Jump to content

Two Identical GPS Units -- Different Accuracies?


Keo1

Recommended Posts

Due to a recent post where I claimed some exceptional accuracy (and luck) when using my GPS unit, and one of the reasons I don't find geocaching all that challenging is because geocaches are so easy to find with a GPS, based on my own experience with my own GPS device. I got to wondering, should a GPS receiver be any different from the same manufacturing variables as all other devices we buy? For example, I have a camera that excels beyond what everyone else has reported for the same make and model of camera. Confirmed through tests of photos as well as audio tracks recorded with it and sent to audiophiles who wouldn't believe an audio recording of that quality came from that make and model of camera. I can now only explain this in manufacturing variables between any two units. Everything just came together the right way that day as that particular camera was going down the assembly line.

 

With my GPS unit, its reporting a 6 ft. accuracy is rather common and locks onto 2 WAAS satellites at full strength and 9-10 others most times whenever there's a fairly unobstructed view of the sky. When I went looking for caches or other benchmarks (I've also used it again to locate another surveyor's benchmark for another friend this winter) it was nearly spot-on when we walked up to where it was. 2 of the 3 Geocaches were within 2 feet of where my GPS told me it was supposed to be. One being right where a friend was standing at the time and I told him he had to move out of the way (the "funny story" post of mine). 2 surveyor's benchmarks were also well within that accuracy.

 

Just by luck of the draw, could I have gotten a GPS unit that excels compared to reports by all others? Are those that are always claiming that nobody should count on more than a 10-20 meter accuracy just happening to get the run-of-the-mill unit that came off the assembly line, or even one of the occasional lemons?

 

The reproducible accuracy with the one I have got me to wondering. It's not my imagination, others who have needed the data from my GPS can also confirm how close it gets to the mark, so I now surmise it has to be something else. My only guess is that I lucked out and got a "gem" instead of a "lemon".

 

I read that the accuracy is much dependent on the quartz crystal inside a GPS device. Maybe some just come with a perfectly tuned crystal by random chance alone, matched with other equally precise circuitry by random chance alone. Others, not so lucky. Has anyone been geocaching with a friend who always uses an identical GPS device and you noticed that yours, or theirs, was always the first to land on the cache or always locks onto more satellites more quickly or stronger?

 

This discrepancy in reported accuracies by others online and my own experience with just the one I own has left me wondering "why".

 

FWIW: Don't assume this "luck" has always held true for me when buying all things. I assure you that my luck in that regard is just as random as anyone else's. I also have a nice lemon-collection from other past purchases :)

Link to comment

No, your unit sounds fairly typical. I would not put much stock into the reported accuracy on the unit. That is what is known as an EPE (Estimated Postional Error). It is a low confidence estimate of the reading on the unit and should be read something like: "the current reading might be within xx feet of the actual coordinates about 60% of the time under the current Satellite geometry and signal strength". Each manufacturer and model varies the formula for calculating this.

 

Since you have absolutely no idea if the hider had the same accuracy or was using WAAS or was using ancient technology, I would chalk it up to happenstance and luck that you found caches with 2 or 3 feet of where your unit said ground zero was.

 

As an experiment, go out into your yard in the middle and mark some spot with a flag, record the coordinates. A few random times per day, go find those same coodinates (as best as you can) and set a new flag. Come back and let us know the results.

 

Most brand new modern high sensitvity units have an average accuracy between 15 and 25 feet (4 to 8 meters or so).

Link to comment

Starbrand, your comments on EPE versus repatability / accuracy are spot on. And you're suggesting the same test I usually do - mark a known spot and return to it several times.

 

But to the OP's more general question: There certainly can be manufacturing differences (and resulting performance difference) between any manufactured devices, even from the same production run.

 

Most folks won't know the difference. Whether you're after a geocache or a benchmark, any GPS that gets you close enough to find the thing is "good enough" -- and most people won't have the opportunity or time to compare multiple units of a given model to spot a few small differences.

 

Anecdotally: Do you recall Rockin Roddy's expeirence with his first and second PN-40? Summarized briefly, he was sure his PN-40 was more accurate than -- well, more accurate than most anyone but the most ardent DeLorme fans around here would believe. He got upset when anyone would mention how unlikely his observations were, or commented that their own PN-40's didn't perform equally well.

 

Then he lost it, and got a replacement. Even he admitted the replacement was good as the original. And when someone found his original and returned it to him, he commented that in side-by-side testing, the original "...locked WAAS most of the day, something my new PN-40 doesn't do nearly as much...I have always known this unit was better and quicker (routing calculations as well as sat lock) than the new replacement..."

(click underlined quote to see original remarks in context)

Edited by lee_rimar
Link to comment

[to: StarBrand] I see. So you're saying that if the antenna is a half centimeter off from being accurately manufactured, or if the quartz-clock circuitry is off by a minute or more a week, then it won't be any less accurate. And that the converse will also never be true.

 

Interesting.

 

When temperatures get warmer though, I will benchmark my mailbox (at -20 to -30 F. I don't bother hiking all the way to my mailbox). This oddly accurate GPS unit has now got me curious to my assembly-line "lemon" and "gem" theory.

 

I'm still curious if anyone else has reported a wide accuracy discrepancy between two well known and well used identical units. This problem exists in reviews and opinions of all other products manufactured. I fail to see how GPS units are magically exempt.

Edited by Keo1
Link to comment

Well, "wide discrepancy" is a bit vague for a reply. Unless you count the folks who bought a GPS and it simply didn't work, out of the box - and they returned it and got one that did. You'll find a good number of folks around here who have stories like that.

