GPS Error? - why different distances

[neilmny]

On a walk yesterday on the "front" face of Mt Dandenong, my GPS measured 10.3km total distance the terrain is steep,
ascent was about 430 metres over 4 km and there is lots of trees.
My walking companions GPS measured around 1 km more (I can't remember the exact figure).
This also happened on a walk on Mt Stirling a while back where his GPS measured further than mine,
from vague memory it was about the same proprotion of difference.
Does anyone have any ideas as to why this would happen?

[perfectlydark]

Was either gps in pants pocket? I wondered if the rise/fall of legs could potentially affect data, bit I dont think most commercial gps are that sensitive anyway. Thats all I got?

[tastrax]

All depends on a number of things - Brand, model etc and also the settings that you have set, plus if you saved the track on the unit. A saved track might be less than the actual track covered as many GPS's will "clean up the data for you" and take out minor deviations etc.

If they were the same brand and model then I would checking the settings on each. Give us a few more details and might be able to give a bit more of an insight into the discrepancy.

[GPSGuided]

The recording mode chosen also have an effect amongst other settings and HW spec differences. Further, even if you are standing still, your GPS plot will still wonder in a resolution circle.

[photohiker]

Did you record a track on both units?

If you did, overlay them on a map to see if there are accuracy issues between the two units and against the map. What does a third party app say each track distance is? (it could be the calculations in the GPS, not the accuracy of the locations)

Are both units set to record at the same interval? Perhaps the variation is as simple as a rounding error, although a 1 in 4 error seems a lot...

Agree with tastrax. Garmin's condense the track when saved, but keep the whole track if it is logged. This shouldn't effect the live display at the end of the walk though.

[phan_TOM]

No doubt a number of things affect the accuracy of a GPS but I think the accuracy setting used has a significant effect on the recorded distances (and also the amount of storage space used). I find with mine I get a longer recorded distance when I am using the highest accuracy setting and from my understanding it's to do with the unit measuring the distance between each point that's recorded to provide a running total. This isn't a big deal if you're travelling in a straight line but basically, if you're going around a bend it's going to chop off the corners, Here's a picture to give you an idea of what I mean

[GPSGuided]

although a 1 in 4 error seems a lot...

I read the difference was closer to 10%. As you suggested, comparing the two recorded tracks would easily identify the source of variation.

[neilmny]

Thanks for the replies, it is appreciated.

We both carried them in a leg pocket so they were sort of under the same installation influences.
We both saved our tracks at a couple of points and the variation was there from the first save.
Mine is an Etrex Vista Hcx and the other a Garmin but not sure of the model
something like a GPSMAP 72s?????
I will compare settings and the resolution setting is probably the culprit.
I have mine set on Record Method - Auto and Interval - Normal not sure what my offsider has his set at.
I'll check this and see what goes.

Thanks to all again. Neil.

[GPSGuided]

When talking of resolution, one really is talking about the performance spec of the unit, involving the electronics and antenna sensitivity etc. All will affect the wonder, most noticeable when standing still.

[photohiker]

When talking of resolution, one really is talking about the performance spec of the unit, involving the electronics and antenna sensitivity etc. All will affect the wonder, most noticeable when standing still.

Yes. and also the orientation of the GPS. I have seen significant variations depending on which side of my pack the GPS is attached to when walking the same track, as well as return trips down the same tracks where the GPS is facing into the valley or into the hillside. Conceivably, there would be differences if one was carried on the right leg pocket and the other on the left...

10% was right btw.

[neilmny]

When talking of resolution, one really is talking about the performance spec of the unit, involving the electronics and antenna sensitivity etc. All will affect the wonder, most noticeable when standing still.

A GPS is a radio receiver and is subject to all the various performance influences that a radio receiver
is subject to.
Resolution has nothing to do with antenna specifications only band width, sensitivity
and reception gain pattern is effected by an antenna.
Then after that the receiver sensitivity and bandwidth are effected by the actual receiver circuitry.
After that the selectivity of the receiver tuning circuitry is what enables a radio device to actually
seperate what it is meant to receive from the plethora of rado frequency garbage that is bouncing around the world 24/7.
Resolution of the GPS track is to do with how often a sample is taken in writing the track.

[GPSGuided]

...Conceivably, there would be differences if one was carried on the right leg pocket and the other on the left...

