Jump to content

Experiment On Accuracy


Recommended Posts

Guinea pigs wanted!!

 

I have just returned from a walk seeking possible cache locations.The most accurate my Legend got was 19ft. Its a damp and gloomy day does this affect it?I had a good number of sats up ,six.

Now here is the experiment,if you pop your gpsr outside now how accurate does it read?Just now mine was only 30ft! With 6 sats but none overhead.....any takers??

Link to comment

My Legend was giving me 25ft most of the day with not much view of the southern horizon. This went down to 16ft when view of southern was available, normally WAAS would kickin after 10mins a give 6ft but today I had a complete 30min session with a good view of sat 37 and the WAAS details didn't appear which it unusual

Link to comment

We have 2 Legends and a Geko.

The geko always seems to perform better than the two Legends.

WAAS works on the Geko, but does not work on either Legend.

The Legends are fairly new. The Legends have never used WAAS, If WAAS is switched on accuracy is worse.

The Geko no problem most days i've been out, a bit of a wait sometimes.

On a day like today with no tree cover i am lucky if my Legend shows less than sixteen feet.

Batteries last longer in the Legend than the Geko tho'.

The Legends accurate enough for me.

Edited by dpoet
Link to comment
my legend is swinging between 18 and 30 ft accuracy with 5 sats at the moment - with a clear night, and on top of the roof - WAAS disabled.

What are you doing on top of the roof?

erm, carrying out vital research i think!

besides, the police cant see me up here! :P

 

more info to add - no change with WAAS enabled!

Oh, I thought you was looking for UFOs. Must be just me that does that :unsure:

Link to comment
Now here is the experiment,if you pop your gpsr outside now how accurate does it read?Just now mine was only 30ft!

Your GPSr does not know how accurate it is. The "Accuracy" number displayed by Garmins is a misnomer.

 

In position-fixing the word 'accuracy' is defined as the degree of closeness to the true value. If your GPSr knows the true value, then surely it will display the true value, in which case the "accuracy" figure will always be zero.

 

To find the accuracy of your GPSr you need to place it on a known point and compare the indicated position with the known position. Look up the co-ords of a trig pillar and input them as a waypoint in your GPSr and then plonk it on the trig pillar or whatever. The indicated distance to go will be your measure of accuracy.

Link to comment
Experiment null and void then!!!!!!!!!!!!!! :laughing:  So why do garmin put it on the display?Surely it has some reference ?

Certainly. They could have put "This figure is a statistically-based estimate of the probability that the co-ordinates displayed will correspond to the true location within a stipulated percentage of given radius."

 

However, thats a bit long for a screen only 1.1 inches wide... <_<

 

In the circumstances, "Accuracy" is a reasonable word to use.

 

-Wlw.

Link to comment

If you wish to experiment then you would require a controlled sample, this would mean recording date time and position, and to be fair probably an open view to the heavens. Anyother method of doing this is meaning less really.

 

If you find a cache and your GPS says it is 16ft from where it is who is to say that your GPS was correct and the setters incorrect?

Link to comment

The Forester done a thread that a lot of people found informative, it was basically based on how accurate your GPS was.

If my memory serves me correctly he found that most GPS's gave an average accuracy of 3 - 5 metres. My GPS says accuracy 5 metres a hell of a lot, and I don't usually have problems with my set co-ords for caches, unless I have stated otherwise. I have found in the past that when my GPS says accuracy 10 metres or above, that any co-ords that I try to take are total pants.

 

So with this knowledge, I still firmly hold the belief that the accuracy displayed is a good guidance to go with. It must have some relevance, if you get a good fix when the numbers are low and a bad fix when they are high?

Link to comment
If you find a cache and your GPS says it is 16ft from where it is who is to say that your GPS was correct and the setters incorrect?

That depends on who the setter is, and how much work and time they put into establishing the correct co-ordinates.

 

If you find one of my caches, the true discrepancy (if there is one) will be a lot less than sixteen feet.

 

-Wlw.

Link to comment
I have just returned from a walk seeking possible cache locations.The most accurate my Legend got was 19ft. Its a damp and gloomy day does this affect it?I had a good number of sats up ,six.

It has been found in the past, that if you allow your track log to build up, once it gets full it can have an effect on your accuracy. Since finding this out, I clear mine after every trip.

 

When I found this out, the person with the problem couldn't get a good fix at all. She was told to check her track log, and on doing so found it was full. She deleted all the logs, and subsequently started getting good readings again.

Link to comment
That depends on who the setter is, and how much work and time they put into establishing the correct co-ordinates.

 

If you find one of my caches, the true discrepancy (if there is one) will be a lot less than sixteen feet.

 

-Wlw.

