Jump to content

Finding caches hard: autoaverage problem?


geomaineiacs

Recommended Posts

In reading some of the FAQs, there was a reference to autoaveraging GPSs having difficultly in finding caches. I have only found 4 caches so far, but I notice that my GPS (Garmin Xtrex Legend) gets me close, but the cache is about 20 to 50 ft off when I do find it. I don't have my GPS in front of me, so I'm not sure if it autoaverages, but is my experience typical, or is it a function of my GPS?

Link to comment

Keep in mind that your GPS is getting a fix +/- to the coords. given to you by another geocacher, whose GPS may not had a good reading, or they did not re-check their coords. after they placed the cache to make sure it was as close as possible before posting them on the site.

Making several aproaches to the cache before logging the coords. will help when cachers place the cache on the site.

Also checking the number of sat. signals and the strength of the signals will help accuracy for hiders and seekers alike.

Link to comment

quote:
Originally posted by Renegade Knight:

You are normal.

 

The autoaverage thing is a Magellan thing. Magellan users tend to overshoot the cache and then end up coming back as they zero out.

 

Garmins tend not to have you do that.

 

=====================

Wherever you go there you are.


 

I learned a long time back to rtfm and the manual does indicate the accuracy of the Magellan does get better when time is allowed for the autoaveraging to work. When I approach the site with my MeriPlat, I tend to sit still for a few minutes and allow the GPSr do its thing.

 

Cheers!

TL

Link to comment

The Legend doesn't auto average. Being 20-50 ft off is quite normal for any unit. You have to realize that your unit may have a 15-30 ft margin of error, and so did the cache placer's unit.So the cache could be as much as 60 feet from what your Legend is telling you is ground zero.

 

Most people put the GPS away when they get within 30-40 ft and just start looking.

 

"It has been my experience that folks who have no vices have very few virtues" -Abraham Lincoln

Link to comment

quote:
Originally posted by BrianSnat:

The Legend doesn't auto average. Being 20-50 ft off is quite normal for any unit. You have to realize that your unit may have a 15-30 ft margin of error, and so did the cache placer's unit.So the cache could be as much as 60 feet from what your Legend is telling you is ground zero.

 

Most people put the GPS away when they get within 30-40 ft and just start looking.

 

_"It has been my experience that folks who have no vices have very few virtues"_ -Abraham Lincoln


 

icon_biggrin.gif Well that's the biggest reason I don't put away the MeriPlat. The autoaveraging (AA) has put me on top of the cache more times than not.

 

My interest is in more than just the search, but the overall accuracy of the GPS with the AA. I want to see if the AA technology is useful or not and surmise whether putting the unit away would speed up or hinder the search.

 

Granted, the number of my finds are very small right now, but the numbers are starting to tell for themselves.

 

In each of the logs I make, you will find I'm putting down the EPE and indicating whether I was on top of the cache or how far off I would have been physically had I stood on top of the coordinates as shown on the GPS.

 

Cheers!

TL

Link to comment

quote:
Originally posted by nincehelser:

But how do you know how helpful it is without knowing the accuracy of the original coordinates?

 

The only way I can see to really test how well auto-averaging works, is to take it to locations where the coordinates are precisely known.

 

George


 

Lookup nearby benchmarks that have the coordinates set by the 'professional' grade GPS that surveyors use. Their accuracy is several orders of magnitude more accurate than the commercial grade units.

You can access the detailed list here

 

Enter your coordinates and radius in miles. Search for GPS Sites and go to one of them and check your GPSr accuracy.

 

These are the coordinates listed for one nearby:

NAD 83(1991)- 42 12 32.56785(N) 121 44 50.17053(W)

NY0977 ELLIP HT 1227.63(m)

 

As you can see, the Seconds decimal point goes way out there.

Link to comment

quote:
Thats about normal. Somedays and some locations it will be even worse. Thats what makes it fun!

 

I agree, I get the best enjoyment out of going as close as I can to the coords and then stand there and think "Now, where would I have hidden a cache?" and look around. I'm always happiest when I actually was right! icon_smile.gif

 

But then, when I'm fooled by a really good cache hider, it really makes my day when I finally track it down! icon_biggrin.gif

 

 

"The hardest thing to find is something that's not there!"

Link to comment

quote:
Originally posted by nincehelser:

But how do you know how helpful it is without knowing the accuracy of the original coordinates?

 

The only way I can see to really test how well auto-averaging works, is to take it to locations where the coordinates are precisely known.

