What accuracy should I expect?

[+jpburcham]

I've already located over 20 Caches, but still consider myself a newbie at this.

Sometimes, I'm out looking, and my Garmin Venture Cx handheld puts me "spot on" the coordinates,

and other times, I find the cache 50' or so from my displayed coordinates.

I realize that not every GPS will read the same, but what is a reasonable accuracy to expect?

What are the rest of you getting?

 

Any tips / hints / etc. are appreciated.

Thanks in Advance

{ Joe }

[+Ike 13]

Depends on the day, the time, the location, and of not only your GPS but also the GPS the hider was using. On average you should expect 20 to 30 feet off error. But I have notice experienced hiders with new cachers are doing a great job of getting coordinates and I find a lot more that are within 10 feet then ones not within 10 feet.

[+frroy]

Depends on the day, the time, the location, and of not only your GPS but also the GPS the hider was using. On average you should expect 20 to 30 feet off error. But I have notice experienced hiders with new cachers are doing a great job of getting coordinates and I find a lot more that are within 10 feet then ones not within 10 feet.

 

With WAAS/EGNOS turned on my Garmin Oregon 550 shows an accuracy of 3m/10ft.

[+Gitchee-Gummee]

True "accuracy" is not the issue, really.

 

A GPSr will put you usually within 20 ft. or so of the coords that you direct it to take you to. Some variance is allowable because -- put simply, those satellites the signals are coming from are a LONG ways away and always on the move besides. There are other, less tangible reasons why accuracy could be off, but those are come-and-go situations.

 

One thing to consider about perceived "accuracy" is that the coords for the cache you are seeking may not be dead on. Original placement could have had an error factor existing at the time, plus over time, many caches seem to "migrate" from one spot to another.

With practice, you will probably hit some caches that are dead-on, it's simply a matter of finding a VERY well camouflaged hide!

 

Any current generation "consumer-grade" GPSr will should show a minimal difference of .001/.002 lat and/or long difference, if any at all. Generally that should be considered to be 6 - 15 ft. (that too, varies depending on your location on this here Earth).

 

End result, each cache placement will be unique in its' own right regarding your "accuracy" factor. Within 20 ft. is good, 30 ft. is OK, 40 ft. is getting a little "pushy".

[+StarBrand]

Our handheld units have a general accuracy in the 15 to 25 foot range most of the time. Don't forget that the hider of the cache also had some unknown error. Generally speaking you will find caches within 30 feet or far less of where your unit says ground zero is but it would not surprise me to have to expand that radius a bit.

[+briansnat]

Even if you had a $6,000 surveyor grade unit that is accurate to a centimeter you will still find yourself off most of the time because the cache owner's unit had a degree of inaccuracy.

 

Being that you don't, your unit will have a degree of inaccuracy, as did the unit of the cache hider, so you will rarely find yourself dead on the cache. Even if you do it may only mean that your unit was off just as much as the cache hider's.

 

For the majority of my finds my unit read about 10-25 feet away. 30, 40 or 50 feet off is not unheard of though, nor is under 10 feet, though I can probably count the number of times my unit zeroed out at the cache on one hand.

[+Harry Dolphin]

I find 25-30 feet at worst, to be acceptable. More than that, I'll post new coords! (If I find it.)

[+dfx]

Between our two Oregons, usually we're able to narrow GZ down to a very small area, leaving maybe 2-3 meters uncertainty. If we find the cache to be well outside that area, with the GPS usually showing a distance of more than 5-6 meters, I consider the coordinates to be bad. All that assuming that reception conditions are somewhat good.

[canadave]

Well, it's fortuitous that this thread came along when it did.

 

My wife and I are fairly experienced geocachers, and we also use a Venture CX. For years it's never been too bad in terms of accuracy.

 

However, the last few times we've been out, we noticed that the GPSr led us pretty far from where the cache coordinates actually were. In one case it was off by about 20 meters. This is in pretty open ground/sky, too. It basically rendered the GPSr unusable--we had to just "get in the vague area" and then hunt around within a large search radius for anything that would likely be a cache hideout. Fun once in a while, but not all the time--especially for micros

 

I just looked at it outside after letting it acquire satellites, and the readout showed "+/- 30 meters" accuracy, with 5 satellites acquired!

 

Is this normal behaviour? Can a GPSr "go bad" over time? It's just so bizarre that the unit would stop functioning normally after three years or so. We tried replacing the batteries, no effect. Everything works fine except the unit is simply inaccurate. Not sure where to go from here with this. I've read other threads that mentioned inaccuracy in the cx units as well, but no one put forward a reason for the inaccuracy

Edited March 19, 2011 by canadave

[+Gitchee-Gummee]

Well, it's fortuitous that this thread came along when it did.

 

My wife and I are fairly experienced geocachers, and we also use a Venture CX. For years it's never been too bad in terms of accuracy.

