Marking Cache Coordinates

[+Indotguy]

I've seen quite a few new caches disabled due to incorrect coordinates. Having hidden almost 100 caches and had quite a few compliments on my coordinates I thought I would give a few tips for getting the best possible coordinates for your hides.

 

1.) First, study your GPSer's manual and go out and find some caches. Take some time to learn how your GPSer works before trying to hide any caches.

2.) Make sure your unit is on with sats. aquired well prior to marking waypoints.

3.) Most GPSers have a display screen indicating the approximate accuracy in feet. From my experience this number can vary depending on which direction you face. Watch the accuracy number and mark waypoints when it shows the lowest reading.

4.) I usually take at least 3 waypoint readings. If possible mark a waypoint on three sides of the cache and then average. In a deep woods situation where sat. signals are poor, it often helps to walk several feet away from the cache in different directions and then walk back to it before marking the waypoints. Either way, always give your unit time to settle before marking.

5.) Before or after you fill out your cache approval form, check your coordinates with an internet a map program like Topozone. Obviously the "X" should fall somewhere near where your cache is hidden.

6.) And lastly, check and double check your numbers as you entering them.

 

Hope these tips are helpful to some of you newer cachers.

Keep on Caching!!!

[+BlueDeuce]

Not bad advice,.

 

For some reason the first coords I take seem to be the most accurate. Maybe because it's after I've been standing at the cache site for a few minutes.

 

I too like to take a couple/three readings, walk away from the cache site (100 feet or so) and see how close I can get.

Edited July 3, 2005 by BlueDeuce

[+Thot]

The webpage for my coordinate averaging utility explains my method and it is much like yours.

 

Soon the folks that think taking more than one reading is a waste of time because there's error in every one you take will drop by. I think that's their reason, but I can't be sure because I don't understand it. You can judge for yourself here.

Edited July 4, 2005 by Thot

[+the hermit crabs]

It also helps to send a guinea pig in to try to find it, before you list it.

[+Indotguy]

Soon the folks that think taking more than one reading is a waste of time because there's error in every one you take will drop by.  I think that's their reason, but I can't be sure because I don't understand it.  You can judge for yourself here.

I'm sure your right and took it into consideration when I posted this. It is true there are many variables involved in getting consistent GPS readings, but common sense dictates that several readings, if properly done, are usually better than one. (I think I looked unsuccessfully for one of those "one reading" hides yesterday!)

[+Kit Fox]

I let my GPS (Meridian Gold) average for 3 to 8 minutes, on my cache hides. To date, with 45 hides, no one has complained about bad coords.

 

1. I don't hide caches next to tall buildings.

2. I don't hide my caches in steep walled canyons.

3. I hold my GPS vertically, when I average.

Edited July 4, 2005 by Kit Fox

[+robert]

I believe Kit Fox Meridian automatically averages when held still for a short amount of time. My 60CS has a button for averaging. When you push the mark button, one of the options is "average". It tells you the coords it has and then repeatedly averages them. So it will count up 1, 2, 3, etc readings it is taking. I usually let it get 50-100 readings before accepting the mark. I set it at the cache location while I get the cache ready to go and it will usually be done by then.

 

Even if you don't have an average feature, letting the GPS set for a little bit at the cache location should do you well. Just make sure your GPS Accuracy (also known as EPE on some units) is as low as possible for the conditions.

[+Hoppingcrow]

And don't forget the unfortunate fact that no matter HOW accurate your readings are, or how many hours you spent obtaining them, there's always going to be at least one person who hunts for the cache who claims your coordinates were "way off" or "out."

[+robert]

And don't forget the unfortunate fact that no matter HOW accurate your readings are, or how many hours you spent obtaining them, there's always going to be at least one person who hunts for the cache who claims your coordinates were "way off" or "out."

VERY TRUE!

 

[+SherwoodForest]

And don't forget the unfortunate fact that no matter HOW accurate your readings are, or how many hours you spent obtaining them, there's always going to be at least one person who hunts for the cache who claims your coordinates were "way off" or "out."

yeah, I always love getting the messages that someone's GPS had them a whole 15 feet away when they were at the cache.