 

George


 

I have gone to benchmarks as well. It is interesting to see that the caches that are hidden seem to be hidden with an average of an EPE of 30 feet while I'm standing on top of them according to my GPS whereas the benchmarks have an EPE of 7-10 feet while I'm standing on top of them. So you see, I am getting a pretty good feel for this.

 

The benchmarks were for a cache hidden under the name of Map Datum here in Puget Sound region, but interestingly enough, the cache hider chose to use two benchmarks to teach about the different map datums. When I located the cache, I had to use the 2nd benchmark to see just how far off it was on the day I went looking for the cache; as the day was giving me best EPE of 30 feet. Using the error I noted when standing on the benchmark, I literally walked straight to the cache and 30 seconds later had the log book out.

 

It doesn't take much to figure it out. As pointed out, you have to work within the confines of the unit you have. But how do you know what those confines really are unless you are willing to spend the extra time to experiment and take notes that mean something?

 

I went to a waypoint yesterday and the area was trampled for an easy 100 foot diameter. I'm sure if the user knew what their confines were, they wouldn't have had to trample so much. I was easily looking in two much smaller spots within that same location and caused less damage to the ecosystem as well as did not contribute to providing any unintended clues.

 

It turned out the waypoint was actually a jump-off to a multi-cache mathmatecal puzzle, but I didn't have my tip sheet with me yesterday and after 10 minutes search, realized I was going to need it to figure out what I was supposed to do. I just happened to drive by and noted on my GPS it was there to discover. icon_rolleyes.gif

 

I also agree with MuzzleBlast!'s assesment of enjoyment at being able to puzzle out where that cache would be hidden once I allow my GPS to get me as close as possible. For some folks the numbers game is all about how many can I find. For me, the numbers game is how close can I get before I have to put away the GPS. With the MeriPlat's AA, it's like playing follow the leader with a global wobble. icon_eek.gificon_biggrin.gif But the reality is, with a little more time spent allowing the averaging to occur, I'm more likely to look down at my feet and say... there it is! Besides, it gives me a chance to look around and enjoy the sights and sounds instead of racing off to the next cache.

 

Cheers!

TL

 

[This message was edited by TotemLake on May 06, 2003 at 08:39 AM.]

Link to comment

quote:
I'm sure if the user knew what their confines were, they wouldn't have had to trample so much.

 

The problem is that things change day to day and often hour to hour. Things like sattelite alignment, cloud cover and even where you are holding the receiver in relation to your body can effect the accuracy.

 

"It has been my experience that folks who have no vices have very few virtues" -Abraham Lincoln

Link to comment

quote:
Originally posted by BrianSnat:

quote:
I'm sure if the user knew what their confines were, they wouldn't have had to trample so much.

 

The problem is that things change day to day and often hour to hour. Things like sattelite alignment, cloud cover and even where you are holding the receiver in relation to your body can effect the accuracy.

 

_"It has been my experience that folks who have no vices have very few virtues"_ -Abraham Lincoln


 

I am in general agreement with you.

 

Mine was a response to the general put away the GPS when within 30-40 feet of the coordinates. That makes for a very large search radius given the error rate on the highside of what I read to be 70% of the time you are that close with a Garmin and from what I have read 50% of the time with the Magellan (not that this is my experience with my MeriPlat).

 

With a 30' EPE, you already have a 60 foot diameter to work with when you are on top of the indicated coordinates. Putting the GPS away when you are 30 feet away from the indicated coordinates dramatically increases your search area. ...at least that's the way I see it based on what I have read on the accuracy.

 

I'm curious, and I direct this to the folks that do put away their GPS 30-40 feet from the coordinates: Do you find you are closer than said distance to the cache most of the time, or further away than the said distance from the cache most of the time?

 

I submit Garmin users put away their GPS because they find they are generally closer, where the Magellan users will keep their GPS out most of the time because it will direct them closer with the autoaveraging.

 

It would be interesting to note this by way of three separate Polls, one for the Garmin users, one for the Magellan users as these two are the major brands out there, with the third Poll to cover all of the others. Would others find this kind of poll useful?

 

Cheers!

TL

Link to comment

You know, I've found a few caches since I've started and I've never thought about the technical side of it. I just start searching at 'ground zero' and work my way out from there. I realize that I may be pretty far away from where my GPS tells me I should be. That's OK. After all, I have know idea of the accuracy of the hider.

 

Its possible that you are making this way too difficult by overanalyzing the technical component.

