Garmin 310xt gps accuracy issues

my last HM my 310xt showed 13.59.. based off of it, i started my "kick" a wee bit too soon... I thought mine was off because it was a trail HM though..

Firstly- I wouldn't assume that the Edges are more accurate than the 310.

I had two identical 405's on my wrist and went for a couple of runs and a couple of bike rides with them.  I did a hard reset on both, and set them to the same settings, with the auto pause feature turned off. 

I was amazed at how different they were.  Firstly, the instantaneous pace - which is about a 12 second average, was way off between the two- often more than 1min/mi.  So- I now no longer use the instantaneous pace display.

I found them often to be up to 160 feet off in total distance in the middle of a long run.  And, it wasn't always the same watch that was coming up short.  I had both set to auto lap at 1 mi, and my mile run splits were as close as 3 seconds and as far off as 12 seconds. 

When you take a sharp corner (acute angle), especially in bike mode, sometimes the garmin will cut the corner off pretty severely, depending on exactly when it takes a position reading I guess.   You can sometimes see this on the plots.  I've assumed that that kind of thing has a lot to do with the errors.

These are truly amazing instruments- but their accuracy is not perfect.

How do you soft reset 310xt?

How about soft reset for Edge 500?

In my 3 Twin City Marathons so far (each done with a different Garmin - long story!), I averaged ~0.5 mile longer than 26.2 - this was also true with 2 other standalone marathons I've done with GPS. 

Just part of the zig-zagging involved with running the long course - you will always go longer than 26.2; the more turns on the course, the worse the effect gets.

Cheers, Chris

I went hiking at Half Dome in yosemite. I gained .50 miles in just from sitting down and resting. I have the garmin 310 xt

First of all, we will all be off when running a race no matter what measuring equipment we use. The race distance is measured at the shortest possible points throughout the race coarse, i.e. running every single tangent. This alone will ad quite a bit of distance for us when we run (most of us need to run around people, can't run the tangent as close as measured, etc.).

Second, the GPS is not going to be accurate enough to give exact distance. For example, you are running under an overpass; the GPS sees you entering the tunnel and when you exit, the distance between will be a "best guess estimation" based on the speed you ran before entering the tunnel and the speed you are running when exiting combined with the time laps when you had no signal. You will have errors in a wide open field as well, going from a triangulation between 3 (or more) satellites to another set of satellites....

And does these errors really matter? The race will always be from start line to finish line anyway... same for everyone.... Use the GPS as an indication, not as an absolute tool.

Thanks for all the feedback everyone.  I will try that soft reset.  I completely understand that it will never be completely accurate, but I do think having a device that loses a tenth of a mile every mile is problematic (this was calculated by running with multiple devices and comparing as well as knowing the distance of the route run).