 

Other than those epic fails, variations in GPS units might been seen in time to first fix, how well it receives or holds a signal under adverse conditions, how wide the estimated error is. Even with a SINGLE unit, you'll see variations of these details day-to-day. And unless you have a chance to do side by side comparisons over a long time it'd be impossible to quantify.

Link to comment

[to: StarBrand] I see. So you're saying that if the antenna is a half centimeter off from being accurately manufactured, or if the quartz-clock circuitry is off by a minute or more a week, then it won't be any less accurate. And that the converse will also never be true.

 

Interesting.

 

When temperatures get warmer though, I will benchmark my mailbox (at -20 to -30 F. I don't bother hiking all the way to my mailbox). This oddly accurate GPS unit has now got me curious to my assembly-line "lemon" and "gem" theory.

 

I'm still curious if anyone else has reported a wide accuracy discrepancy between two well known and well used identical units. This problem exists in reviews and opinions of all other products manufactured. I fail to see how GPS units are magically exempt.

I seriously doubt the manufacturing tolerances are anywhere close to as big as what you are suggesting.

 

I am sure there are some (very minor) differences in manufacturing that cause accuracy to be affected by a meter or 2 here and there but I do not believe you would see any one unit that would be dramatically better than the average. It seems far more likely to have few bad lemons in the lot that were way off rather than 1 or 2 highly exceptional units. There are just way too many variables in the GPS calculations to allow that to happen.

 

Just to be clear on the experiment. Don't get daily readings on a fixed object. Rather return to the same coordinates as displayed on your unit and place a marker. After a few days you'll typically see a scatter pattern with points somewhere between 4 and 7 meters or more.

 

BTW - the clock calculations in your unit used for location calculations are not very dependant on a quartz crystal at all but rather calculations from the atomic clocks aboard the sats. Your unit is literally as accurate as any atomic clock when it is able to calculate a position!

Link to comment

Due to a recent post where I claimed some exceptional accuracy (and luck) when using my GPS unit, and one of the reasons I don't find geocaching all that challenging is because geocaches are so easy to find with a GPS, based on my own experience with my own GPS device.

 

Humm, quite a claim from someone who has found "0" caches, or do you just don't log them?

:>)

 

Some caches are "easy" and some are just down right difficult. It is not a pinpoint coordinates, more like a 20 foot to as much as a 100 foot circle.

 

On a clouldy day the coordinates may be off 50 to 100 feet, clouds slow down the siginal just a nano second but enough it is different from a sunny day.

Edited by rayt333
Link to comment

Just to be clear on the experiment. Don't get daily readings on a fixed object. Rather return to the same coordinates as displayed on your unit and place a marker. After a few days you'll typically see a scatter pattern with points somewhere between 4 and 7 meters or more.

I fail to see how that's any better than ticking-off a new waypoint for a stationary object every time. Other than that using electronically recorded waypoints is easier and less time consuming than making a bunch of little stakes to poke in the ground for some bizarre ritual.

 

(I'm starting to wonder if most of the people posting advice to this forum can even think clearly. I still think that high post counts are a strong clue to their "armchair-expert" experiences in life. It keeps being proved, time and time again.)

Edited by Keo1
Link to comment

2 of the 3 Geocaches were within 2 feet of where my GPS told me it was supposed to be. One being right where a friend was standing at the time and I told him he had to move out of the way (the "funny story" post of mine). 2 surveyor's benchmarks were also well within that accuracy.

 

 

So what you're actually saying is either...

 

1. Your GPS is almost as accurate as that used by the hiders when they placed those caches?

or

2. Your's and the hider's GPSs are both off by the same margin of error?

Link to comment

On a clouldy day the coordinates may be off 50 to 100 feet, clouds slow down the siginal just a nano second but enough it is different from a sunny day.

A common misconception. Clouds are essentially "transparent" to the frequency used by the GPS system. It was specifically chosen for that reason. A cloudy day will not cause a 50 - 100 foot error.

Edited by Prime Suspect
Link to comment

A common misconception. Clouds are essentially "transparent" to the frequency used by the GPS system. It was specifically chosen for that reason. A cloudy day will not cause a 50 - 100 foot error.

 

Do you have reference to that information? I know from experience that I have taken several trips and traveled at 700 to 800 miles an hour and never left the spot I was standing. When the storms are really close ..... The only way I knew I made those trips was the "fastest speed" and "average speed" showed those numbers.

 

Humm, but the storms or clouds didn't affect it at all? I wouldn't have known that if you hadn't shared your vast knowledge, I have been fooled all this time by that "misconception"

Link to comment

Clouds and storms can affect your GPSr. Just not in the way you think.

 

Heavy clouds can attenuate the signal, but not much. If your GPSr is barely able to distinguish a particular satellite, it is possible that it will drop below detection on a heavily clouded day. Storms, with associated electrical disturbances, can create enough "noise" to obscure a signal momentarily.

 

The GPS signals are digital and either there or not, no "weak" signals.

Link to comment

All these "armchair-expert" speculations aside (though electrical storm activity interfering in the signal being the only one that's plausible so far, due to its broadband behavior) ... except for lee_rimar who posted a 3rd-party hearsay anecdote, I'm still waiting for some people to address the original important question with a first-hand account or accounts of their own.

 

"Has anyone been geocaching with a friend who always uses an identical GPS device and you noticed that yours, or theirs, was always the first to land on the cache or always locks onto more satellites more quickly or stronger?"

 

Without that base of data-points of comparing two identical units behaving the same or different, then the often-claimed base accuracy for all GPS devices by all these armchair-experts is just wild speculation on their part. ... Or more plausible, their trying to justify to themselves why they paid so much for a GPS lemon.

Link to comment

 

"Has anyone been geocaching with a friend who always uses an identical GPS device and you noticed that yours, or theirs, was always the first to land on the cache or always locks onto more satellites more quickly or stronger?"

