for accumulated distance travelled, particularly if any of that distance was under tree cover. 200 meters over 5000 is not suprising to me.

I concur - GPS is great for measuring your position (to a reasonable accuracy), and your velocity (to a reasonable accuracy). Unfortunately measuring distance travelled along an arbitrary path is an activity which tends to accumulate the errors in the above activities. My GPS has given me substantially different numbers for distance elapsed travelling from A to B and the distance between A and B where the route between A and B is a straight line!

The issue is compounded where the speed of motion is slow relative to the GPS error (such as walking or running), or where changes in direction are frequent. Most GPS units use proprietary smoothing algorithms to determine what your route has been, and these can lose detail.

If you want to get into the details of any of this stuff, look at either of these two public newsgroups:

sci.geo.satellite-nav (http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&oe=UTF-8&safe=off&group=sci.geo.satellite-nav)

alt.satellite.gps (http://groups.google.com/groups?group=alt.satellite.gps)


I've always been able to get technical questions answered there.

Well, what we are touching on here is that the example given above is just a poor application for a GPS. That doesn't mean it would be any less useful for you, depending on what you are actually going to do with it once you get it. Can you give us an example of what you want to do?

Well, what we are touching on here is that the example given above is just a poor application for a GPS. That doesn't mean it would be any less useful for you, depending on what you are actually going to do with it once you get it. Can you give us an example of what you want to do?

I'm just looking for a reasonably accurate measure of real-time pace and distance traveled over an arbitrary course. That's the whole point for me - I want to make the course arbitrary, thereby freeing myself from measured trails, tracks, etc. But I still want to know how far I've gone with a decent degree of confidence.

I went to the website to look at the specific product: it's called the Garmin Forerunner 201 and it is marketed for exactly this purpose. Looks like I could get one for about $125 (new in box, shipped).