 

But we must forgive the noob, sometimes they will learn the error of their ways (and of the concept of position error in general). The problem is the multihundred find cachers that suffer the same delusion... egads....

 

[+Thot]

And don't forget the unfortunate fact that no matter HOW accurate your readings are, or how many hours you spent obtaining them, there's always going to be at least one person who hunts for the cache who claims your coordinates were "way off" or "out."

I guess I've been lucky. I've never had anyone complain about my coordinates.

[+fizzymagic]

Soon the folks that think taking more than one reading is a waste of time because there's error in every one you take will drop by.  I think that's their reason, but I can't be sure because I don't understand it.  You can judge for yourself here.

OK, I am officially stunned speechless.

 

And to think that some of those people have been caching for more than a couple of years!

[ImpalaBob]

I'd say the best way to determine your coordinates is to go to the location several times at different times of the day and do some averaging. You should also try coming at it from different directions.

 

If you have asked someone to hike a considerable distance to find your cache, and the reason for your cache was to introduce the hike and view to the cachers, then a nice colored ribbon in a tree within 15 feet of the cache is a nice touch! Micros hidden in the middle of the woods .... yeah I like a visible marker on these so I at least know I am in the right area! Urban micros .... a close landmark is nice to know! If I know I am in the right general area for a micro, I will keep searching for quite some time before giving up.

 

Being FTF is like being a gunia pig ..... and I always expect to search a bit. It is a pleasant suprize when it is spot-on! I always give feedback on a personal E-mail about the accuracy of the coordinates especially if they are off by 50' or more. Any cache within 30' should be considered OK.

 

Listing your GPS accuracy at the time you took the reading is also helpful. You can edit it out as you get the coordinates zeroed in. ImpalaBob

[+DaveA]

Soon the folks that think taking more than one reading is a waste of time because there's error in every one you take will drop by.  I think that's their reason, but I can't be sure because I don't understand it.  You can judge for yourself here.

OK, I am officially stunned speechless.

 

And to think that some of those people have been caching for more than a couple of years!

OK, let's not get into a pointless debate, let's just summarize the facts in this controversy.

 

Why would one assume averaging coordinates would improve accuracy?

 

The idea behind this belief is that the errors in position are random. If it is true that the positional errors are random then averaging can improve accuracy. The idea is that coord 1 is dead on, coord 2 is off by 15 feet to the north, coord 3 is off to the south etc. Using enough random errors what happens is the errors cancel themselves out. In other words if the errors are random we can expect one coord to be off to the north and another off to the south. These two errors would bring the result more to the middle thus improving accuracy.

 

Why would one assume averaging coords would not improve accuracy?

 

The idea behind this belief is that the errors are not random. If they are not random then all coords are likely to be off in the same general direction. Averaging multiple sets of coords that are off in the same direction won't produce a better result and could produce a worse result (assuming the additional coords are further off than the first).

 

The question then is "Are the positional errors random?"

 

The answer is 'probably not'.

 

Back in the days of selective availability the US government intentionally sent coords that were innaccurate to reduce the precision of GPSrs. These errors were random. SA has been off for a long time now. Currently the errors we deal with are not random, but are caused by multipath (signal bouncing off objects) and satellite error among other potential factors. Because the errors are not random, it would seem averaging is at best, pointless.

 

However, that isn't the end of the story.

 

If we approach the cache site at different times and from different locations we might be able to improve accuracy via averaging. Why is this? Well satellite error ends up getting corrected. Multipathing may be reduced when not approaching thru the heavy tree cover or large rock wall.

 

The question you have to ask yourself is getting slightly better coords worth visiting the site multiple times at different times of day on different days and from multiple directions? Wouldn't it be easier to simply look at the GPS indicated EPE and do what you can while on site to get it as low as possible and then save the coords?

 

It has come to my understanding that Magellan units have some sort of goofy auto averaging that requires one to sit still for a few seconds before the coords 'catch up' with the actual location, but this should not be confused with averaging. GPS ''settling' is the term used to describe the need to sit still for a few seconds to get the best coords under such circumstances. I dont understand the algorithms the GPS manufactures are using so I can't explain why this is or isn't necessary for any given unit.

