Choosing a GPS

NO, I never pointed a rocket launcher at innocent people just cos I don't like burgers and chips (oh
and democracy) !!!

"geoff" <[email hidden]> wrote in message "]news:[email hidden]...

Quoted message said:

In message <[email hidden]>, yup <[email hidden]> writes

Quoted message said:

Andy Mabbett Reply to [my first name] [at] pigsonthewing.org.uk USA imprisons children without
trial, at Guantanamo Bay:

Why dont you keep this [censored] to yourself ? !

Quoted message said:

Lucky it's not you this time eh ?
--
geoff

Andy Dingley said:

Quoted message said:

Quoted message said:

A surprising number have large inaccuracies in their conversion to GB OS grid refs though,

Inaccuracies, yes. Large, no.

1/2km is a bit too big, IMHO.

Can't remember the model, this is a friend's (a Garmin, I think) bought just a few years ago. It
was enough to put us on the wrong side of a woodland when out walking.

Sounds like he was using the wrong datum.

Paul
--
The October Project 2003
wildwales.fsnet.co.ukoctober.htmlOpen ↗
wilderness-wales.co.ukwilderness-wales.co.ukOpen ↗
photosig.comuserphotosOpen ↗

"Andy Dingley" <[email hidden]> wrote in message
"]news:[email hidden]...

Quoted message said:

Paul Saunders said:

Andy Dingley said:

A surprising number have large inaccuracies in their conversion to GB OS grid refs though,

Inaccuracies, yes. Large, no.

1/2km is a bit too big, IMHO.

Can't remember the model, this is a friend's (a Garmin, I think) bought just a few years ago. It
was enough to put us on the wrong side of a woodland when out walking.

--

Wrong set-up, or very old unit. From the tracklog of my Garmin Vista you can usually see which side
of the road I walked.

Jim Hutton

Andy Dingley said:

On Tue, 21 Oct 2003 00:42:48 +0100, "Paul Saunders" <[email hidden]> wrote:

Quoted message said:

Quoted message said:

Andy Dingley wrote:

Quoted message said:

Quoted message said:

Quoted message said:

A surprising number have large inaccuracies in their conversion to GB OS grid refs though,

Quoted message said:

Quoted message said:

Inaccuracies, yes. Large, no.

Quoted message said:

1/2km is a bit too big, IMHO.

Quoted message said:

Can't remember the model, this is a friend's (a Garmin, I think) bought just a few years ago. It
was enough to put us on the wrong side of a woodland when out walking.

With that size of error it was your friend's set-up of the GPS that was faulty, not the GPS. Google
will reveal lots of folk complaining of similar problems, the problem being diagnosed, the
complainer being told how to fix it, and the complainer reporting it fixed.

If it was easy to use a GPS correctly everyone would be doing it :-)

--
Chris Malcolm [email hidden] +44 (0)131 651 3445 DoD #205
IPAB, Informatics, JCMB, King's Buildings, Edinburgh, EH9 3JZ, UK
["]http://www.dai.ed.ac.uk/homes/cam/]

"Andy Dingley" <[email hidden]> wrote in message
"]news:[email hidden]...

Quoted message said:

Paul Saunders said:

Andy Dingley said:

A surprising number have large inaccuracies in their conversion to GB OS grid refs though,

Inaccuracies, yes. Large, no.

1/2km is a bit too big, IMHO.

Can't remember the model, this is a friend's (a Garmin, I think) bought just a few years ago. It
was enough to put us on the wrong side of a woodland when out walking.

I've met people on mountains with a Garmin Etrex (the basic yellow one) reporting a position around
a km wrong.

Answer - incorrect setup. The instruction manual is at fault, telling people to use a particular
datum. Quite possibly the correct setting for the USA, but wrong for UK OS maps.

I changed the setup for them, and the position reported was where we were, reporting to a higher
accuracy than any of us could measure from the maps we held.

Nigel. (not a GPS owner).

URL: said:

In article <[email hidden]>, Andy Dingley
<[email hidden]> writes

Quoted message said:

Paul Saunders said:

Andy Dingley wrote:

> A surprising number have large inaccuracies in their conversion to GB OS grid refs though,

Inaccuracies, yes. Large, no.

1/2km is a bit too big, IMHO.

Can't remember the model, this is a friend's (a Garmin, I think) bought just a few years ago. It
was enough to put us on the wrong side of a woodland when out walking.

That was not due to an error in the GPS conversions to OS grid. The datum transformations are
accurate to a few metres (<10).

Most likely it was due to either poor satellite geometry / obstruction or user error in having the
wrong datum set.

The most common error with setting up a Garmin for use with the OS is to select British Grid
position format but not change the datum to the required Ord Srvy GB from the default WGS-84. That
will lead to an extra 100m of error which is consistent. People also commonly miss read the grid
ref reported by Garmins until they realise that it is a 10 figure ref.

Yes I've accidently been caught out in the past by that oversight. In my case I was using Swiss grid
with the wrong datum! The latest versions of the software in Garmin units like the Summit and also
the Geko automatically select the most appropriate datum when you switch the grid setting.