 

Without that base of data-points of comparing two identical units behaving the same or different, then the often-claimed base accuracy for all GPS devices by all these armchair-experts is just wild speculation on their part. ... Or more plausible, their trying to justify to themselves why they paid so much for a GPS lemon.

 

Again - all you are going to prove is that one of the two GPSs you are comparing is out by a similar amount to the hiders GPS...... unless you are assuming that every cache you try to find has co-ords measured by a survey quality GPS???

Edited by keehotee
Link to comment
Don't get daily readings on a fixed object. Rather return to the same coordinates as displayed on your unit and place a marker. After a few days you'll typically see a scatter pattern with points somewhere between 4 and 7 meters or more.
I fail to see how that's any better than ticking-off a new waypoint for a stationary object every time. Other than that using electronically recorded waypoints is easier and less time consuming than making a bunch of little stakes to poke in the ground for some bizarre ritual.
You're right, there's not much difference between a bunch of tickmarks on a map versus a bunch of stakes in the ground. Just remember the exercise is to identify the size of the scatter pattern over time, not simply average out EPE readings. If you go back to your mailbox and say "today the EPE is 2 metres," you're not recording the same data as "The GPS says it's 2 metres west of where it said yesterday."

 

The reason for doing this is there's no universally agreed to way to cacluate and present an EPE number -- at best it's a fuzzy guess of how "confident" the GPS is at any given moment. A scatter pattern collected over time is a more useful and revealing data set -- more than EPE readings or one cacher saying to another "I think your GPS got closer than mine today."

 

(I'm starting to wonder if most of the people posting advice to this forum can even think clearly. I still think that high post counts are a strong clue to their "armchair-expert" experiences in life. It keeps being proved, time and time again.)
Well, you might not be happy with the answer anyone around here gives you then. If you personally dismiss high post counts (because you think they come from armchair experts) and high numbers of cache finds (because you think caching is too easy), who are you planning to look to for information here? Why even show up to ask questions?

 

But it might happen, sometimes, that folks with high post counts* spend a lot of time working with these things, and consequently writing about them. And those who find lots of caches might know a bit about GPS devices and how to use them.

 

---

* PS: What's a high post count? 2200+ in a something under 10 years averages out to less than one a day :)

Edited by lee_rimar
Link to comment

 

"Has anyone been geocaching with a friend who always uses an identical GPS device and you noticed that yours, or theirs, was always the first to land on the cache or always locks onto more satellites more quickly or stronger?"

 

Without that base of data-points of comparing two identical units behaving the same or different, then the often-claimed base accuracy for all GPS devices by all these armchair-experts is just wild speculation on their part. ... Or more plausible, their trying to justify to themselves why they paid so much for a GPS lemon.

 

Again - all you are going to prove is that one of the two GPSs you are comparing is out by a similar amount to the hiders GPS...... unless you are assuming that every cache you try to find has co-ords measured by a survey quality GPS???

 

This is so obvious, I am amazed that most fail to understand that !

 

The famous claim for "accuracy" is simply that the GPS happens to have the same "inaccuracy" as the placer's GPS ! :)

Edited by Suscrofa
Link to comment

On a clouldy day the coordinates may be off 50 to 100 feet, clouds slow down the siginal just a nano second but enough it is different from a sunny day.

A common misconception. Clouds are essentially "transparent" to the frequency used by the GPS system. It was specifically chosen for that reason. A cloudy day will not cause a 50 - 100 foot error.

 

I am still waiting for the proof of your ststement, everything I find show it does make a difference, sometimes a BIG difference.

Below is an exert of what I found when I tried to get more information about your "misconception"

 

To determine an object's location, the GPS system must receive a radio signal from at least three satellites. Since each satellite emits a unique signal, the receiver can then match the signal to the satellite and its orbital position. Distance from the receiver is then calculated (for each satellite), and from that data, the receiver accurately calculates its geographic position. But just how accurate is GPS?

 

What affects GPS accuracy

 

GPS accuracy is affected by a number of factors, including satellite positions, noise in the radio signal, atmospheric conditions, and natural barriers to the signal. Noise can create an error between 1 to 10 meters and results from static or interference from something near the receiver or something on the same frequency. Clouds and other atmospheric phenomena, and objects such a mountains or buildings between the satellite and the receiver can also produce error, sometimes up to 30 meters. The most accurate determination of position occurs when the satellite and receiver have a clear view of each other and no other objects interfere.

 

Obviously, mountains and clouds can not be controlled or moved, nor can interference and blockage from buildings always be prevented. These factors then, will affect GPS accuracy. To overcome or get around these factors, other technology, AGPS, DGPS, and WAAS, has been developed to aid in determining an accurate location.

 

AGPS (Assisted Global Positioning System) is a system that assists conventional GPS when reception of the radio signal from the satellite is poor or non-existent (line of sight is blocked). To aid in GPS accuracy, the AGPS gains information via a wireless network, such as the GPS receivers on cell towers, to relay the satellite information to the receiver. With this assistance, the GPS doesn't have to calculate the satellite's orbit, which shortens initialization time, and increases battery life.

Differential GPS

 

To further increase accuracy, DGPS (Differential Global Positioning System) technology was developed. Like the AGPS, the DGPS uses a fixed GPS location (such as a cell tower) to send information to the GPS receiver. DGPS, however, looks at both the satellite and the fixed location adjusts for any difference between the two, and then sends that information to the receiver. DGPS is particularly helpful when atmospheric conditions interfere with reception.

