Sent: 7/13/2001 10:38:55 AM

RE: [gpsstash] multiple GPS readings for average

This is a really difficult question to answer fully - partly because I don't
have all the answers. :)

The only way you can figure out how accurate your GPS is, is to have it take
a reading at a spot where the exact GPS location is known. The only place I
know where you can get that, is a 'survey point'. If you're social enough,
you can ask one of those survey guys that you see on the side of the road
sometimes where to find one of those.. I imagine they would be able to give
you an answer. (Actually, this is something I've been considering.)

You can test averaging, for your GPS, by doing the following: Take your GPS
into your backyard, set it down and write down the readings every 30
seconds, for an hour. Do this, at the exact same spot, over several days,
at different times of a day. That should tell you how averaging affects the
reading on your particular GPS.

More things to note:

The GPS manufacturers won't release the algorithms they use to determine
your location. Each manufacturer has its own, proprietary, algorithm.

There are things called 'survey grade' GPS receivers. They're more accurate
and expensive.

An antenna might change the accuracy of your GPS. Only testing will tell
you. There doesn't seem to be any hard data out there. Although it will
definitely help you in leaf cover.

Search for 'GPS averaging' on Google. Lots of links. Many of them are
pre-SA (Selective Availability) being turned off, so be sure you note that.

The best reading I've ever gotten on my Garmin Emap has been where it said
15' of accuracy. That's with about 11 satellites visible, most of them with
a strong signal.

I've seen a chart of someone who did multiple averages with an Emap of a
known location - all the points clustered around a reading that was actually
about 2 meters off the real location. So unless you do all averaging and
figuring for your GPS at a known location you will never really know how far
you're off.

You could get a friend's GPS, put the two units side by side and compare,
although remember: "A man with one watch knows what time it is, a man with 2
watches is never sure."

That's about all I know. Have fun.

> ----------
> From: GC Scout[]
> Reply To:
> Sent: Thursday, July 12, 2001 11:56 PM
> To:
> Subject: [gpsstash] multiple GPS readings for average
> I have often heard the advice that to get a really accurate reading from a
> GPS, you should walk away, come back, take another reading, and repeat at
> least three times.
> I've never seen an explanation for the "walk away" part of this advice. I
> have a suspicion that walking away does nothing more than kill time, which
> is, I think, the real benefit of taking multiple readings. The delay
> between
> readings allows time for transistory effects (e.g., atmospheric
> conditions)
> to change.
> Wouldn't you get the same benefit from sitting down and having a cold one
> between readings rather than walking away and back? After all, if you are
> going to be accused of behavior like a drunken bee anyway, you might as
> well
> have a beer.