What GPS to use?

Rick Onanian wrote:
On Wed, 30 Jul 2003 15:05:08 +0200, Roberto Divia
wrote:
Yes and yes. Garmin sells a special bike holder. The best position seems
to be on the helmet, a "soft" holder from Garmin can be used for that
(heard of, never tried myself).

That never occured to me, but it would probably yield more accurate
results and less signal loss.

However, there, you won't know when you've lost a signal; and also,
it's subject to getting destroyed in an accident.

Not to mention the fact that you can't read the info (e.g. trip computer page)
without taking your helmet off :-(

Mine's been safe on the handlebar.

So is mine ;-)

Ciao,
--
Roberto Divia` Love at first sight is one of the greatest
============= labour-saving devices the world has ever seen.
Mailbox: C02110 CERN-European Organization for Nuclear Research
E-mail: CH-1211 GENEVE 23, Switzerland

Ditto. I have a Garmin eMap and it sucks for elevation. Otherwise it's
pretty cool. An external antenns helps when in the woods.

Actually, the altitude solution from a GPS is less accurate than the
latitude/longitude. This is mostly due to the fact that the GPS system
is up in the sky only. If there was a satellite or two at the center
of the Earth, the 3D mapping would be much more accurate. This has
been discussed in sci.geo.satellite-nav before, and there is an old
(outdated due to SA being turned off) Realaudio file at
http://www.tapr.org/tapr/ra/dcc97.intro.gps.ram if you are really
intrested in the technical details and a good overview of navagation
in general.

The e-Trex Vista has a built in barametric altimeter and magnetic
compass. The altimeter works very well, but I don't believe the
tracklog uses it (the tracklog uses GPS altitude). I still find it
very useful while climbing, though. The big improvement on the Vista
over the others is the large memory available for downloading maps.
For example, I keep most of the street level maps of the state of PA
loaded all the time.

One problem is that due to it bouncing around the countryside for
about 7000 miles on my handlebars, the compass no longer works.
However, I know that Garmin will repair it for a flat fee (basically
they send a new unit). I just don't think it is worth fixing, since I
rarely used it anyway.

Eric

archer wrote in message ...
In article , says...
On Tue, 29 Jul 2003 14:59:06 -0400, Jim Sherman wrote:
A few more questions. Does the Venture calculate elevation? The
comparison chart shows that only the Vista has a barometric altimeter. Or
is that just an advanced version?

Now that you mention it, I believe it does not use an altimeter,
but just figures it from the satellite signal. I don't know just
how accurate this method is.

The altitude accuracy is the same as the horizontal accuracy: typicallt
10 to 30 ft, depending on conditions. It simply determines your position
in 3-dimensional space with respect to the satellites.

....

Chris Malcolm wrote:
I don't have anything to contribute other than a question. A friend has one
of these (I don't even know the brand), and it seems to give wildly
inaccurate elevations (below sea level in the hills). Do some do a better
job of elevation and do any have a useful slope gauge?

Those with barometric altimeters, e.g. Garmin Summit, Vista, MAP76S, &
Geko 301.

I get fine altitudes with Garmin eTrex (variously $99-$125), the original.
Since the page button fell off after a couple years, I bought another and
now have two on the bike. Scotch electrical tape over a small wad of
durable plastic packing peanut material (there are various kinds, you need
a very flexible kind that won't break when you wad it up) fixed the missing
button, so it looks like I have two forever.

They agree about position to within a place in the last digit (might be
handlebar width! and altitude by about ten feet. Altitude will slowly
change but not wildly. Usually it goes slowly up when you pedal slowly
up and down when down. A tree canopy though will ruin it because the
number of satellites seen is greatly reduced.

Barometric altitude can be less accurate due to wind, which GPS altitude
is unaffected by.

I would say that if you bike in the open, so mostly you can see horizon to
horizon, the GPS altitude is fine. It will still change but is better at
detecting a slope than you are.
--
Ron Hardin


On the internet, nobody knows you're a jerk.

