5 ms·
"The best fix so far was ±1 metres at a very open location using 12 satellites; but accuracy is typically ±5 metres in poorer locations with fewer satellites."
by bainsfather 11y ago
"The best fix so far was ±1 metres at a very open location using 12 satellites; but accuracy is typically ±5 metres in poorer locations with fewer satellites."
That seems very good accuracy. The off-the-shelf GPS I have (https://www.adafruit.com/products/746 https://www.adafruit.com/products/746) is always >= ±5m, even with 360 degree view of the sky.
(the specs for my GPS claim < 3 meters, and also say "(all GPS technology has about 3m accuracy)").
- CONTRARlAN 11y agoCommercial-grade technology has about 3m accuracy, but different methods can yield down to 10cm accuracy: https://en.wikipedia.org/wiki/Global_Positioning_System#Precise_monitoring https://en.wikipedia.org/wiki/Global_Positioning_System#Prec... And then there's a whole segment of survey-grade equipment that can get within a centimeter or two: http://water.usgs.gov/osw/gps/ http://water.usgs.gov/osw/gps/
- jhayward 11y agoThat MediaTek chip is capable of better. Do you have WAAS (SBAS) enabled? 5m is about what you'd expect without the improved orbit/clock/ionosphere corrections. It also pays off to use a good antenna, or at least put a ground plane beneath a mediocre one to block multipath reflections from the surrounding ground objects.
- bainsfather 11y agoI just used the default settings. Didn't know about WAAS - thanks for pointing me to it. From reading some forum posts, it seems you have to explicitly enable WAAS - I'll give that a try and see how much it helps.