Co-ordinate Issue.

[+The Duckers]

If I were to have a possible co-ordinate issue with a recently published cache, should I contact the cache owner or the friendly reviewer who published the aforementioned cache?

[Izzy and the Lizard King]

If I were to have a possible co-ordinate issue with a recently published cache, should I contact the cache owner or the friendly reviewer who published the aforementioned cache?

 

Most definitely the cache owner

[+MartyBartfast]

If I were to have a possible co-ordinate issue with a recently published cache, should I contact the cache owner or the friendly reviewer who published the aforementioned cache?

 

Most definitely the cache owner

 

And post a note on the cache page saying what you think the co-ords are, then anyone else looking for it will have an idea there might be a problem before going after it.

[+Bear and Ragged]

If I were to have a possible co-ordinate issue with a recently published cache, should I contact the cache owner or the friendly reviewer who published the aforementioned cache?

 

Most definitely the cache owner

 

And post a note on the cache page saying what you think the co-ords are, then anyone else looking for it will have an idea there might be a problem before going after it.

So long as it's not the final of a Multi, or a Puzzle cache!

[+Graculus]

We won't update cache coordinates based on information from another cacher who has found a problem. As said above you should post a log to the cache with the details or email the cache owner directly (click on their name). The cache owner is responsible for making sure their cache is properly maintained which includes updating coordinates when someone reports an error. Of course we will update coordinates if asked to by the owner if the update moves them further than is allowed when an owner tries to make such a change.

 

If it is a cache that has had a problem for some time then you can also post a 'Needs Maintenance' log on the cache with the details to alert the owner. If the problem persists and the owner doesn't seem to be paying attention to the logs you could post a 'Needs Archived' log which alerts us reviewers to the problem. We will then try and get the owner to sort the coordinates out. If they don't then we may archive the cache for lack of maintenance.

 

Chris

Graculus

Volunteer UK Reviewer for geocaching.com

UK Geocaching Information & Resources website www.follow-the-arrow.co.uk

Geocaching.com Knowledge Books

[Izzy and the Lizard King]

If it is a recently published cache which you have attempted but has yet been unfound, it might be an idea to post a DNF as well as emailing the CO thus alerting other cachers to a possible problem.

[+The Duckers]

Thanks all. I contacted the cache owner who has now modified the listing. It was a new cache and although I had not attempted it, I could see from Google Maps there there was an issue with one of the waypoint co-ords. All is well now.

[+JUSTHEJOB]

Thats a good point. The co-ords issue could be down to the newer GPS available which are alot better, also one must take into account the cache details, i have seen afew caches were the owner ofsets the co-ords as part of the quest and enter details in the cache rating

[+dfx]

i have seen afew caches were the owner ofsets the co-ords as part of the quest and enter details in the cache rating

 

that would not be a valid "traditional cache" though.

[+dave-harris]

There is a cache not to far from me where the owner has stated that the cache is within a 10 foot radius of the co-ords. He has reflected this in the ratings by upping it one star. When its like this it shouldn't be a problem. I found one cache today actually, that was over 40 foot from the published co-ords. When i found and logged it, my first words were "the co-ords for this cache are way out...". All future caches will now be aware that this is the case. I also asked the owner to re-visit this problem. Hopefully he will, and hopfully future cachers will chase him.

[+MBFace]

There is a cache not to far from me where the owner has stated that the cache is within a 10 foot radius of the co-ords. He has reflected this in the ratings by upping it one star.

Why - surely 10 feet is within the combined error margins on the setter's and finder's equipment and definitely not worth an extra star?

[+The Blorenges]

If the co-ords are within a 30ft radius of the cache location then we're happy.

 

More than 50ft off and we might note our own reading in our log for the cache owner's info.

 

More than 100ft off... Pray there's a good hint... Search likely spots... Mutter under breath... Give up... Log DNF.

 

MrsB

[+Amberel]

There is a cache not to far from me where the owner has stated that the cache is within a 10 foot radius of the co-ords.

How does he know that? Commercial GPSrs aren't good enough to justify even moderate confidence of 10ft accuracy in most cases, and they are certainly not accurate enough for you to KNOW. There are techniques that justify improved confidence, but very rarely to 100%.

 

He has reflected this in the ratings by upping it one star.

Very odd indeed. If he is correct and it really is accurate to 10ft it would be more appropriate to REDUCE it by one star.

 

When its like this it shouldn't be a problem. I found one cache today actually, that was over 40 foot from the published co-ords. When i found and logged it, my first words were "the co-ords for this cache are way out...".

How do you know it was 40ft out? On the basis of one reading each, all you know is that your two solutions showed a 40ft difference. It doesn't tell you which was the more accurate, or by how much.