Link to comment

It's not very often that I start at "ground zero" with my search. Often as I get within 50 or so feet of the coordinates, I start looking for obvious hiding spots ahead of me, and find it quickly.

 

It's only when it's hidden really well do I have to resort to spiraling out from the zero point. That's not too often.

 

George

Link to comment

quote:
Originally posted by TotemLake:

.... where the Magellan users will keep their GPS out most of the time because it will direct them closer with the autoaveraging.


 

No there's no real truth in that statement these days due to many reasons and many different variables at different times.

 

In affect (despite what the manual might says) autoaveraging can actually make the position worse.

 

Autoaveraging is one of those things that if it feels better using it (but then one really doesn't have a choice with some units) then do it but it's really only a feel good thing as there's nothing to indicate at the time if the results are better or worse.

 

Cheers, Kerry.

 

I never get lost icon_smile.gif everybody keeps telling me where to go icon_wink.gif

Link to comment

quote:
I'm curious, and I direct this to the folks that do put away their GPS 30-40 feet from the coordinates: Do you find you are closer than said distance to the cache most of the time, or further away than the said distance from the cache most of the time?

 

It's about equal. Sometimes I hit 30-40 feet away and I'm right on the cache and other's I'm 50-60 feet off. Most of the time however, I can spot the hiding place immediately, either way.

 

"It has been my experience that folks who have no vices have very few virtues" -Abraham Lincoln

Link to comment

quote:
Originally posted by smithdw:

Lookup nearby benchmarks that have the coordinates set by the 'professional' grade GPS that surveyors use. Their accuracy is several orders of magnitude more accurate than the commercial grade units.

You can access the detailed list http://www.ngs.noaa.gov/cgi-bin/ds_radius.prl


 

OR, you can go to http://www.geocaching.com/mark/ and put in your zip code. Just keep in mind that if the description says that "location is SCALED", then the coordinates may not be all that accurate. Look for benchmarks that say "location is ADJUSTED".

Link to comment

quote:
Originally posted by Kerry:

quote:
Originally posted by TotemLake:

.... where the Magellan users will keep their GPS out most of the time because it will direct them closer with the autoaveraging.


 

No there's no real truth in that statement these days due to many reasons and many different variables at different times.

 

In affect (despite what the manual might says) autoaveraging can actually make the position worse.

 

Autoaveraging is one of those things that if it feels better using it (but then one really doesn't have a choice with some units) then do it but it's really only a feel good thing as there's nothing to indicate at the time if the results are better or worse.

 

Cheers, Kerry.

 

I never get lost icon_smile.gif everybody keeps telling me where to go icon_wink.gif


 

The only reason I can see why it would be worse instead of better is if you would find yourself in a very narrow window of satellites instead of the optimal wide geometry of same.

 

My observations on the averaging is it takes the normal swings of the estimated position and begins to settle it down to a much more narrow position.

 

Yes, there will still be swings, largely due to the signal being handed off from one satellite to the next as it goes over the horizon and I have seen swings as large as 350 miles but I don't think this is just a Magellan problem.

 

Granted, I haven't the means to compare the Magellan against the Garmin to say which is better, but I can say I am confident that the laws of averaging are going to get me closer to the target more times than without it.

 

Cheers!

TL

Link to comment

quote:
Originally posted by TotemLake:

The only reason I can see why it would be worse instead of better is if you would find yourself in a very narrow window of satellites instead of the optimal wide geometry of same.

 

My observations on the averaging is it takes the normal swings of the estimated position and begins to settle it down to a much more narrow position.

 

Yes, there will still be swings, largely due to the signal being handed off from one satellite to the next as it goes over the horizon and I have seen swings as large as 350 miles but I don't think this is just a Magellan problem.

 

Granted, I haven't the means to compare the Magellan against the Garmin to say which is better, but I can say I am confident that the laws of averaging are going to get me closer to the target more times than without it.


 

TL, Averaging say a narrow window of satellites (like in a creek, canyon etc) will still only be an average of the available data. If the geometry is weak, the position solution will generally be weak and so will any average of that position solution. Averaging bad data basically only produces a bad data average.

 

It's hard to manage those type of swings, which in reality aren't all that common but that's where the other 5% of the accuracy spec fits in.

 

These days the position solution track is slower moving and if it's wandering around say X number of metres from the real world coordinates then it takes a little while to make much of a difference as far as averaging is concerned.

 

It's not as if the position solution is X metres north then suddenly X metres south then X metres West then X metres East as that simply doesn't happen, which is why averaging is an unknown quanity.