 

Anyway, that's all I have to say on this subject.

[+Tharagleb]

I guess a good way to kinda settle the average vs. no-averaging argument is to try it on a known coordinate. This would have to be a coordinate whose lat/long is know to high degree of accuracy (or several such waypoints) and then try using the average/no-average approach on it/them. I bet averaging wins out. I can't think of any good reason for averaging not to perform better.

 

Anybody out there own a Trimble (with sub-meter accuracy)? That would help settle this.

[+fizzymagic]

Wouldn't it be easier to simply look at the GPS indicated EPE and do what you can while on site to get it as low as possible and then save the coords?

Ok, I am even more stunned now.

 

It's even worse than I thought.

 

Wow.

[+briansnat]

Soon the folks that think taking more than one reading is a waste of time because there's error in every one you take will drop by. I think that's their reason, but I can't be sure because I don't understand it. You can judge for yourself here

 

Here I am. I make sure I have a good sat lock, hit mark and I'm done. I consistently get compliments about how good my coordinates are so to me averaging is a waste of time.

 

Even if averagingt does improve your coords (and averaging can make them worse in some instances) is 5-10 feet of additonal accuracy really worth all that effort? Nah.

[+DaveA]

Wouldn't it be easier to simply look at the GPS indicated EPE and do what you can while on site to get it as low as possible and then save the coords?

Ok, I am even more stunned now.

 

It's even worse than I thought.

 

Wow.

Perhaps you could articulate the reasons for your being stunned. Y'know, to further a discussion.

[+sbell111]

I use a variation of Brian's method.

 

When I get to ground zero, I set down my GPSr. While I'm placing the cache, its doing its thing. When I'm done, I pick it up and make sure it's got a good lock. If it does, I mark the waypoint and drive to Sonic where I get a chili dog and some tots. Usually, I'll get a cherry limeade, also.

Edited July 6, 2005 by sbell111

[+Tharagleb]

Wouldn't it be easier to simply look at the GPS indicated EPE and do what you can while on site to get it as low as possible and then save the coords?

Ok, I am even more stunned now.

 

It's even worse than I thought.

 

Wow.

Perhaps you could articulate the reasons for your being stunned. Y'know, to further a discussion.

Yeah, cause I think DaveA made a valid point, it shouldn't stun you (unless someone making a valid point stuns you).

 

I think waiting till you get a good constellation, and thus better accuracy, is one way to reduce the error. Don't have the patience for it myself but I tend to use the averaging method. If I know I am getting lousy accuracy then I give help with the cache location in the hints.

[+fizzymagic]

Perhaps you could articulate the reasons for your being stunned.  Y'know, to further a discussion.

No point. We've been around and around this topic too many times and I am sick of it. I am stunned because, despite consistently good evidence (and even gasp data) to the contrary, people keep bringing up the same flawed arguments (e.g. "averaging can make it worse," "there is no random error," "the searcher's GPS has a big error anyway, so it doesn't matter," etc.)

 

My conclusion? There are some people who, for whatever reason, don't consider it important to follow a pretty simple quality-control protocol to validate the measurements of the position of their caches. Nothing I say will ever change their minds. Unfortunately, newbies need to be constantly reminded that good coordinates do matter, and so we are blessed by people, like Thot, who help educate them.

Edited July 6, 2005 by fizzymagic

[+sbell111]

I'm not arguing with you because I'm sure you are correct. However, I think there is a very thin line separating the awesome coords you get with your method and the good coords that I get with my 'low epe and go get lunch' method. In my opinion, this argument is like those surrounding WAAS. Your coords are better, but people find my caches and I find theirs.

Edited July 6, 2005 by sbell111

[+fizzymagic]

However, I think there is a very thin line separating the awesome coords you get with your method and the good coords that I get with my 'low epe and go get lunch' method.  In my opinion, this argument is like those surrounding WAAS.  Your coords are better, but people find my caches and I find theirs.

I think that cache hiders generally have the obligation to those who seek their caches to get the best coordinates possible.

 

I have known some hiders who have considered making a cache more difficult to find by having poor coordinates "just part of the game." Those kind of hides make me grouchy.

