Forum: GPS Tech Center
Topic: Determining Accuracy
started by: OG (Old Guy)
Posted by OG (Old Guy) on Jan. 10 2003,6:37 pmThrough an interesting turn of events at Christmas I have two new SporTrak Pros.
Any time I compare the coordinate readouts between the two, I get different readings ( sometime quiet different).
My question is how does one determine the accuracy of these or any GPSr?
I am going to return on of the units and I would like to keep the best one.
Posted by Scout on Jan. 10 2003,8:03 pm
Good question. You could put your faith in the surveying done for some local benchmarks. Hunt a dozen or so and pick the GPS receiver that has the least deviation from the surveyed locations.
If you don't have faith in local benchmarks, you could do the same thing with local caches, relying on the accuracy of the hiders' GPS receivers and the people using them.
Or you could go to a caching picnic and get a dozen of your friends together in as tight a circle as you can manage without getting slapped and compare your GPS readings with the others.
Posted by Morseman on Jan. 11 2003,11:11 am
As Scout suggests, you need to compaire them to a know and accurate standard.
However, it may not be as simple as that! One may be closer than the other at one time and then vice versa another time!
The proble is that it's not the internal workings of the GPS receiver that is causing you a problem, it's the geometry of the satellites around you.
In theory they should both show the same possition when side by side, but we are talking about Microwave Radio signals, and they can suffer from fading, reflection and other phenomenon like that. So, you can't really say one is going to be better than the other for a different day, time and place.
Just sell one and keep the other.
If your unit shows the DOP and EPE values, you will get an idea of how far away you might be from a place that your GPS is displaying. As you hunt for Caches, you will learn that, when you get to withing about 10 metres or so (30 feet if you prefer), you will probably get a 'jitter' as the signals from the sats change. It will be worse in tree cover! This means that you will be confidently watching the distance to the cache decreasing and suddenly you are further away than you were a second ago! Probably approaching from a totally different direction as well!
You then need to use your detective skills (or decode the clues) to find the cache.
Posted by Morseman on Jan. 11 2003,11:19 am
The problem there, Scout, is that the person hiding the cache was probably showing some innaccuracy as well!
So, if the person hiding was actually 30 feet away from the "real" location (here we could start a thread about precision and accuracy - ) and your GPS is 30 feet out as well, then the combination is a huge area of uncertanty.
Also, even if the person who placed the cache was, by some piece of luck, very close to the real location, when you go there the geometry of the satellites will be very different! They will be in different possitions and you may be blocking signals with your body if you approach from a different angle, or hold the GPSr in a different way...
I don't think people realise that the sats are actually orbiting and moving around up there. In fact, if you watch the number of the sats you are receiving, you will see that they change all the time as some go over the horizon, to be replaced by others.
Posted by Scout on Jan. 11 2003,1:10 pm
Good point. I suggested using a dozen caches to perform this test. If one of your GPS receivers is consistently "closer" to the zero point than the other, keep it and return the other. If, as I suspect, one GPS receiver will perform better on some of the caches and the other receiver will perform better on the others, then flip a coin to decide which to keep. Both receivers are probably operating within spec and neither one is more accurate than the other, even though they don't always agree.
Posted by Morseman on Jan. 11 2003,3:27 pm
I'd go with the 'flip a coin' idea. You have a 50/50 chance of getting the "right" one, and either could be as good, or bad, as the other.
The problem is one of chance.
The fact is that, every time you go out there's a 50/50 chance that one or the other will seem more accurate. Because of this, it may appear that one is more accurate than the other every time you go out. Neither of the units has a memory, the starting point is the same for each every time you go.
In fact, neither is better, nor worse, than the other, but one appears to be better because you think it is.
Consider a coin, every time you flip it there's a 50/50 chance it will land head side up. If you flip it 3 times and it comes up heads, would you put your money on tails a fourth time? If you do, you still have a 50/50 chance of winning, and it could still come up heads. In fact, it could go on being heads for another 100 times, because each time you throw it in the air, there's a 50/50 chance it will be heads.
Posted by headmj on Jan. 12 2003,2:28 pmMorseman if you really thinkthee is a significant difference then I would try at least 3 or 4 trig points (your in the UK right?). If there was no statistically valid difference sfter that I would flip a coin or go with your gut feel or returm the one that was the most expensive!
Posted by Morseman on Jan. 13 2003,8:33 am
Erm, I think you are confusing me with the person who asked the question.
However, you are supporting my point of view.
Posted by Morseman on Jan. 14 2003,6:37 am
If you are ever in the UK try looking for one of the Passive GPS markers that are listed on < www.gps.gov.uk >
These are maintained by the Ordnance Survey, and are maintained to ensure accuracy.
There are still over 6000 Triangulation Pillars (Trig Points) dotted around the UK, but the majority are no longer maintained. Any of the old pillars which form part of the maintained system are marked with a metal plate saying "This monument forms part of the Ordnance Survey National GPS Network". Any without this mark are not part of the system, but are useful for walkers as many of them are near footpaths, and are marked on many OS maps that walkers tend to use.
For walkers, the fact that the pillar may have moved a few millimetres or so is not important.