Here's a test run, two Garman eTrex's on the handlebars, bike
parked out front, reading position and altitude every five minutes:
(decimal degrees, x's agree, alt feet, lat 40 degrees, average/good view of sky)

11:15 lat xx.xxx08 lon xx.xxx05 alt 1165 lat xx.xxx05 lon xx.xxx06 alt 1166
11:20 lat xx.xxx07 lon xx.xxx02 alt 1152 lat xx.xxx07 lon xx.xxx05 alt 1162
11:25 lat xx.xxx07 lon xx.xxx03 alt 1162 lat xx.xxx07 lon xx.xxx05 alt 1157
11:30 lat xx.xxx06 lon xx.xxx03 alt 1160 lat xx.xxx07 lon xx.xxx04 alt 1162
11:35 lat xx.xxx07 lon xx.xxx04 alt 1157 lat xx.xxx06 lon xx.xxx03 alt 1169
11:40 lat xx.xxx08 lon xx.xxx03 alt 1146 lat xx.xxx06 lon xx.xxx04 alt 1164
11:45 lat xx.xxx07 lon xx.xxx03 alt 1153 lat xx.xxx06 lon xx.xxx04 alt 1163
11:50 lat xx.xxx07 lon xx.xxx03 alt 1154 lat xx.xxx07 lon xx.xxx03 alt 1142
11:55 lat xx.xxx07 lon xx.xxx03 alt 1142 lat xx.xxx06 lon xx.xxx03 alt 1158
12:00n lat xx.xxx08 lon xx.xxx04 alt 1151 lat xx.xxx05 lon xx.xxx04 alt 1162

the variation is slow (ie this captures it well, it's not changing faster)
corresponding to satellite movement and (probably) deficiencies in the refraction
correction, rather than measurement noise. The two eTrex's are different system
releases, current and then previous.

Now, over 45 minutes, you've gone over ten miles, and 20 feet over ten miles
is two feet over a mile, or a pretty good grade estimate.

Probably, if you see fast altitude variation, it's trees cutting out satellites,
which lets the errors in the refraction correction cut in and out at the same speed.
--
Ron Hardin


On the internet, nobody knows you're a jerk.

Ron Hardin writes:

Here's a test run, two Garman eTrex's on the handlebars, bike
parked out front, reading position and altitude every five minutes:
(decimal degrees, x's agree, alt feet, lat 40 degrees, average/good view of sky)

11:15 lat xx.xxx08 lon xx.xxx05 alt 1165 lat xx.xxx05 lon xx.xxx06 alt 1166
11:20 lat xx.xxx07 lon xx.xxx02 alt 1152 lat xx.xxx07 lon xx.xxx05 alt 1162
11:25 lat xx.xxx07 lon xx.xxx03 alt 1162 lat xx.xxx07 lon xx.xxx05 alt 1157
11:30 lat xx.xxx06 lon xx.xxx03 alt 1160 lat xx.xxx07 lon xx.xxx04 alt 1162
11:35 lat xx.xxx07 lon xx.xxx04 alt 1157 lat xx.xxx06 lon xx.xxx03 alt 1169
11:40 lat xx.xxx08 lon xx.xxx03 alt 1146 lat xx.xxx06 lon xx.xxx04 alt 1164
11:45 lat xx.xxx07 lon xx.xxx03 alt 1153 lat xx.xxx06 lon xx.xxx04 alt 1163
11:50 lat xx.xxx07 lon xx.xxx03 alt 1154 lat xx.xxx07 lon xx.xxx03 alt 1142
11:55 lat xx.xxx07 lon xx.xxx03 alt 1142 lat xx.xxx06 lon xx.xxx03 alt 1158
12:00n lat xx.xxx08 lon xx.xxx04 alt 1151 lat xx.xxx05 lon xx.xxx04 alt 1162

the variation is slow (ie this captures it well, it's not changing faster)
corresponding to satellite movement and (probably) deficiencies in the refraction
correction, rather than measurement noise. The two eTrex's are different system
releases, current and then previous.

Looks fairly typical to me. It's the sort of altitude dithering which
a barometric altimeter will greatly reduce.

Now, over 45 minutes, you've gone over ten miles, and 20 feet over ten miles
is two feet over a mile, or a pretty good grade estimate.

Fair enough. Some folk are fussier about altitude profiles and move
slower, e.g., folk climbing hills on foot (and hand), and who might
want to use small ups and downs in altitude as navigational clues.

Probably, if you see fast altitude variation, it's trees cutting out satellites,
which lets the errors in the refraction correction cut in and out at the same speed.

And this can happen if you're stationary near a big tree and
satellites move through its large branches. It can happen if a wind is
moving the tree branches around. Rapid ( few minutes) variation won't
necessarily be observed in any particular half hour, but it's common
enough in a trip that you never know whether it's happened or not. For
example, suppose in a trip at one point in ten minutes you climbed 15
feet. But the trip log shows two such incidents. Which one is the real
one and which is the dither?








--
Chris Malcolm +44 (0)131 650 3085 DoD #205
School of Informatics, Edinburgh University, 5 Forrest Hill,
Edinburgh, EH1 2QL, UK. [http://www.dai.ed.ac.uk/homes/cam/ ]