 

If you're hiding a truly evil micro that is going to take people hours to find, even with perfect coords, then I think you should make the coords as good as possible. OTOH, if you are hiding an ammo can next to the only tree within 100 feet, precise coords are probably not necessary.

 

But don't tell me that taking a single measurement is going to give coords as good as can be obtained with multiple measurements, because it just isn't true. Coords taken just once will probably be "good enough," but is that really the standard we want to set for ourselves?

 

It might just be that, as a scientist, I am accustomed to verifying that my equipment is working as well as possible whenever I use it to take a measurement. I am used to collecting extra data just so that, in the unlikely event that something was not right, I can diagnose the error and correct the data. I've set up far too many experiments where you only get one shot at the data, and if anything goes wrong you are screwed.

 

Mental picture of a Deep Impact team member saying "Why do we need three course corrections? One should be enough!"

[+tozainamboku]

Fizzymagic is probably right that there is no point in trying to convince those who think that averaging won't give you a better chance of having accurate coordinates, but I'll still try. EPE (Estimated position error) is a value calculated by your GPSr that indicates that there is a 90% chance (manufacturers won't disclose this value but this seems to be the accepted value) that the location it reports is within the given distance of the actual location. So if you try to minimize the EPE it means there is a 90% chance you are at least that accurate. There is a 10% chance that you are less accurate. If you take one reading there is a 90% chance you will be within the stated EPE but there is a 10% chance you will be further away from the actual coordinates than the stated EPE. So one out of 10 times your coordinates won't be as good as the EPE. Now take 10 readings. Do like Thot says and take these readings over a period of time, on different days if possible. This will remove some of the non-random errors due to satellite geometry and atmosperic conditions. Average the 10 readings. If you have a similar small EPE for all your readings, the approximately 9 readings within the EPE with contribute much more to the average than the approximately 1 reading not within the EPE. So your confidence that you are with a given distance of the actual location will go up. With a big enough sample, it will approach 100%.

[+Night Stalker]

Ok, maybe I am a little slow. Some people are stating that averaging takes quite a bit of time. I have not found this to be true. I set my GPS down by where I am hiding the cache. I start it averaging and then proceed to hide the cache. By the time I have adequately hidden the cache and written whatever notes I need I pick up my GPS and tell it to save the coordinates. This seems to be long enough to get good coordinates. It takes maybe 5 minutes. Is this to long to wait for better coordinates? I don't think so.

 

[+DaveA]

Perhaps you could articulate the reasons for your being stunned.  Y'know, to further a discussion.

No point. We've been around and around this topic too many times and I am sick of it. I am stunned because, despite consistently good evidence (and even gasp data) to the contrary, people keep bringing up the same flawed arguments (e.g. "averaging can make it worse," "there is no random error," "the searcher's GPS has a big error anyway, so it doesn't matter," etc.)

So in other words you won't address the argument I made?

 

I don't believe I have ever weighed in on this discussion before so I am not one of those who discounts evidence. I am interested in your evidence.

 

Please present it. I presented what I believe to be true based upon the knowledge I presently have. If you have additional or contradictory knowledge I would appreciate learning of it.

 

Here is what I claimed based upon my present knowledge. I will make a claim, and then support it with my argument. I will label each part to make it easy for you to identify the part that is incorrect and explain why it is incorrect.

 

CLAIM: Averaging does not improve positional accuracy.

 

SUPPORT: Averaging can improve accuracy if the errors are random. SA produced random errors which averaging could minimize. SA is no longer present. Multipath and upper atmosphere conditions do not produce random errors therefore averaging cannot correct for them.

 

Please show where I am going wrong.

[+DaveA]

Ok, maybe I am a little slow. Some people are stating that averaging takes quite a bit of time. I have not found this to be true. I set my GPS down by where I am hiding the cache. I start it averaging and then proceed to hide the cache. By the time I have adequately hidden the cache and written whatever notes I need I pick up my GPS and tell it to save the coordinates. This seems to be long enough to get good coordinates. It takes maybe 5 minutes. Is this to long to wait for better coordinates? I don't think so.