 

Rgds, Andy

[+dave-harris]

There is a cache not to far from me where the owner has stated that the cache is within a 10 foot radius of the co-ords.

How does he know that? Commercial GPSrs aren't good enough to justify even moderate confidence of 10ft accuracy in most cases, and they are certainly not accurate enough for you to KNOW. There are techniques that justify improved confidence, but very rarely to 100%.

 

He has reflected this in the ratings by upping it one star.

Very odd indeed. If he is correct and it really is accurate to 10ft it would be more appropriate to REDUCE it by one star.

 

When its like this it shouldn't be a problem. I found one cache today actually, that was over 40 foot from the published co-ords. When i found and logged it, my first words were "the co-ords for this cache are way out...".

How do you know it was 40ft out? On the basis of one reading each, all you know is that your two solutions showed a 40ft difference. It doesn't tell you which was the more accurate, or by how much.

 

Rgds, Andy

 

First point. The co-ord take yo to a parking spot in a car park with no tree cover, and one can assume that to co-ords are pretty accurate.

 

third point. My gps tells me how accuarte my position is in feet. The co-ordinates for the cache take me to the middle of a shopping chains car park. After searching using the hint (behind a concrete post) of which there are dozens around the car park. I found the cache eventually just over 40 feet away on a footpath in woods behind the shop. The accuracy of my GPS under trees was 40ft. My gps told my i was 44 feet from the cache co-ords. Allowing for the + - margin of error that places the cache between 4 and 84 foot radius. Plus there are several DNF's published for this cache. I know one cacher that has visited the place 3 times, each time logging a DNF

[+Amberel]

First point. The co-ord take you to a parking spot in a car park with no tree cover, and one can assume that to co-ords are pretty accurate.

Not to 10ft, nor anything like it. If you were to average the readings over half an hour, spread over more than one visit, using a good quality GPSr, ensuring that the readings had a good HDoP, then in a completely open space with no tree cover you could justify an expectation that the co-ordinates are within 10 feet. A single reading without knowing the HDoP you should NOT have a reasonable expectancy that they are. They might be, but you have no way of knowing if they are or not.

 

third point. My gps tells me how accuarte my position is in feet.

No it doesn't. It can't, because it doesn't itself know how accurate it is in feet.

 

GPSr accuracy is an exceedingly complex subject. There are dozens of sources of error, some in the space segment, some in the receiver, some in between and some down to untrained users, and they affect the accuracy in different ways. Some GPSr manufacturers have tried to simplify the large number of complex metrics the system provides as an indication of relative accuracy, by condensing them into a single number and presenting it as an accuracy in feet (or metres, etc). This results in people who don't understand how the system works taking this number at face value. It cannot be emphasized enough that this figure does NOT guarantee that the solution the unit is presenting is accurate to the stated distance. It is merely a relative indication and it is grossly misleading for them to associate any units with it. All it should be used for is a guide - the lower the number, the more dependable is the solution.

 

Rgds, Andy

[+Graculus]

Understanding GPS accuracy requires fairly serious mathematical ability otherwise it is just a black art. This article will make everything clear....

 

Seriously though, all you need to understand (as Andy has said) is the figure shown as the 'accuracy' is not a true figure. It is approximate. If it says 15 feet then you can be reasonably sure your GPS has a good satellite signal and you could be within say 15 to 30 foot of where you think you are. 30 foot is perfectly normal for a GPS. With the WAAS/EGNOS option turned on that improves the accuracy and your GPS may show a figure of 10 feet or even less. It still isn't precise but you can be sure your GPS is slightly more accurate than without it turned on.

 

Here are two simple tests you can do. Switch on your GPS (or smart phone if you use that). Wait outdoors for at least 5 minutes until you've got a good signal and the 'accuracy' figure has settled down. Now take a walk down the road a few hundred yards to say a lampost and when you arrive, mark the coordinates. If your GPS shows an accuracy of say 15 feet then you'd think you are within 15 feet of the lampost. Now, over the next few days switch on your GPS and after waiting for it to settle down navigate to the coordinates for the lampost. When the GPS says you have arrived (0 feet) then you may find you are standing by the lampost but you may end up standing in the middle of the road or even on the other side of the road.

Second test is to make sure the track log is running and switch to the map screen. Put the GPS out in your garden for about 20 minutes. Then look at the track log (zoom in to the max on the map). You'd expect a single dot as the GPS hasn't moved but in fact you'll have a track showing as though a fly with ink on its feet had walked around the screen. This is caused by the variation in the satellite signal due to various things including the GPS slowly downloading the satellite data and 'improving' its accuracy as it 'locks' on to more satellites but it does illustrate how the GPS accuracy 'wanders'.