 

The most recent innovation in GPS technology is the WAAS (Wide Area Augmentation System) developed by the FAA and DOD to augment GPS for air navigation. Utilizing a network of ground-based stations (WRS or Wide-area Reference Stations) which are protected from the public, WAAS transmits corrections to geosynchronous communications satellites, which then transmit the corrections to the user. WAAS was designed to allow aircraft to rely on GPS for all phases of flights, including precision, or "instrument only" landings. Specifications for WAAS require accuracy of 7 meters or better both vertically and laterally, 95% of the time. In practice, WAAS achieved a lateral accuracy of 1 meter and of 1.5 meters vertically when over the contiguous United States. Read more on how WAAS works at WAAS Explained.

 

Many GPS manufacturers market their products as the more accurate, or having greater sensitivity than their competitors, but the bottom line is that GPS accuracy depends on the GPS technology in use.

Edited by rayt333
Link to comment

Just to be clear on the experiment. Don't get daily readings on a fixed object. Rather return to the same coordinates as displayed on your unit and place a marker. After a few days you'll typically see a scatter pattern with points somewhere between 4 and 7 meters or more.

I fail to see how that's any better than ticking-off a new waypoint for a stationary object every time. Other than that using electronically recorded waypoints is easier and less time consuming than making a bunch of little stakes to poke in the ground for some bizarre ritual.

 

(I'm starting to wonder if most of the people posting advice to this forum can even think clearly. I still think that high post counts are a strong clue to their "armchair-expert" experiences in life. It keeps being proved, time and time again.)

Wow. I was suggesting a highly visual experiment so you could visually "see" the scatter over time instead of a bunch of little numbers on a piece of paper.

 

Believe what you will about my post count but the fact remains that I seriously doubt that your unit is dramatically better than other units (similar or not). I own 14 different GPS units and have taught numerous GPS classes and shown dozens how to use thier units for geocaching. I have read whitepapers about the technology and numerous websites in addtion to my experiences here in the forums. I have attended 2 classes on GPS technologies. I like to think I know what I am talking about when it comes to the "accuracy" of these handheld units.

 

I own 2 Garmin Legend HCx GPS units. The EPE reported on the units does vary even when sitting within a foot of each other but the biggest difference I have seen was 15 feet. However the the displayed coordinates were identical or off by .001 minutes (about 6 feet). So what does that mean?? I know it doesn't mean one was more accurate than the other.

Link to comment
... except for lee_rimar who posted a 3rd-party hearsay anecdote ...
And it's time for the daily whack over the head with a dictionary.

 

If I had remarked "Rod said his old GPS was better than his new one" and left it at that, it would have been hearsay. But posting a pointer to his original words makes it a sourced reference. In case anyone missed/didn't notice the underlined portion of post #3 was actually a clickable link back to the source comment, I'll go back to make that a bit clearer.

Edited by lee_rimar
Link to comment

The only established (out here in the Real World) way of accurately determining how accurate a particular GPS unit is , is to compare that unit's coordinates for a specific Benchmark (Only Benchmarks with with OFFICIALLY ESTABLISHED and ADJUSTED coordinates qualify) , and then only with repeated visits over multiple days.

 

Otherwise you're just FITW.

Link to comment

On a clouldy day the coordinates may be off 50 to 100 feet, clouds slow down the siginal just a nano second but enough it is different from a sunny day.

A common misconception. Clouds are essentially "transparent" to the frequency used by the GPS system. It was specifically chosen for that reason. A cloudy day will not cause a 50 - 100 foot error.

 

I am still waiting for the proof of your ststement, everything I find show it does make a difference, sometimes a BIG difference.

Below is an exert of what I found when I tried to get more information about your "misconception"

 

To determine an object's location, the GPS system must receive a radio signal from at least three satellites. Since each satellite emits a unique signal, the receiver can then match the signal to the satellite and its orbital position. Distance from the receiver is then calculated (for each satellite), and from that data, the receiver accurately calculates its geographic position. But just how accurate is GPS?

 

What affects GPS accuracy

 

GPS accuracy is affected by a number of factors, including satellite positions, noise in the radio signal, atmospheric conditions, and natural barriers to the signal. Noise can create an error between 1 to 10 meters and results from static or interference from something near the receiver or something on the same frequency. Clouds and other atmospheric phenomena, and objects such a mountains or buildings between the satellite and the receiver can also produce error, sometimes up to 30 meters. The most accurate determination of position occurs when the satellite and receiver have a clear view of each other and no other objects interfere.

 

Obviously, mountains and clouds can not be controlled or moved, nor can interference and blockage from buildings always be prevented. These factors then, will affect GPS accuracy. To overcome or get around these factors, other technology, AGPS, DGPS, and WAAS, has been developed to aid in determining an accurate location.

 

AGPS (Assisted Global Positioning System) is a system that assists conventional GPS when reception of the radio signal from the satellite is poor or non-existent (line of sight is blocked). To aid in GPS accuracy, the AGPS gains information via a wireless network, such as the GPS receivers on cell towers, to relay the satellite information to the receiver. With this assistance, the GPS doesn't have to calculate the satellite's orbit, which shortens initialization time, and increases battery life.

Differential GPS

 

To further increase accuracy, DGPS (Differential Global Positioning System) technology was developed. Like the AGPS, the DGPS uses a fixed GPS location (such as a cell tower) to send information to the GPS receiver. DGPS, however, looks at both the satellite and the fixed location adjusts for any difference between the two, and then sends that information to the receiver. DGPS is particularly helpful when atmospheric conditions interfere with reception.