 

This isn't the kind of averaging we are talking about when we say it takes too much time to be worthwhile for many. We are talking about returning to the same place on different days and different times of day from different directions and averaging those coords.

 

The type of averaging you are talking about uses proprietary algorithms that I know nothing about and probably also includes some limited multipath detection as well to toss out some coords and prefer others in order to display the best possible position fix.

 

The type of averaging you are talking about may very well produce a more accurate set of coordinates. To know whether it does or not would require understanding proprietary info about each manufacture's units as well as an ability to comprehend them. It seems Magellans do this by default when stationary and Garmins do it optionally if you tell them to. Don't know whether or not this is at all significant.

[+Night Stalker]

You may be right in your assumptions. I have never used the manual averaging method. I know other cachers who have, and their coordinates are really close. Whether they are any better than if they had not done their averaging I cannot say. I have used my Magellan's and my Garmin's averaging features and have found no significant difference between the result obtained by either one. I can say that any effort to improve coordinates by the hider (including double checking their typing) is really appreciated.

[+DaveA]

  I can say that any effort to improve coordinates by the hider (including double checking their typing) is really appreciated. 

well, consider this scenario before you go appreciating the efforts too much :

 

joe cacher goes and places his cache and his GPSr gives him a set of coords. His EPE (estimated position error) according to his GPSr is 10ft.

 

Joe then returns to the site the next day and takes coords again. The new coords are 50ft away from the first coords and his EPE is 75ft.

 

Which set of coords do you think would be the more accurate? If you answer the way I do that the coords with the 10ft EPE are statistically more likely to be the more accurate of the two coordinates then here is the million dollar question:

 

How does averaging these two coordinates improve the accuracy of the coords Joe Cacher lists on his cache page? Shouldn't he have just moved around a bit, faced different angles, tilted his GPSr and done whatever he could to get the lowest EPE possible and use those coords? Which approach would have yielded the more accurate coords?

[+fizzymagic]

CLAIM:  Averaging does not improve positional accuracy.

 

SUPPORT:  Averaging can improve accuracy if the errors are random.  SA produced random errors which averaging could minimize.  SA is no longer present.  Multipath and upper atmosphere conditions do not produce random errors therefore averaging cannot correct for them.

Sigh.

 

Multipath errors are geometry-dependent. Moving towards the waypoint from a different direction will give different readings for most GPS receivers because the satellites used will vary. Multiple readings made this way can help correct for multipath errors. As satellite geometry changes, multipath errors change; averaging multiple measurements taken over long time scales helps correct for it.

 

Ionospheric errors are effectively random, but major fluctuations occur over relatively long timescales of several minutes to an hour or so. Making multiple measurements of a waypoint over the appropriate timescale can help correct for these errors. Smaller fluctuations occur at short timescales, as well, of course. Averaging over short timescales can help reduce those errors. Making measurements at night can help minimize the errors associated with ionospheric models.

 

A single waypoint measurement contains no information about any errors. Making multiple measurements allows the user to detect if the GPS is giving inconsistent readings and may therefore be giving inaccurate coordinates. Since most of the sources of these errors do not show up in the EPE estimate of the GPS, simply waiting until the EPE looks good and taking a single measurement does not address them.

 

There. You got me to show you the problem with your claims. Are you happy now?

Edited July 6, 2005 by fizzymagic

[+Night Stalker]

If that person who got his second set of coordinates that were way off also reentered the previous coordinates and they took him right to the cache then of course not, but as we all know position seems to change depending on what time of day and all sorts of other issues. I have found that more errors occur because the person entering the cache information transposed his number then whether they averaged or not.

 

With that said we had a problem with a local cacher who's coordinates were consistently off. It turns out that this cacher was using a Magellan and was using the coordinates before the GPS settled down. This was causing 20 to 40' errors. After talking to this person about letting it settle for a few minutes these errors went away.

[CoyoteRed]

Not only that, but a real world experiment using a SporTrak Map and a Pro resulted in a +-0.001 readings. Put the unit in a convenient spot and let it sit for about 5 minutes. Break averaging and put it back. Wait 10 minutes, take a waypoint. Compare to the rest of the list taken over days and weeks. +-0.001 minutes.

 

Nope, averaging doesn't work.