Absolutely! All comes down to how many satellites the GPSr can "see". Right side being a cliff face or dense canopy with an overhang and the body (70% water) blocking signal to the left can easily increase the error rate.

[GPSGuided]

Resolution of the GPS track is to do with how often a sample is taken in writing the track.

I guess I may be using the term loosely but GPS's resolution depends on the triangulation of more than 3 satellites along with the amount of available data received, as such the resolution/precision is affected. Further and as we know, US Defence Dept also controls and alters this parameter for their military circumstances. With the wide availability of GLONASS and satellite constellations from other countries, this may be less of an issue in time.

[frenchy_84]

Further and as we know, US Defence Dept also controls and alters this parameter for their military circumstances. With the wide availability of GLONASS and satellite constellations from other countries, this may be less of an issue in time.

Selective variability has been turned off for over a decade and will not be turned back on. GPS forms such a critical part of human life and economy these days that it do more harm than good turning selective availabilty back on.

[GPSGuided]

Selective variability has been turned off for over a decade and will not be turned back on. GPS forms such a critical part of human life and economy these days that it do more harm than good turning selective availabilty back on.

True. I was presenting it as a potential, one that the US military can still activate at their discretion if the circumstance calls.

[MrWalker]

I've carried two Garmin GPS units on walks down the Launceston Gorge several times and often have difference of up to a km between them in 3km. One is a newer model but there is no consistency in which has the higher reading. I assume it is due to one of them sometimes not finding the satellites for a while in the Gorge and taking a straight line to the last known point, as I don't get these differences in open country.

[michael_p]

We both carried them in a leg pocket.

I've had this happen. Only a few months back we did a walk where I had my GPSr in the side pocket of my pack. I would normally leave it in the top (flap) pocket but needed it at hand. It never had a good view of the sky and lost the satellites several times. End result was the track log was way off when overlaid on the topo map. With nearly 10 years experience using GPS you would think that I would know better.

Poor reception:

Cheers,
Michael.

[GPSGuided]

With nearly 10 years experience using GPS you would think that I would know better... Poor reception

I don't mind the poor tracking but the stationary wonder annoys me no end. When cleaned up, they typically represent considerable distance.

[Ent]

Hi

I notice the difference when using OSM tracks compared to actual claimed distance. It as mentioned is largely due to the lower the number of points the straighter the track, hence shorter. Also the small antenna and poor placement of the Fenix means it over calculates distance. I have either the Rino 65O or 62s clipped high on the shoulder straps and both will give different distanced covered but generally within 10% of each other.

We tend to think that GPSs are brilliantly accurate but tracking performance can vary considerably. Still light years ahead compared to a map and compass.

Cheers

[GPSGuided]

Narrow comparison. GPS puts a fix on a map (paper or electronic) and that's what counts most. Thought process of map/compass use is different to that of GPS use, as such, track distance is hardly critical.

[madmacca]

With an accuracy of 5-10 metres, raw GPS data can be pretty messy at walking speeds. The algorithms that "fill in the blanks" between breadcrumbs are probably the culprit here.

Tree canopy and the like can interfere with the GPS signal. As GPS satellites move fairly quickly, your mate following a few seconds behind you may get a different angle to the satellite, and picks up a signal where yours got blocked.

[GPSGuided]

Recently did a full track recording of my SYD-SFO-BOS return flights. Pretty clean. No random jumps.

[Myself]

I've been tesing the Oregon 650 last few weeks, just taking it with me when I'm out and about suburbia, and I can not see any noticable difference in accuracy between GPS+GLONASS and GPS ONLY. Of course, there're no significant overhead obstructions - which GLONASS was designed to improve accuracy with - but I've seen a few threads around different forums lately where posters have speculated on added 'general' accuracy of GLONASS, and have found that for the e tra battery consumption, GLONASS hasn't provided comparable improvements in "open air" accuracy.

I wish we did have decent tree-canopied areas around here, cos I'd happily go walk around for half an hour or so with the two different settings but the only trees around my neiborhood are the typical "road-liners' scattered out the front of houses.

The heaviest test of GLONASS+GPS I've tried so far was taking it into a shopping centre, where it lost reception almost the entire time, but connected the holes in the track afterwards.

I *do* however, get continuous sat reception walking around inside the house, be it at a reduced accuracy 8-10m generally inside.