 

The most recent innovation in GPS technology is the WAAS (Wide Area Augmentation System) developed by the FAA and DOD to augment GPS for air navigation. Utilizing a network of ground-based stations (WRS or Wide-area Reference Stations) which are protected from the public, WAAS transmits corrections to geosynchronous communications satellites, which then transmit the corrections to the user. WAAS was designed to allow aircraft to rely on GPS for all phases of flights, including precision, or "instrument only" landings. Specifications for WAAS require accuracy of 7 meters or better both vertically and laterally, 95% of the time. In practice, WAAS achieved a lateral accuracy of 1 meter and of 1.5 meters vertically when over the contiguous United States. Read more on how WAAS works at WAAS Explained.

 

Many GPS manufacturers market their products as the more accurate, or having greater sensitivity than their competitors, but the bottom line is that GPS accuracy depends on the GPS technology in use.

 

Like I said, it's a common misconception. That you found something on the net (I assume) that also makes the same mistake doesn't surprise me.

 

Clouds have virtually* no effect on GPS signals. Links: http://tinyurl.com/ybvgnw8

 

* not noticeable by humans using consumer-grade GPS units.

Link to comment

Well, I guess my post count and Geocache find count makes me only a passable "armchair intermediate" :) , but from reading this thread it would seem Keo1 is going to remain upset until more people start agreeing with him :anibad:

 

Seriously from my (humble & amateur) experiences, it would seem most GPSr manufacturers roll literally thousands of "nearly identical" units off their assembly lines and, the most significant variance would be that yes they all unfortunate do wind up with a few duds in the collection, which hopefully get remedied by their warranty/return/exchange programs. The thought of them also generating a small handful of genious gems that are waay better than the rest is quite unlikely.. in my amateur opinion.

 

.. but I guess in a car analogy I'd be reminded of movies such as Christine and Herbie the Love Bug :anibad: so OK I stand corrected and strange things can & do happen.

Link to comment

Like I said, it's a common misconception. That you found something on the net (I assume) that also makes the same mistake doesn't surprise me.

 

Clouds have virtually* no effect on GPS signals. Links: http://tinyurl.com/ybvgnw8

 

* not noticeable by humans using consumer-grade GPS units.

 

OK lets agree to disagree. I can find info where it states it does make a difference and you can find info it says it doesn't. You are right I am wrong along with the tons of users who have experienced total loss of signal in a heavy rainstorm.

 

Dave Patton, Canadian Coordinator of the Degree Confluence Project had posted the best answer as of this morning…

 

The GPS signals will not be affected by clouds, rain, fog,

snow, etc., because of the combination of their passing

through water vapour, and the signal wavelength meaning

they "pass through" rain etc.

 

On the other hand, water will block/attenuate GPS signals.

For example, you can’t receive GPS signals underwater, although

having a GPS receiver antenna very close to the water surface

may allow some reception.

 

Most cloth that would be covering a GPS receiver’s antenna

will pass GPS signals(e.g. backpack, jacket, etc.), but cloth

with a metallic component will block/attenuate the signals.

 

Cloth that covers a GPS receiver’s antenna and gets wet

(e.g. from rain) can also block/attenuate the GPS signals.

In fact, you don’t even need the cloth – water on the case

that covers the GPS receiver antenna can block/attenuate

the GPS signals. That water could be drops, and/or a film

of water. I’ve seen that happen myself, when using my etrex

Venture in the rain – the signals were weak/blocked, but

a quick drying-off of the GPS case over the antenna brought

the reception back, which then degraded again as water

accumulated on the case. Similarly, GPS signal reception

is degraded in wet forest canopy conditions compared to

when the forest canopy is dry.

 

Not affected by rain, but is affected by water???? I always thought rain was water, what a fool I was???

Rain is not water.

 

:>)

Link to comment

Like I said, it's a common misconception. That you found something on the net (I assume) that also makes the same mistake doesn't surprise me.

 

Clouds have virtually* no effect on GPS signals. Links: http://tinyurl.com/ybvgnw8

 

* not noticeable by humans using consumer-grade GPS units.

 

OK lets agree to disagree. I can find info where it states it does make a difference and you can find info it says it doesn't. You are right I am wrong along with the tons of users who have experienced total loss of signal in a heavy rainstorm.

 

Dave Patton, Canadian Coordinator of the Degree Confluence Project had posted the best answer as of this morning…

 

The GPS signals will not be affected by clouds, rain, fog,

snow, etc., because of the combination of their passing

through water vapour, and the signal wavelength meaning

they "pass through" rain etc.

 

On the other hand, water will block/attenuate GPS signals.

For example, you can’t receive GPS signals underwater, although

having a GPS receiver antenna very close to the water surface

may allow some reception.

 

Most cloth that would be covering a GPS receiver’s antenna

will pass GPS signals(e.g. backpack, jacket, etc.), but cloth

with a metallic component will block/attenuate the signals.

 

Cloth that covers a GPS receiver’s antenna and gets wet

(e.g. from rain) can also block/attenuate the GPS signals.

In fact, you don’t even need the cloth – water on the case

that covers the GPS receiver antenna can block/attenuate

the GPS signals. That water could be drops, and/or a film

of water. I’ve seen that happen myself, when using my etrex

Venture in the rain – the signals were weak/blocked, but

a quick drying-off of the GPS case over the antenna brought

the reception back, which then degraded again as water

accumulated on the case. Similarly, GPS signal reception

is degraded in wet forest canopy conditions compared to

when the forest canopy is dry.

 

Not affected by rain, but is affected by water???? I always thought rain was water, what a fool I was???

Rain is not water.

 

:>)

 

The fact that vapor, water decreases the signal strength does not imply it affects the accuracy ! It's the speed change that can affect it !

Link to comment

Like I said, it's a common misconception. That you found something on the net (I assume) that also makes the same mistake doesn't surprise me.