 

I'd post my findings if I hadn't lost them in a crash. However, anyone is free to do the same experiment for themselves.

[+Quoddy]

A recent "test" I watched at Geo Jamboree 3 should prove useful to many cache setters. In order to get REALLY accurate readings the GPSr should remain at a position for a full 12 minutes. This time frames allows for all available birds to establish the correct coordinate readings. After viewing it, this seems to be far superior to the somewhat established paradigm of multiple marks.

[+DaveA]

 

Multipath errors are geometry-dependent.  Moving towards the waypoint from a different direction will give different readings for most GPS receivers because the satellites used will vary.

 

I am sorry, but this is incorrect. Yes, multipath errors are geometry dependent, but not because facing a different direction will result in different satellites being used (It generally won't, the birds are high enough that they don't care which way you are facing). Geometry dependent means that multipath signal is coming from a particular direction. Hence, the error will not be random, but consistent. Multiple readings cannot correct for a consistent, non random error.

 

for more information on multipathing please see this

 

Ionospheric errors are effectively random, but major fluctuations occur over relatively long timescales of several minutes to an hour or so.  Making multiple measurements of a waypoint over the appropriate timescale can help correct for these errors.  Smaller fluctuations occur at short timescales, as well, of course.  Averaging over short timescales can help reduce those errors.  Making measurements at night can help minimize the errors associated with ionospheric models.

 

Partially true. However ionospheric errors are corrected for via WAAS. If you are getting a WAAS lock you don't need to concern yourself with it. It is true (as I understand it) that this type of error is reduced at night.

 

There.  You got me to show you the problem with your claims.  Are you happy now?

 

Yes, I am happy that instead of giving dogmatic, non backed up derisive comments you are presenting information which can be discussed. Thank you for that.

[+fizzymagic]

I am sorry, but this is incorrect.  Yes, multipath errors are geometry dependent, but not because facing a different direction will result in different satellites being used (It generally won't, the birds are high enough that they don't care which way you are facing).

 

This is why I hate these kinds of conversations.

 

I didn't say "facing in another direction!" I said "approaching the site from a different direction." If you really want to have a discussion about this, it is key that you actually read what others write with the intent of understanding it.

 

In order to avoid derisive comments, I will stop at that. I will not be misquoted. This discussion is over.

Edited July 6, 2005 by fizzymagic

[+sbell111]

...But don't tell me that taking a single measurement is going to give coords as good as can be obtained with multiple measurements, because it just isn't true.  Coords taken just once will probably be "good enough," but is that really the standard we want to set for ourselves? ...

I clearly posted that I have no doubt that your way likely gives more accurate coords.

 

However, this is a game about finding a box of junk in the woods. If your coords are dead on and mine are 8 feet away, they are both good enough to play. Does anyone expect to go to ground zero, reach down, and pick up the cache every time? If so, that is the expectation that needs to be changed.

 

Therefore, to answer your question, 'Yes. Good enough is good enough.'

 

As the man said, 'This ain't rocket science.'

Edited July 7, 2005 by sbell111

[ImpalaBob]

I totally agree with FizzyMagic and he may have misunderstood my previous post. I think we all agree that with solar activity and tree cover we can get differtent degrees of accuracy at different times. I would go back to the spot several times and mark the waypoint noting my sat accuracy. I would then pick the waypoint with the least deviation (best accuracy). I would NOT average my readings!

 

I assisted in a geocaching class and marked a micro as I placed it in an open area. Class 1 found it at 2 feet, class 2 at 4 feet, class 3 at 3 feet. I'd say that one mark did it just fine! ImpalaBob

[+DaveA]

I am sorry, but this is incorrect.  Yes, multipath errors are geometry dependent, but not because facing a different direction will result in different satellites being used (It generally won't, the birds are high enough that they don't care which way you are facing).

 

This is why I hate these kinds of conversations.

 

I didn't say "facing in another direction!" I said "approaching the site from a different direction." If you really want to have a discussion about this, it is key that you actually read what others write with the intent of understanding it.

 

In order to avoid derisive comments, I will stop at that. I will not be misquoted. This discussion is over.