 

Chris

Graculus

Volunteer UK Reviewer for geocaching.com

UK Geocaching Information & Resources website www.follow-the-arrow.co.uk

Geocaching.com Knowledge Books

[+dave-harris]

I do agree with what both of you say, but my gps does show an accuracy, and should be used as a guide, and should be fairly accurate, depending on how many satellites you are homed in on. I can walk along any path, turn around and walk back, or re-trace the final part of a walk along a path and the track log will be slightly left or right of my original track. But only by feet. Leaving a gps standing still for a prolonged period of time could indicate you have walked around 1/4 or 1/3 of a mile without moving, but this depends on how long you leave it. It does not count for an error in logging co-ords by 40+ feet. When placing a cache myself i spend some time walking away and back to the hide, then take the most common / avergate position. That way i can be happy that my published co-ords are as close as they can be to any cache. On the account i previously mentioned the owner i fear, has not done this and only taken possibly one reading on possibly a smaller scale map, rather than zooming in on the gps to the largest scale which would give a better indication on positon. Even under tree cover, with a good gps you can be pretty sure you are not that far out.

 

I did not intend to rattle any cages with my first post. I was just making a point which has been taken the wrong way. My apologies if this has happened

[+Amberel]

I do agree with what both of you say, but my gps does show an accuracy, and should be used as a guide, and should be fairly accurate ...

Indeed, it can be used as a guide, but not as an indication that the solution is accurate to any particular number of feet. It should just be used as a relative indication - if your GPSr says 10 feet, it is a more reliable solution than if it was saying 15 feet. Note that I don't use the term accuracy here - it's quite possible for any single solution that has an "accuracy" figure of 15 feet to be, by chance, more accurate than one with a figure of 10 feet - but you don't know. The lower the figure, the more confidence you may reasonably assume.

 

I can walk along any path, turn around and walk back, or re-trace the final part of a walk along a path and the track log will be slightly left or right of my original track. But only by feet.

But in that situation many of the errors will be the same for both readings. You are likely to be using the same set of satellites, so any satellite ephemeris inaccuracy, or clock drift or other such errors will be the same for both readings, ionospheric distortion will be the same and any reflections are likely to be similar. You are confusing short term repeatability with accuracy, and they are not at all the same thing.

 

Leaving a gps standing still for a prolonged period of time could indicate you have walked around 1/4 or 1/3 of a mile without moving, but this depends on how long you leave it. It does not count for an error in logging co-ords by 40+ feet. When placing a cache myself i spend some time walking away and back to the hide, then take the most common / avergate position. That way i can be happy that my published co-ords are as close as they can be to any cache.

That's not "as close as they can be", though it should normally be entirely adequate for caching purposes.

 

On the account i previously mentioned the owner i fear, has not done this and only taken possibly one reading on possibly a smaller scale map, rather than zooming in on the gps to the largest scale which would give a better indication on positon.

Again, how do you know? You haven't said that you performed multiple readings over several days and at different times of the day when you found this cache, only when you hid your own, so it's easily possible your reading on this one was 40 ft out. Note that I'm not saying it WAS, I'm saying there's not enough information for us to make a judgement, so it's premature to assume that yours is "right" and his is "wrong".

 

I did not intend to rattle any cages with my first post. I was just making a point which has been taken the wrong way. My apologies if this has happened

I didn't take it as "rattling cages", I was primarily responding to two things:

 

1) with any difference between a single pair of solutions, most people seem to assume that theirs is right and the other is wrong. But it takes a great deal of work and knowledge of how the system works to make any reasonable judgement like that, and these days a lot of receivers don't even give you the low level information you need to assist that judgement.

 

2) the issue of the "accuracy" figure quoted by some GPSrs being taken literally, when it most certainly should not be. The GPSr chipset manufacturers do not offer any accuracy metric in the form of "x feet", this is added by the consumer receiver manufacturers. One can only assume they do this to make it easier for non-technical users to understand, but in my view this is quite wrong because it actually encourages non-technical users to believe the wrong thing, which is worse than not understanding it at all. Furthermore there is no standard algorithm for calculating this particular metric, so there is pressure on GPSr manufacturers to show optimistic figures because to the uninitiated it makes their units appear to be more accurate than their competitors.

 

Apologies if my tone sounds slightly acerbic, I spent 5 years as the UK tech support for a GPSr chipset manufacturer and most of that time was spent trying to correct misunderstandings of how reliable and accurate the system is. There are few subjects so prone to being misundersood - it's probably because the receivers are so easy to use that people thing the system itself is simple to understand and many will argue that black is white from a position of total ignorance. And receiver manufacturers quoting accuracy in feet doesn't help anyone's understanding of the subject.