 

However, the last few times we've been out, we noticed that the GPSr led us pretty far from where the cache coordinates actually were. In one case it was off by about 20 meters. This is in pretty open ground/sky, too. It basically rendered the GPSr unusable--we had to just "get in the vague area" and then hunt around within a large search radius for anything that would likely be a cache hideout. Fun once in a while, but not all the time--especially for micros

 

I just looked at it outside after letting it acquire satellites, and the readout showed "+/- 30 meters" accuracy, with 5 satellites acquired!

 

Is this normal behaviour? Can a GPSr "go bad" over time? It's just so bizarre that the unit would stop functioning normally after three years or so. We tried replacing the batteries, no effect. Everything works fine except the unit is simply inaccurate. Not sure where to go from here with this. I've read other threads that mentioned inaccuracy in the cx units as well, but no one put forward a reason for the inaccuracy

We are entering a time of heavy sunspot activity. It may have an effect on GPS signals.

 

BUT, what you are seeing with the +/- 30 meters, is only ESTIMATED. If could be better or worse, in reality. Ya get to go with whatcha got!

 

EDIT: Don't forget to take into account that the cache was hidden by GPSr coords that were probably just as bad, or as good, as yours. It makes for a compound differential.

Edited March 19, 2011 by Gitchee-Gummee

[+StarBrand]

20 meters off is fairly extreme.

 

Check to make sure your unit is still set for WGS84 datum and the decimal minutes format DDD MM.mmm

[+E = Mc2]

You should expect (to answer your question) roughly 3m/10ft in theory. In practice, I consider any cache I find within 30 feet of the posted coordinates (according to my GPSr) to be acceptable coords. The tolerances between units are such that it's possible to have two good units with good reception be off 15" in either direction easily.

 

...and disregard that 'Accuracy' notation on your GPSr. If you're getting accuracy readings in the high double or even triple digits, then there's a problem. Other than that, it means little most of the time.

 

For some reason, it irks me when I see "My GPSr was 10' of accuracy when I hid this". Ditto "I averaged this 2000 times with 8' accuracy".

 

I still prefer the old Magellan's 'accuracy' notation: EPE, or Estimated Placement Error.

[+DazeDnFamily]

I posted a similar answer a few minutes ago, but this topic is more refined, so I'll add a few cents here. Here are some things to keep in mind:

 

1. You can not really say a coord/map/etc. is "accurate". They are accurate -to a scale-. A coord may be dead-on at 1:500k, but appear horribly off at 1:10k. Likewise, features on a map may shift, depending on the scale you are viewing them at. Also, mapmakers have "Cartographer's License", which allows us to move features if we need to.

2. When discussing the relative accuracy of a coordinate/map/etc., it is common to use the terms "circular error" and "linear error", and reference a percentage. It will look something like this: CE90 10 meters. What that means is that 90% of the time, that coordinate will be within 10 meters of the ground truth location. CE refers to horizontal (X,Y) accuracy. LE refers to vertical (Z) accuracy.

3. The best topographic maps are only "accurate" to about 50 meters, unless they are specifically created to be more accurate. This accounts for CE, LE, cartographer's license and translation from globe to paper (called projection in mapmaking terms.) I've been in the imagery/mapping business for over 13 years, and the best maps (I use the term map generically) I've ever seen were 1 ft zero visibility navigation charts (technically, a chart is a map with navigational information on it) of San Fransisco harbor. They were custom made, and took a LONG time and a LOT of effort, including guys on the ground with tape measures. But, as an infantryman, I was trained to call for fire support using 1:50k topo maps that had an assumed accuracy of about 50 meters. (We also had to use map/compass, and the first generation of GPS units [circa 1996] to find ammo cans in the woods. LOL Funny how I do that for fun now.....)

4. Starbrand referenced this, but the datum you have selected can make a HUGE difference in where the coord plots. Again, this doesn't mean the coord isn't accurate. It means it isn't accurate in the datum you are trying to use. BTW, a datum is a mathematical model of the surface of the Earth. A common datum used for maps in the US was the North American Datum 1927 (commonly written NAD27). The center point of the NAD27 datum is on a farm in Kansas. If your map data is older than 1984, it could well be in that datum. (Notice I didn't say if your -map- is older. If the data that was used to make your map is in NAD27, you will prolly need to convert it, or switch the datum. If you have a paper map, it should be marked on the map.) The World Geodetic System 1984 (WGS84) was created in 1984, after the GPS system became more reliable. (They had WGS 82 as well, but it is rare to see it.) The "ball" at the center of the WGS84 datum was about 10cm, in 2002. (That was the last formal training I went to.) It's prolly better that that now. If you think of the Earth as a sphere (datums generally average out mountains and valleys and form a spheroid. It isn't really a sphere, because it's "smushed" down some along the polar axis, and wider at the Equator) the absolute center of the spheroid is accurate to about 10cm. Also, a datum can be horizontal (WGS84) or vertical (mean sea level). The GPS system defaults to WGS84, and I'm pretty certain that any digital maps you find that work on any modern GPS will be in WGS84 as well.

 

Now, most cachers will never need to know the info here. But, if anyone was curious, now you know. [] Hope nobody glazed over too hard.

 

Later!