 

Your temper tantrums aren't going to make you more correct.

 

I addressed your points. If you are unable to deal with them then you are correct, the discussion, with you, is over. If you prefer change what I said from "facing direction" to "direction of approach" and it makes no difference. Net result is the same. Multipath and ionosphere errors are not random and therefore they cannot be corrected for with averaging. Averaging non random errors is pointless. Any mathematicians in the house?

[+DaveA]

I totally agree with FizzyMagic and he may have misunderstood my previous post. I think we all agree that with solar activity and tree cover we can get differtent degrees of accuracy at different times. I would go back to the spot several times and mark the waypoint noting my sat accuracy. I would then pick the waypoint with the least deviation (best accuracy). I would NOT average my readings!

I agree with you that we can get different accuracy at different times. I am unclear on what FizzyMagic said that you are agreeing with since you are rejecting the averaging of non random errors (which it appears he is arguing for) in favor of selecting the waypoint with the least deviation.

[+Thot]

. . . ionospheric errors are corrected for via WAAS.  If you are getting a WAAS lock you don't need to concern yourself with it. 

Given you believe that much in the value of WAAS, you need to weigh-in against the group that feels WAAS is of little to no value. Click here for an example.

 

Usually the people who feel taking multiple coordinates is of little or no value are the same people who think that about WAAS too. You seem to be an exception. Next time the frequent discussion of the merit of WAAS comes up you can join those of us who think it's helpful. That way you, I and fizzymagic can be on the same side.

[+DaveA]

. . . ionospheric errors are corrected for via WAAS.  If you are getting a WAAS lock you don't need to concern yourself with it. 

Given you believe that much in the value of WAAS, you need to weigh-in against the group that feels WAAS is of little to no value. Click here for an example.

 

Usually the people who feel taking multiple coordinates is of little or no value are the same people who think that about WAAS too. You seem to be an exception. Next time the frequent discussion of the merit of WAAS comes up you can join those of us who think it's helpful. That way you, I and fizzymagic can be on the same side.

ok

[ImpalaBob]

As a pilot who has flown IFR (Instrument Flight Rules) and done 3 missed approaches in dense fog looking for the end of a runway .... before GPS was in the panel, I can assure you that every pilot using GPS with WAAS is appreciative of ANY deviation correction no matter how slight it is. ImpalaBob

[+StarBrand]

I use averaging, approach from different directions and WAAS. Have compared all of that to taking one simple reading. It is better.

 

Actually, since WAAS uses an atmospheric "model" for corrections in most locations and the ionosphere is in a constant state of change - ionospheric errors are at least somewhat random. Also, multipath errors are somewhat random in that the angle to the sat makes a big difference in how much of a multipath error is present. Not really "random" but if I take readings at semi-random intervals over a period of time - there does exist at least a small amount of semi-randomess in the errors.

 

'nuff said.

[+Tharagleb]

I use averaging, approach from different directions and WAAS. Have compared all of that to taking one simple reading. It is better.

Okay, this is a little ambiguous, *which* is better?

Edited July 7, 2005 by Tharagleb

[+Tharagleb]

Thought experiment:

 

Joe claims that a single reading is better.

Jane claims that averaging is better.

 

On 10 successive days both Joe and Jane go to a known location and take a single reading.

 

For each of his 10 readings joe has an error, the 10 errors average say N feet.

 

Jane *averages* here readings gets a point P, which has an error of M feet.

 

I am not ever sure it is possible for M > N. (If Joe and Jane's readings are the same each day).

 

[Edit: Added last sentence]

Edited July 7, 2005 by Tharagleb

[+DaveA]

there does exist at least a small amount of semi-randomess in the errors.

 

'nuff said.

The errors are not random. True randomness and pseudo randomness are not the same.

 

Averaging only corrects for true randomness. The idea behind true randomness is that over a long enough period of time the errors will become fairly uniform. Imagine a dot with a circle around it. That circle represents a diameter of 60ft, 30 feet to every angle from the dot in the center. This is the area in which all of our collected coords will occur.

 

In a true random scenario like when SA was on by averaging over time, the coords, if plotted on our imaginary circle would be all over it in a reasonably uniform pattern. Averaging these coords would result in many of the errors canceling each other out thus improving accuracy.