 

One might start averaging when X metres north and if after several minutes of averaging the position solution has been slowing wandering further north then the average is worse than the first initial reading. Of course if the reverse applied then the final position might be better but one doesn't know this at the time.

 

quote:
....I am confident that the laws of averaging are going to get me closer to the target more times than without it

 

In GPS terms those laws of averaging are 50/50.

 

Cheers, Kerry.

 

I never get lost icon_smile.gif everybody keeps telling me where to go icon_wink.gif

Link to comment

The idea of studying, and learning the behavior of your receiver is a great one. To be honest, one of the main reasons I geocache is to play with, or compare my receivers.

 

Having found a full hundred caches now using both my Legend and sportrak side by side, I've found the auto averaging to be a mixed blessing. Sometimes, the more stable position is easier for me to work with, and if I give it a fair bit of time in flat heavy tree cover it seems to help more often than it hurts. In the canyons, the averaging coupled with the heavy filtering seems to kill accuracy unless you're willing to let the unit set for at least 10 minutes or more after you stop. Personally, I'd like to have the option of turning my averaging off on the Sportrak, and have the option of averaging on the Legend.

 

What I've come to consider the biggest problem with auto averaging is that it seems to reduce my ability to predict, and appropriately respond to possible reception problems. Especially in canyons, or heavy trees, I've found the sat strength bars, EPE numbers etc aren't always very good indicators of how accurate you can expect the receiver to be. The stability of the Altitude and position however seem to tie in a little better. If it's jumping around a lot, accuracy will most likely not be so great. I often find that moving location just a bit, or changing the way my receiver is positioned will make a big difference. Being unable to turn off the averaging I don't have this option with my sportrak.

Link to comment

Hi Kerry and Searching_ut,

 

All valid points and mind you, I don't take averaging as gospel written in stone as well as EPE and satellite signal strength.

 

I have seen conditions where each of these can provide grossly inaccurate readings. Here's what I have seen thus far using this particular unit in optimal conditions - and remember, your mileage may vary:

 

1. The average cache hider has averaged between 25-35 EPE with their coordinates.

 

2. If the hider is more meticulous, they probably used a known benchmark to gain smaller EPE averages as I have noticed EPE 10 feet or less usually had a benchmark nearby. I used one to gain specific information on how far off mine was when locating the final piece to a multicache.

 

3. Until somebody comes up with a better way of gauging accuracy i.e. EPE, satellite strength, and yes, autoaveraging - we do end up using the tools available to us to gauge.

 

Now, I have read 3 approaches to hunting down caches.

 

1. Within 30-40 feet, some folks are prone to stop looking at the GPS and begin searching. With more experience, I may employ this same technique.

 

2. Some folks rely solely on the GPS and complain the cache was too hard to locate.

 

3. Some folks like to try to zero in on the coordinates, then begin their search just to see how far off their GPS really is. I live in this group.

 

You can throw numbers and technicalities at me all day long and it won't mean much as long as when I look at my GPS at ground zero, and at my feet, is the cache; and I will add, more than 50% of the time. I gain satisfaction at finding the cache in this manner than in the hunt and peck style of #1 while keeping in check the reality of #2.

 

I won't be one of those yahoos that completely trust the gadget, but I will be one to know how to work and read my GPS when the chips are down and not wait for the emergency to happen upon me before I crack open the manual to figure out how to do something. icon_wink.gif

 

Cheers!

TL

Link to comment

quote:
Originally posted by nincehelser:

It's not very often that I start at "ground zero" with my search. Often as I get within 50 or so feet of the coordinates, I start looking for obvious hiding spots ahead of me, and find it quickly.

 

It's only when it's hidden really well do I have to resort to spiraling out from the zero point. That's not too often.

 

George


 

You're right, of course. As I approach the cache location, I am scanning the area to look for obvious spots. Often this does turn up the cache before arriving at 'ground zero'. If it doesn't, I do a quick scan to find any obvious places I missed. I then begin a more complete search.

Link to comment

quote:
As I approach the cache location, I am scanning the area to look for obvious spots. Often this does turn up the cache before arriving at 'ground zero'.

 

I remember one cache I visited (in England)... it was along a well defined path (hedges / walls / fences on both sides!) and at 400 metres from the cache, I looked ahead a guessed 400 metres.... "ah, look there's a lone tree down there, I bet it's within a couple of feet of it."

 

It was!

 

My wife uses a cache box to take her sandwiches to work... how odd!

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...