Mike <URL:http://www.path.cam.ac.uk/~mrc7/>
--
o/ \\ // |\ ,_ o Mike Clark <\__,\\ // __o | \ / /\, "A mountain climbing, cycling, skiing, "> ||
_`\<,_ |__\ \> | immunology lecturer, antibody engineer and ` || (_)/ (_) | \corn computer user"

jake said:

Except, of course, if you have purchased Memory-Map software and OS maps to go with your
new Garmin.

Then you need to have the datum in your Garmin set to WGS84.

No you don't. Garmin units always send and receive data in WGS84 format, no matter what the datum is
set to. Setting the OS datum simply affects the coordinates that are displayed on the unit, not how
they are sent to the software.

I don't know about Memory Map, but with Ozi Explorer I set the upload/download datum to WGS84, so
that the data is transferred correctly, but set the map datum to Ordnance Survey, so that the data
is correctly shown on the map.

Setting your Garmin to WGS84 will give you incorrect readings.

Paul
--
The October Project 2003
wildwales.fsnet.co.ukoctober.htmlOpen ↗
wilderness-wales.co.ukwilderness-wales.co.ukOpen ↗
photosig.comuserphotosOpen ↗

Quoted message said:

In article <[email hidden]>, jake <[email hidden]> writes

Quoted message said:

Quoted message said:

<[email hidden]> writes [snip]

Quoted message said:

The most common error with setting up a Garmin for use with the OS is to select British Grid
position format but not change the datum to the required Ord Srvy GB from the default WGS-84.
That will lead to an extra 100m of error which is consistent. People also commonly miss read the
grid ref reported by Garmins until they realise that it is a 10 figure ref.

That caught me out too.

XY 12345 BNG 67890

equates to XY 123678 and BNG doesn't mean bearing :-)

Quoted message said:

Quoted message said:

Except, of course, if you have purchased Memory-Map software and OS maps to go with your
new Garmin.

Then you need to have the datum in your Garmin set to WGS84.

Not for the transfer of waypoints, routes and tracks as these use the Garmin Protocol which always
uses WGS-84 no matter what datum the user has selected in their GPS.

However if you are using a program which takes the real-time NMEA information from the GPS to
display the position on a map you may need to set the GPS datum to WGS-84 in order to have the
position displayed correctly.

Oh yes! Been there done that. BTTS :-) In the end I figured out how to convert a OSGB NMEA plot
(from Cetus) into WGS 84 in Ozi. I was using a spreadsheet to convert the Cetus output to Ozi format
and had to change the header of the Ozi .plt file to specify Ord Svy Grt Btn. Ozi then automatically
recalculated all the points for me and saved it as WGS 84 when I closed it.
--
Phil Cook looking north over the park to the "Westminster Gasworks"

jake said:

In message <[email hidden]>, Paul Saunders <[email hidden]> writes

Quoted message said:

jake said:

Except, of course, if you have purchased Memory-Map software and OS maps to go with your new
Garmin.

Then you need to have the datum in your Garmin set to WGS84.

No you don't. Garmin units always send and receive data in WGS84 format, no matter what the datum
is set to. Setting the OS datum simply affects the coordinates that are displayed on the unit, not
how they are sent to the software.

I don't know about Memory Map, but with Ozi Explorer I set the upload/download datum to WGS84, so
that the data is transferred correctly, but set the map datum to Ordnance Survey, so that the data
is correctly shown on the map.

Setting your Garmin to WGS84 will give you incorrect readings.

Quoted message said:

Yes. Seems your right.

Just checked. My Garmin Map Datum is set to 'Ord Srvy GB'.

Now when memory-map displays the position where the cursor is sitting, it always appends '(WGS84)'

So a waypoint on a map is SP 75562 54701 (WGS84), but when down-loaded is displayed on the Garmin
as SP 75561 54704.

Hmm ....... must try and understand all this one day ;-)

Datums datums datums...

Does Memory Map have a user selectable display datum? I can get Oziexplorer to display positions in
all sorts of datums but it really makes sense to display an OS grid reference in OSGB datum.
--
Phil Cook looking north over the park to the "Westminster Gasworks"

jake said:

So a waypoint on a map is SP 75562 54701 (WGS84), but when down-loaded is displayed on the Garmin
as SP 75561 54704.

Actually that doesn't sound like WGS84, that sounds like the conversion to OSGB inaccuracy that Andy
Dingley mentioned.

I've tried inputting a grid ref in WGS84 on my GPS and on OziExplorer, then changing the datum to OS
and they both produce slightly different results. I think Ozi is more accurate because the
conversion in a GPS is relatively simple.

I have two highly accurate lists of trig point waypoints in the UK, produced by the OS themselves,
one in WGS84 Lat/Long format, the other in OS BNG format. When I enter the WGS84 waypoint (in either
my GPS or OziExplorer) then switch to OS BNG, the result is different to the OS BNG coordinates (the
OS use a much more complex conversion). The difference is usually only a few metres, sometimes more,
but always less than 10 metres.

Paul
--
The October Project 2003
wildwales.fsnet.co.ukoctober.htmlOpen ↗
wilderness-wales.co.ukwilderness-wales.co.ukOpen ↗
photosig.comuserphotosOpen ↗