 

Clouds have virtually* no effect on GPS signals. Links: http://tinyurl.com/ybvgnw8

 

* not noticeable by humans using consumer-grade GPS units.

 

OK lets agree to disagree. I can find info where it states it does make a difference and you can find info it says it doesn't. You are right I am wrong along with the tons of users who have experienced total loss of signal in a heavy rainstorm.

 

No, I won't agree to disagree, because this is not a matter of opinion. You're entitled to your own opinions, but you're not entitled to your own facts.

 

We were discussing clouds and overcast conditions. If you get drops of water, or even a thin film of water on you antenna, then yes that will most definitely have an effect. I have never said otherwise. But clouds? The answer is NO.

Link to comment

Interesting discussion in it’s intensity. As for a few inputs:

 

The quartz crystal in a handheld gps unit is used to keep time for the internal clock within the receiver so it will have a rough time hack when it’s first powered up. It isn’t used in position calculation. A clock accurate enough to be used for position indication would cost more than any of us could afford. You can search and read about the calculated time used by gps receivers.

 

Years ago, I set up instrumentation used to transmit and log satellite data when testing gps devices. From having looked that sort of data over, I'm confident in knowing upper atmospheric conditions have a fairly large effect, but you’re not going to see visually whats going on there to guess if it might affect you or not. Lower atmosphere conditions to include clouds, rain and snow have affects so slight you can barely measure them. It’s not significant enough you’ll notice anything with a consumer grade unit. I haven’t looked for the data in years, but some universities and cors stations used to put out data you could look at to see just how much signal strength varied over time, as well as atmospheric effects that would affect accuracy.

 

Regarding the accuracy rating for gps units is universal across units because it’s based on the atmospheric affects on the signal. There are ways to process the signals differently to try and filter or calculate out those errors, but that is going to remain constant amongst same type receivers and only going to be of difference when comparing different software, or receivers. As mentioned, EPE indications are just wild guess having no way of being based on actual accuracy. They seem to influence peoples perceptions though.

 

Finally, something well known in the test world is the tendency of people so see what they expect to see. Instrument your tests, record lots of data, and trust in the data rather than your perceptions of what you saw. A fun way to see how gps accuracy varies over time is to set up track log recording for time intervals, set the unit in a stationary position, then look at the data to see how it walks around over time.

Link to comment
The only established (out here in the Real World) way of accurately determining how accurate a particular GPS unit is, is to compare that unit's coordinates for a specific Benchmark (Only Benchmarks with with OFFICIALLY ESTABLISHED and ADJUSTED coordinates qualify) , and then only with repeated visits over multiple days.
I agree with repeated visits to a fixed, known location over multiple days. But how is "an established benchmark with officially established/adjusted coordinates" better than any other fixed, known locaction?

 

The only advantage I see with an "established" benchmark is that I know what my GPS should read before I get there the first time. But after that? My mailbox isn't a moving target. If a GPS will routinely give the same coords for that spot (within a couple of metres), I think that's a good enough test.

Edited by lee_rimar
Link to comment

I fail to see how that's any better than ticking-off a new waypoint for a stationary object every time.

I don't get it, either. That approach also adds the error of the numerical accuracy of the coordinate system we use (0.001 minutes). That's 6 feet N/S and at my latitude, about 4 feet E/W. In the N/S direction, you can't get any closer than that without adding some more digits to the right of the decimal.

 

The stationary object makes more sense in any case. What we're interested in here as geocachers is whether, within whatever accuracy radius, a given GPS unit will produce the correct value when located at the cache - since that's the best we can ever hope for with the coordinate system we're using.

 

I just opened a different thread on my Dakota 20 vs. Summit HC to which I will continue to add as the experiments continue. There's something VERY different about what my Dakota is reporting vs. what I've grown accustomed to seeing on my eTrex.

Link to comment

...(I'm starting to wonder if most of the people posting advice to this forum can even think clearly. I still think that high post counts are a strong clue to their "armchair-expert" experiences in life. It keeps being proved, time and time again.)

 

From my armchair, the way I see it the chips and components that go into your GPS have slight varitions that will fit within an acceptable range of tolerance.

 

Those tolerences will come together to make some GPSs slightly better and some slightly worse but all within the acceptable design of the GPS to begin with.

 

This of course has nothing to do with your luck at finding a cache because that tolerance is to put down your GPS and start looking at 20' or so and I suspect that (without using google to verify this) is well outside the variation you are gong to see between GPS units that meet spec.

 

I made all that up from my armchair. Good luck out there.

Link to comment
But how is "an established benchmark with officially established/adjusted coordinates" better than any other fixed, known locaction?
It depends upon whether you are trying to determine accuracy or repeatability. I know a local cacher who would have done well to understand the difference. He had a GPS that was, for whatever reason, adding about 25' to the N position before he replaced it. The repeatability was excellent -- all of his placements for a couple of months were off by the same amount! If he'd done the test to an arbitrary static position, he'd likely have seen good results (repeatability was good). Meanwhile, he was killing the rest of us (accuracy was off by 0.004 north).

 

So a static test to an arbitrary position is just fine for seeing how well a GPS unit does at getting a lock and putting you in the same spot. Nothing wrong with that that test. But ultimately, we're looking at the difference between the CO and finder's GPS units, and accuracy is a real issue. But for caching purposes, knowing of any consistent OFFSET in your own unit is really important.

 

As I noted, I'm seeing some definite differences in the readings between my old eTrex Summit HC and my new Dakota 20. One of those things is what appears to be an offset of 0.003 between them in N/S in side-by-side testing. That it seems to be pretty consistent indicates a repeatable inaccuracy of at least one of them. That isn't a deal killer for geocaching providing you establish the offset of your particular GPS unit since you can always compensate by 6' N/S or (depending upon where you are) about 4' E/W for each 0.001 minutes of repeatable error during your searches.

