[time-nuts] GPS altitude somewhat wrong?

[Van Horn, David]

Some of that is probably the difference between the geoid (what your surveyed maps report height relative to) and the WGS84 ellipsoid (what your GPS reports heights relative to).  At Boulder that difference is only about 15 meters, though.

Generally with VDOP < 2 and a reasonably modern receiver the accuracy of a GPS altitude measurement should be better than about 20 meters.
I'm not sure if the Thunderbolt counts as such.

How flat is Boulder?  Do you have a proper surveyed elevation of your location?

I do not.

The readout is plausibly right, but it is displaying to the nearest tenth of a millimeter which I thought was somewhat optimistic.  :)