8,153
edits
Changes
no edit summary
I have come to suspect that the 620 has two issues.
* Firstly, Garmin has compromised GPS accuracy for size and battery life. This is then compounded by high levels of smoothing in an attempt to compensate.
* I believe that The 620 (and the newer [[Garmin Fenix 2]] watches have changed from a SiRFstar chipset to a MediaTek chipset. This is backed by the way the newer Garmin (These watches support the pre-caching of satellite position data using an "EPO.BIN" file that is a MediaTek trademark. ) I wondered if problems in the EPO ([http://www.gtop-tech.com/LNG/product/AGPS-Assisted-GPS/Software_Services_07.html Extended Prediction Orbit]) data could be causing some of the problems, so I ran a test without any EPO data. To do this, I did a factory reset and did not connect the watch to the internet while testing. The obvious impact was that the 620 now takes an age to acquire a satellite signal. As you can see from the accuracy above, the 620 appears to be slightly more accurate without EPO data. However, one run I did without EPO data recorded only 19.5 miles on a 21 mile run, and analyzing the file shows a general poor quality rather than a specific section of the run being bad. (If that run is excluded, the 620 without EPO showed Trueness of 2.68% and SD of 5.62%).
I thought that the 620 had a problem with its [http://en.wikipedia.org/wiki/Wide_Area_Augmentation_System WAAS] processing, but I have it on good authority that no Garmin Forerunner has WAAS support. Below are a couple of examples where you can see that it appears that the track has been offset, rather than the 620 simply becoming 'lost', which I'd attributed to WAAS error.
{| class="wikitable"
|[[File:Fenix2 Getting Lost6.jpg|none|thumb|x400px|This GPS track looks reasonable until marker #54, and then the track gets offset, but strangely it stays offset until the last marker.]]
|}
=Polar V800 GPS Accuracy=
As you can see from the numbers above, the V800 is remarkably accurate. The V800 uses the latest SiRF chipset, rather than the MediaTek chipset that has caused so many problems in the [[Garmin 620]] and [[Garmin Fenix 2]]. This SiRF chipset includes satellite prediction to reduce the time it takes to acquire the first satellite lock. This generally works pretty well, but is not as fast as the MediaTek chipset.
=GPS Accuracy and Pace=
[[File:AccuracyAndPace.jpg|none|thumb|500px| A plot of GPS precision against pace. The red line is the correlation.]]
There have been reports of GPS accuracy changing with pace, but as you can see from the graph above, my testing does not show this.
=GPS Accuracy and Sampling Rate=
GPS watches default to recording a sample frequently enough that accuracy is not compromised. However, several devices off the option of recording less frequently to improve battery life at the cost of accuracy. The images below are from the [[2014 Badwater 135]] using the [[Suunto Ambit2 R]] with recording set to one minute intervals. As you can see, accuracy suffers on curves, but is fine on the straights. For a course like Badwater, the one minute recording interval was fine as the course has few turns.
{| class="wikitable"
|- valign="top"
|[[File:GPS Sampling Curve.jpg|none|thumb|x300px|On a curve, the infrequent samples tend to 'cut the corners' and are quite inaccurate.]]
|[[File:GPS Sampling Straight.jpg|none|thumb|x300px|On the straight sections, the one minute sampling does not lose any accuracy.]]
|}
=Device Specific Notes=
For those interested in some of the details of how devices are configured for testing, here are some additional notes.