I would agree to this to some extent, but even experience can't allow for on the day issues such as lack of satellites, seasonal changes, user error due to time or frame of mind, GPS make and / or model etc

 

Unless you have some expensive DGPS then I would be careful with your claim of 16ft someone might try and prove this wrong, and who is to say that 2 GPSr's stood at the same location, "Yours is incorrect"

Link to comment
I would agree to this to some extent, but even experience can't allow for on the day issues such as lack of satellites, seasonal changes, user error due to time or frame of mind, GPS make and / or model etc

Absolute nonsense...

 

1) "Time of day" - I survey my caches at least three times on different days, before submitting them. Don't you?

 

2) "Seasonal changes" - what seasonal changes? Have you even the remotest idea what you're talking about?

 

3) "Errors due to time or frame of mind." - I don't make errors due to either of these causes.

 

The claim stands - and you or anybody else is welcome to try and disprove it.

 

Bloody cheek... :laughing:

 

-Wlw

Link to comment
I would agree to this to some extent, but even experience can't allow for on the day issues such as lack of satellites, seasonal changes, user error due to time or frame of mind, GPS make and / or model etc

Absolute nonsense...

 

1) "Time of day" - I survey my caches at least three times on different days, before submitting them. Don't you?

 

2) "Seasonal changes" - what seasonal changes? Have you even the remotest idea what you're talking about?

 

3) "Errors due to time or frame of mind." - I don't make errors due to either of these causes.

 

The claim stands - and you or anybody else is welcome to try and disprove it.

 

Bloody cheek... :tired:

 

-Wlw

So there I was the other day caching with Moote, Pengy & Tigger and Mattwaggie and all our GPSr's gave different differances. All good enough to find the cache.

 

As to seasonal changes, he may have meant the weather, rather picky.

 

As for cheek, thought he was just giving an opinion, or is it that only one persons opinions count.

Link to comment
I would agree to this to some extent, but even experience can't allow for on the day issues such as lack of satellites, seasonal changes, user error due to time or frame of mind, GPS make and / or model etc

Absolute nonsense...

 

1) "Time of day" - I survey my caches at least three times on different days, before submitting them. Don't you?

 

2) "Seasonal changes" - what seasonal changes? Have you even the remotest idea what you're talking about?

 

3) "Errors due to time or frame of mind." - I don't make errors due to either of these causes.

 

The claim stands - and you or anybody else is welcome to try and disprove it.

 

Bloody cheek... :tired:

 

-Wlw

Time of day = There are times when the satellite coverage is not a good as others.

 

Seasonal differences = Tree cover, increased cloud, rain or moisture (Yes the signal is absorbed by water)

 

Frame of Mind = No one is perfect; or am I incorrect in this assumption?

 

I'm sure you do check your caches many times; I also do, but I admit mistakes/problems. One cache which I own has particular bad coordinates, due to several reasons:

 

Moisture

Tree Cover

and another factor as yet unmentioned electrical interference

 

Now I'm not shouting out that my coordinates are brilliant I never would as it will come back to haunt

Link to comment
Your GPSr does not know how accurate it is. The "Accuracy" number displayed by Garmins is a misnomer.

Ludicrous, that say like saying that when you weigh a kilo of sugar on some kitchen scales, you have measured exactly 1 kilo. Measurement is only as accurate as the device / situation allows.

 

All measuring devices, including GPSr's have a degree of error, this can be calculated, and the way the GPSr does it is by monitoring time signals from several satellites and averaging them to provide an indication of position. There for it can provide a measurement of accuracy and in the GPSr's case this means ft or meters from the present point.

Link to comment
I'm sure you do check your caches many times; I also do, but I admit mistakes/problems.  One cache which I own has particular bad coordinates, due to several reasons: (etc)

 

I understand now.

 

You have an attitude to the presentation of your caches, which you assume everyone else adopts, also.

 

This is not always the case.

 

If there is a factor which adversely affects the co-ordinates, I either eliminate it or calculate a correction for it.

 

In the unlikely event of a mistake, I don't have to admit it - I just fix it. Once all this is done and checked, and checked again, the cache is ready to be listed. :tired:

 

Because of this attention to detail, I can make the claim that you originally objected to AND stand over it.

 

-Wlw

Link to comment
I'm sure you do check your caches many times; I also do, but I admit mistakes/problems.  One cache which I own has particular bad coordinates, due to several reasons: (etc)

 

I understand now.

 

You have an attitude to the presentation of your caches, which you assume everyone else adopts, also.

 

This is not always the case.

 

If there is a factor which adversely affects the co-ordinates, I either eliminate it or calculate a correction for it.

 

In the unlikely event of a mistake, I don't have to admit it - I just fix it. Once all this is done and checked, and checked again, the cache is ready to be listed. :tired:

 

Because of this attention to detail, I can make the claim that you originally objected to AND stand over it.

 

-Wlw

I don't have an attitude the presentation of my caches, and it is rude and untoward to say that.