Link to comment

...The only advantage I see with an "established" benchmark is that I know what my GPS should read before I get there the first time. But after that? My mailbox isn't a moving target. If a GPS will routinely give the same coords for that spot (within a couple of metres), I think that's a good enough test.

 

Exactly. That said, there is an issue of accuracy vs. precision at play that only the known point will help you figure out- if that's an issue. If your GPS is always 12' left for no particular reason, your mailbox will not help you figure that out.

 

Over time you would average out the random error and create a fairly accurate point on the ground that is the "location of your mail box". Then your surveyor buddy would come over and just for kicks give you an actual location of that mailbox. When you compared the numbers you would find the 12' of error. Known locations have the advantage of your surveyor buddy already having done that work for you going in.

 

In survey terms (off the top of my head since i"m not leaving my armchair to figuer it out) this is a systematic error. You are off an easily calucated amount amount every time and so it's easy to compensate for.

Link to comment

...Not affected by rain, but is affected by water???? I always thought rain was water, what a fool I was???

Rain is not water....

 

Think drop size. Coud dropletts are too small to create a problem. However your wet jacket, the drops running off the GPS antantea, the droplets on the trees etc. are.

 

Even if the drop size is an issue 'rain' is mostly empty space, but 'wet' is 100%.

 

Egads man, will I never get out of this armchair today?

Link to comment
one of the reasons I don't find geocaching all that challenging is because geocaches are so easy to find with a GPS, based on my own experience with my own GPS device.
So ignore any cache with less than a 3-star difficulty. While I've walked right up to many caches & had my GPSr zero out within arm's reach & the cache has been incredibly obvious, harder caches should give you more of the challenge you're looking for. I've had several caches where the GPSr zeroes out and I spent 30 minutes or more scratching your head.

 

You're also assuming, as others have pointed out, that because your particular device happens to be very accurate, every cache hider's GPS must perform exactly the same as yours because you just happened to get the same results on a particular day, eliminating all guesswork.

Link to comment
The only established (out here in the Real World) way of accurately determining how accurate a particular GPS unit is, is to compare that unit's coordinates for a specific Benchmark (Only Benchmarks with with OFFICIALLY ESTABLISHED and ADJUSTED coordinates qualify) , and then only with repeated visits over multiple days.
I agree with repeated visits to a fixed, known location over multiple days. But how is "an established benchmark with officially established/adjusted coordinates" better than any other fixed, known locaction?

 

The only advantage I see with an "established" benchmark is that I know what my GPS should read before I get there the first time. But after that? My mailbox isn't a moving target. If a GPS will routinely give the same coords for that spot (within a couple of metres), I think that's a good enough test.

 

Come on Lee, surely with your experience you are smarter than that. If your GPS repeated EXACTLY at your mail box 237 times in a row, that proves ABSOLUTELY NOTHING about accuracy. It would only prove that your unit is consistent with itself. Consistently wrong is just as probable as consistently right.

 

Only when you compare your "consumer grade GPS" coordinates to official Benchmark coordinates ("adjusted") which were established to accuracy tolerances that consumer grade equipment is incapable of achiving, can you start determining the accuracy of your unit.

Also, only checking once, proves nothing. Why? Because "one in a row" does not make a pattern.

Link to comment
The only established (out here in the Real World) way of accurately determining how accurate a particular GPS unit is, is to compare that unit's coordinates for a specific Benchmark (Only Benchmarks with with OFFICIALLY ESTABLISHED and ADJUSTED coordinates qualify) , and then only with repeated visits over multiple days.
I agree with repeated visits to a fixed, known location over multiple days. But how is "an established benchmark with officially established/adjusted coordinates" better than any other fixed, known locaction?

 

The only advantage I see with an "established" benchmark is that I know what my GPS should read before I get there the first time. But after that? My mailbox isn't a moving target. If a GPS will routinely give the same coords for that spot (within a couple of metres), I think that's a good enough test.

 

Come on Lee, surely with your experience you are smarter than that. If your GPS repeated EXACTLY at your mail box 237 times in a row, that proves ABSOLUTELY NOTHING about accuracy...

 

Oh, but I would love that unit. It would have precision unheard of in other GPS's. With a GPS that precice a simple calebration (if your GPS could do that) and you would have a perfectly accurate GPS.

 

When I have a wide grouping when shooting I like to tell my friends that on average I hit the target. When my buddy shows me his tight little grouping off to one side, I like to tell him, he didn't even hit on average. Of coure then he makes an adjustment and nails it, while I need more practice so I can hit in reality instead of on average.

 

There is a relationship between precision and accuracy. You can use it to your advantage. Even with a mailbox.

 

Lunch is over, time to get out of the armchair.

Link to comment

Hmm, several folks offering similar scenarios describing a specific kind of GPS inaccuracy

  • ecanderson: "...consistent offset..."
  • Renegade Knight: "...always 12' left for no particular reason..."
  • GrassCatcher "...Consistently wrong is just as probable as consistently right..."

I don't mean this argumentatively -- I'm really asking informationally. If any of you can expand my knowledge, I will thank you for it.

 

The only ways I know of to get consistent, repeatable, and precisely WRONG answers from a GPS is through user error - like using the wrong datum or other system setting, or just reading the results incorrectly. And I do know all of those are possible from my own direct experience :unsure:

 

What I haven't experienced, and can't think of a cause for, is GrassCatcher's summation that "consistently wrong is just as probable as consistently right." If the device is set up and used properly, my own experience is a "consistent" answer is the right one. With a consumer grade device that best "right answer" can be off by a few metres at any given time. Repeated readings will let you draw scatter pattern, small or large, that shows how good or bad the thing is.