 

In a pseudo random scenario like multipath the errors may, coincidentally, have the same effect as random errors where they will cancel each other out, but if this occurs it was by chance. We can't know if it occurs. Each reading may affect the coords in an eastern direction (potentially degrading accuracy), or perhaps one time in an eastern direction and the next in a northern direction, but never in our readings is there a counter error to the west and south.

 

So while averaging might improve accuracy in any particular case, it is by coincidence and it is just as likely to degrade the accuracy if the first coord was closer to the mark than the subsequent ones which get averaged in.

 

I can understand that it appears multipath and ionospheric errors are random, but they are not really random, they are pseudo random. This whole idea of averaging was born in the days of SA when the primary errors were genuinely random and has carried over to today (I think) because old ideas die hard and it is counter intuitive to believe 5 coords averaged out wouldn't be better than just one. While it may be counter intuitive, it is true.

 

To believe averaging is a good thing we pretty much have to believe that our first coord is the worst and subsequent coords will be better. I can't think of any reason to believe this would be true except by chance.

[+StarBrand]

To believe averaging is a good thing we pretty much have to believe that our first coord is the worst and subsequent coords will be better. I can't think of any reason to believe this would be true except by chance.

To believe that a single reading is a good thing we pretty much have to believe that our first coord is the best and most reliable reading and subsequent ones are worse. I can't think of reason to believe this would be true except by chance.

 

[+tozainamboku]

When I mark coordinates for a cache I've hidden, I'm not really concerned so much that I have a accurate abosulte location. Rather I want a repeatable reading, so that others that come looking for my cache will be able to find it close by. DaveA argues that most of the errors are not random and therefore averaging is not going to improve my reading. Good. Lets assume that is correct. Assume that both I and the finder have WAAS. This will correct for most of the atomospheric error. What remains can be considered random and averaging can help. Multi-path error will vary depending on the geometry of satellites at the time I take the reading. Since I don't know what position the satellites will be in when the finder is at the cache site the error (between me and the finder) is random. By averaging the location at different times of the day I can improve (on average) the reading for any finder. It still makes sense to take several readings (perferably over a period of several hours) and average them.

[+DaveA]

DaveA argues that  most of the errors are not random and therefore averaging is not going to improve my reading. Good. Lets assume that is correct.

 

OK, let's assume that.

 

Since I don't know what position the satellites will be in when the finder is at the cache site the error (between me and the finder) is random.

 

Hey, no fair! You said you were going to assume the errors were non random

 

The error from multipath is not random, it is pseudo random. You have to understand that mathetmatically speaking, random has a meaning. Only errors that are mathematically random can be reliably corrected for with averaging and even then it requires many datasets, not 2 or 3. Perhaps I should use the term mathematically random to hopefully reduce confusion on this subject.

 

Ionospheric and multipath errors are non mathematically random, SA errors are. We no longer deal with SA therefore we no longer deal with mathematically random errors therefore averaging is of no help except occasionally, by pure chance/luck.

[CoyoteRed]

To believe averaging is a good thing we pretty much have to believe that our first coord is the worst and subsequent coords will be better. I can't think of any reason to believe this would be true except by chance.

All of this postulating misses one point, you don't know if any one waypoint, taken by itself, is closer to the actual point on the surface of the planet than another.

 

Secondly, all of the readings a GPS will give you will be within a certain distance a certain amount of time, but you don't know which one is "more accurate" at any one time.

 

To help understand how GPS reading move around, a program like SA Watch is helpful. You can actually see each reading and watch how the readings drift around.

 

While I agree GPS signals aren't completely random, they are not organized to the point where averaging is worthless either. Yes, I agree that an averaged reading may be worse than a single reading, but just as likely it also may be much better. You wouldn't know unless you take other readings to compare it to.

 

Regardless, "accuracy" is not a real issue in geocaching. "Repeatability" is. There is a difference. You are wanting a reading than another geocacher can duplicate so they can find a box in the woods. Who cares if the coordinates given are the actual coordinates on the face of the Earth? As long as my reading and your reading matches or is pretty close, we're good.