[Qgis-user] (no subject) (elevation question)

[Greg Troxel]

To the OP: agreement within 11m for a pseudorange solution is entirely
ok.  I don't call that an error; I call it matching.   As an experiment,
find a physical point you can reoccupy, and on 10 days not all at the
same time of day go take a measurement and then tell us the statistics
of the data set.

Nicolas Cadieux <njacadieux.gitlab at gmail.com> writes:

> Interesting but I very much doubt they are not applying a geoid on top
> of the ellipsoid. If they can put maps of the earth on the device,
> they can put a geoid with a posting of 15´x15´!

You would think, but it seems that the geoid model in some receivers is
very coarse.

I've found that with a u-blox F9P (and I think 8 series), the internal
model gives me a value that's about 4m off from EGM2008.  And it's not
an EGM96/2008 thing -- there is some variance (around me) between those,
but far smaller.

So it seemed like the geoid model was nowhere near 15', more like 2
degrees, but I haven't been the least bit rigorous in checking it.

Still, all of this is maybe 4m, which if you aren't doing RTK is really
hard to discern.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 194 bytes
Desc: not available
URL: <http://lists.osgeo.org/pipermail/qgis-user/attachments/20210421/b3983bf8/attachment.sig>