 

You have never done a single cache of mine and to pass judgment without doing one is not on.

 

I'm not perfect that is what I'm saying, and the fact that I admit that show a lack of having such an attitude, it just shows that I accept my weaknesses and live with them.

 

So tell me how do you set an accurate position without spending £10,000+ on an precision DGPS system? I would love to know.

Link to comment
Your GPSr does not know how accurate it is.  The "Accuracy" number displayed by Garmins is a misnomer.

Ludicrous, that say like saying that when you weigh a kilo of sugar on some kitchen scales, you have measured exactly 1 kilo. Measurement is only as accurate as the device / situation allows.

 

All measuring devices, including GPSr's have a degree of error, this can be calculated, and the way the GPSr does it is by monitoring time signals from several satellites and averaging them to provide an indication of position. There for it can provide a measurement of accuracy and in the GPSr's case this means ft or meters from the present point.

That's interesting to know, I have always stood by the fact that the GPS accuracy was there not just for show. I have never been happy with the thought that Garmin and Magellen would spend so much money putting in a programme onto every handheld GPS that they make, just for it to look good.

 

Now I don't know if what you are saying is fact, but in my mind it certainly holds more weight with me than it's a misnomer, as I have stated what large company in their right mind is going to spend money on something that they know is useless.

Link to comment
Your GPSr does not know how accurate it is.  The "Accuracy" number displayed by Garmins is a misnomer.

Ludicrous, that say like saying that when you weigh a kilo of sugar on some kitchen scales, you have measured exactly 1 kilo. Measurement is only as accurate as the device / situation allows.

 

All measuring devices, including GPSr's have a degree of error, this can be calculated, and the way the GPSr does it is by monitoring time signals from several satellites and averaging them to provide an indication of position. There for it can provide a measurement of accuracy and in the GPSr's case this means ft or meters from the present point.

That's interesting to know, I have always stood by the fact that the GPS accuracy was there not just for show. I have never been happy with the thought that Garmin and Magellen would spend so much money putting in a programme onto every handheld GPS that they make, just for it to look good.

 

Now I don't know if what you are saying is fact, but in my mind it certainly holds more weight with me than it's a misnomer, as I have stated what large company in their right mind is going to spend money on something that they know is useless.

HH

 

It is just a sample and averaging calculation which works out approximately where you are and gives average each side of the mean for accuracy.

 

If these devices were so accurate, then the OS and the like would spend £300 or less rather than their usual £10,000+ on GPSr equipment.

 

As they say it ain't rocket science, but what is being used above is :P

 

Moote

Link to comment
The SA Watch screen image below shows data collected at the NGS benchmark, S159, located near Slaughterhouse bridge, Aspen, CO. Note that the mean of all samples collected was just 2.2 meters from the USGS surveyed location of the benchmark and 99% of all samples were within 5.6 meters of the benchmark!

 

Funnily enough, as I stated above, that's what my accuracy reads most of the time, so I guess it must be accurate, and the misnomer statement is more Urban Myth than fact.

Link to comment
In position-fixing the word 'accuracy' is defined as the degree of closeness to the true value. If your GPSr knows the true value, then surely it will display the true value, in which case the "accuracy" figure will always be zero.

 

To find the accuracy of your GPSr you need to place it on a known point and compare the indicated position with the known position. Look up the co-ords of a trig pillar and input them as a waypoint in your GPSr and then plonk it on the trig pillar or whatever. The indicated distance to go will be your measure of accuracy.

So, if I understand this correctly in the first paragraph accuracy is defined as how far the GPSr thinks it is from the true position, and the second paragraph is how far you measure the GPSr as being from the true (known) position?

 

I get your point, even if I find it a tad pedantic. All we need to know as geocachers is that if it's reading accuracy 12 feet we don't have to search such a large area as when it's reading accuracy 100 feet. As the caches have been set by other people using GPSrs then we are all working to the same sort of boundaries anyhow.

 

I appreciate that when my GPSr is reading 12 feet, this does not mean that I should get too excited about finding the cache within 12 feet as it depends upon the conditions and also cross checking of the posted co-ords by the setter.

 

Ultimately, if my GPS took me to within 1 foot of the cache every time, I think the game would be too easy. All to often the trampled trail points to the piles of sticks in the obvious rotten stump, so the GPSr co-ords are just and means to an end anyhow.

Link to comment
All we need to know as geocachers is that if it's reading accuracy 12 feet we don't have to search such a large area as when it's reading accuracy 100 feet. As the caches have been set by other people using GPSrs then we are all working to the same sort of boundaries anyhow.

 

But maybe the GPSr of the cache owner was only "accurate" to 50ft and you still have to look 62ft from your position.

Still I agree that it might be a boring if GPSrs were 100% accurate all the time! :)

Link to comment
Guest
This topic is now closed to further replies.
×
×
  • Create New...