NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
From: Geoffrey Kolbe
Date: 2017 Feb 26, 23:43 -0800
Back in 2008 I was doing some experimental work on alignment of Egyptian pyramids and wanted to know if Google Earth was as accurate as it pretended to be. Google Earth (at that time) gave lat/long to a precision of four decimal places, which is equivalent to about +/- 0.25" or about 7m in latitude and 4m in longitude (at a latitude of 55 degrees, where I live).
I picked a number of points around the North of England, like corners of parking lots or waste bins in road laybys, where it would be easy to identify as precise spot on the ground which could be located on GE despite the resolution limits, and set off with a Garmin GPS III receiver to see how accurate GE was.
I allowed the Garmin GPS III receiver to average for 10 minutes at each location. The Garmin gives (I still have it) lat/long to five decimal places. The bottom line was that it agreed with GE to within 3 metres (10 feet) on average.
Is this accuracy true world-wide? Subsequent work in Eygpt found that GE could 100 metres out in some locations. That was back in 2008 and no doubt the good people at Google have up'd their game since then. But I suspect that for the "developed world" GE accuracy is as good as a $100 GPS.
Geoffrey Kolbe