 

Rgds, Andy

Edited January 3, 2011 by Amberel

[+dave-harris]

[+FantasyRaider]

 

Apologies if my tone sounds slightly acerbic, I spent 5 years as the UK tech support for a GPSr chipset manufacturer and most of that time was spent trying to correct misunderstandings of how reliable and accurate the system is. There are few subjects so prone to being misundersood - it's probably because the receivers are so easy to use that people thing the system itself is simple to understand and many will argue that black is white from a position of total ignorance. And receiver manufacturers quoting accuracy in feet doesn't help anyone's understanding of the subject.

 

Rgds, Andy

 

Rgds, Andy

 

I have read this thread with interest but ...... I'm still a bit confused.

 

Why is it that on a lot of occasions my GPS has gone down to 0 feet and 'BINGO' there is the cache.... right at my feet?

On some it goes down to 15 feet at the most and if I move in either direction it goes up .... so I go back to the lowest and bingo .... the cache.

Then On some of the logs (of that cache), people have noted it was about 15 feet out?

 

Also, when setting some of my own caches, people have logged that the co-ords were spot on (down to 5 feet or less)

Again, on some of my caches i couldn't get a good average due to tree cover (my Colorado says 20 feet at best) these are the caches that other people have noted are off by a few feet.

 

So why, if GPS's are not accurate do most finders notice the same difference in feet?

[+Amberel]

I have read this thread with interest but ...... I'm still a bit confused.

 

Why is it that on a lot of occasions my GPS has gone down to 0 feet and 'BINGO' there is the cache.... right at my feet?

On some it goes down to 15 feet at the most and if I move in either direction it goes up .... so I go back to the lowest and bingo .... the cache.

Then On some of the logs (of that cache), people have noted it was about 15 feet out?

 

Also, when setting some of my own caches, people have logged that the co-ords were spot on (down to 5 feet or less)

Again, on some of my caches i couldn't get a good average due to tree cover (my Colorado says 20 feet at best) these are the caches that other people have noted are off by a few feet.

 

So why, if GPS's are not accurate do most finders notice the same difference in feet?

I did say that the ease of reading co-ordinates off a GPSr display makes an extremely complex subject seem deceptively simple . Firstly, go back to what I said before. What you have just described is "repeatability" - a set of accurate readings should be repeatable, but a set of repeatable readings may or may not be accurate. They most often will be, but the the difficulty is in knowing when they are not. I'll try to give you a very simplified example to help make it clearer.

 

You are searching for a cache that is hidden at the base of a 6ft high wall to the south, with a house 20ft away to the north. The signals you get from satellites to the south may reach the receiver directly or may have bounced off the side of the house. You may fairly repeatably get a reading 40 feet north of where you really are, and just occasionally get a more accurate reading, depending on the satellite geometry at the time. I repeat, this is very simplified and is just to give you some sort of idea of the difference between repeatability and accuracy.

 

Secondly, I've never said that a GPSr is NEVER acccurate to 10 feet - it very often will be - what I've tried to get across is that you don't necessarily know when it is and when it isn't.

 

The accuracy of GPS solutions is a statistical measurement. For example, x% will be accurate to 10ft, y% will be accurate to 30ft and z% will be accurate to 100ft, where z > y > x. But, taken in isolation and without external reference, you don't know if any single solution is 1ft out or 1 mile out. Generally when we are caching, because we already know roughly where we are, we will know if a solution has a gross error and can disregard it. But small errors are much more frequent than large errors, and they are harder to detect.

 

Averaging is one technique for improving the confidence in a solution. Averaging over a short term (10 minutes) is far from ideal because many sources of error will remain fairly constant for this period. Taking an average of readings spread over several days, at different times of day, is a better technique.

 

Coming back to your point about often seeing a close correlation between your co-ordinates and the cache, of course that will often be so. But it is most certainly the case that the vast majority of people are unconciously selective about what they recall. They are more likely to remember co-ordinates that agree exactly or which are a long way out than they are to remember those which are 20ft out. To get a reliable indication of how well your co-ordinates agree with the cache page you would have to scrupulously record the reading on your GPSr at every cache you find, without being selective in any way, and then analyse the results after a period of time. But while this would show a more representative indication of the difference distribution, I can't see it would be of any use because it would not provide any prediction for specific future caches.

 

Going back further to the original issue of a difference between the co-ordinates on a cache page and the co-ordinates your GPSr indicates at the cache, and deciding which is likely to be the more accurate, the best way is for everyone who gets a significant difference to post their co-ordinates. It's not enough just to say "40ft out" - different finders might get a 40ft error in opposite directions! But even then you need to be very careful about how you use those posted co-ordinates - they are likely to show a bias in the opposite direction because solutions with a smaller difference are less likely to have been posted.

 

I'll finish up by saying that it would be easy to fill up a whole weeks training course on the subject of GPS errors, and that's assuming you already have a solid technical background. The above therefore necessarily contains simplifications and ommissions.

 

Rgds, Andy

Edited January 4, 2011 by Amberel