 

When I've done that kind of testing, a small (good) pattern has always been centered on the right place; and a wide (poor) pattern doesn't show enough consistency to look for an offset.

 

But a small pattern showing a "consistent offset"? I haven't seen this and don't understand how it can happen. Other than user error, what kind of conditions bring that about?

 

And repeating my New Years Resolution to be nicer and less compabative -- I really would like to know this. I'm not just asking to be argumentative. Honest!

 

---

edit to add: Renegade Knight, I saw your answer to GrassCatcher after I posted this originally. I see you meant something different from him.

Edited by lee_rimar
Link to comment

But a small pattern showing a "consistent offset"? I haven't seen this and don't understand how it can happen. Other than user error, what kind of conditions bring that about?

Wish I had just a little more info on the innards of these chips. One could only imagine that static offset error is caused by some component within the analog components prior to the DSP. The usual suspects external to the receiver (e.g., multipath errors) aren't static. Further, there's no user error in our region that can produce something on the order of a 25' N static error. Incorrect datum selection (e.g., NAD27) is going to create a less consistent result. Moreover, in our area, the larger errors are much more likely to appear in the E/W direction. That varies by region of the country, of course.
Link to comment

Lee,

No arguments here either. Maybe a difference in my choice of words should have been there's a 50/50 chance of being right (correct as to official coordinates) or being wrong.

 

RK,

I too would love to have a GPS that would repeat exactly "327" times in a row....HA! ...Not likely to happen!

I should have said "Even if....."

 

Lee, see inserted comments.....

" If the device is set up and used properly, my own experience is that a "consistent" answer is the right one. With a consumer grade device that best right might still be off by a few metres at any given time, and repeated readings will let you draw scatter pattern, small or large, that shows how good or bad the thing is.

Now THAT is a measure of your accuracy (small or large)

 

When I've done that kind of testing, a small (good) pattern has always been centered on the right place; and a wide (poor) pattern doesn't show enough consistency to look for an offset.

You have to have accurately established coordinates to "define" what the "right place" is.

 

But a small pattern showing a "consistent offset"? I haven't seen this and don't understand how it can happen. Other than user error, what kind of conditions bring that about?"

I agree, this is the one that is most likely NOT to happen in the real world.

 

By modifying methods and procedures and setup choices, users can noticeably reduce what you are calling "user errors" . That's not necessarily a correct term. Wouldn't you agree that "possible operator induced variations" might be more correct? However, without an established accuracy "baseline" , you wouldn't even know that changes in methods and procedures even need to be made.

 

Example:

Carrying multiple GPSrs at the same time, I've got multiple tracks from multiple GPSrs on exactly the same narrow (18") single track trail. In three separate places around the 5 mile loop, one GPS consistently says the trail is "here" and the other consistently says the trail is "there". The problem appears to be caused by multipath error....it is NOT operator error. I've changed carrying positions , used and not used external antenna, etc. ....I'm still working on it.

 

I'm trying to determine which is "accurately" logging the track. Both units are repeating (with themselves) just Not with each other.....and only in those three locations.

Link to comment

ECAnderson, it sounds like you and your friends never pinned down the cause for that one unit's consistent offset.

 

It's a shame your friend replaced that GPS, it would be interesting to lay hands on it for evaluation. While he had it, did you identify the problem only from his cache coords always being off? Or did you also do some of the other checks we're talking about here -- going to an established benchmark, comparing it against other GPS units at the same time, etc?

Link to comment

But a small pattern showing a "consistent offset"? I haven't seen this and don't understand how it can happen. Other than user error, what kind of conditions bring that about?

More to the point - what kind of USER error could cause this??? No jive - we had a cacher out here in Colorado that was posting new cache locations at darned near exactly GZ 25' north of actual on a consistent basis for several months. Numerous previous caches had been dead on the money (usually within no more than 10') under optimal satellite/weather and WAAS conditions. (He swears he wasn't playing with the numbers!) Traded in his GPS for a new Oregon, and we haven't seen the goofy offsets since. Whatever happened to his GPS unit wasn't there from the beginning -- the problem seems to have appeared overnight after a long period of apparently accurate readings.

 

Two years ago, we had another cacher with a more peculiar error. Just about 50% of his hides would vary 30 to 35 feet either east OR west of posted position (never north or south).

 

The entire local caching community, with our very diverse set of hardware and techniques, pretty well agreed on the numbers above and often commented on it in logs. So something was up with those two GPS units. What? I have no idea.

 

Here's a possibility - is there some chance that if a cacher starts to get a bit hurried (read: sloppy) when placing hides - and takes a single reading instead of taking a decent average - that there is some chance a particular GPS might have a built-in bias in one direction or another before adequate readings are accumulated to produce an accurate result?

Link to comment

......Here's a possibility - is there some chance that if a cacher starts to get a bit hurried (read: sloppy) when placing hides - and takes a single reading instead of taking a decent average - that there is some chance a particular GPS might have a built-in bias in one direction or another before adequate readings are accumulated to produce an accurate result?

Older Magellean units were infoamous for the boomerang effect. They kept you moving in a particular direction of travel even after you stopped moving. Then after a few moment it would realize you had stopped and settle down to correct coordinates. Early Colorado and Oregon Firmware revisions had a similar effect.

Link to comment

ECAnderson, I have to apply Occam's Razor to your most recent observation. User error, misreading, or "hurried/sloppy" work (or dare I say suggest, fudging the numbers just to mess with you?) are all simpler possibilities than a systematic hardware error consistently adding an arbitary offset to the reported position.

 

Repeating what I mentioned earlier -- are all of these observations based on posted coords from a cacher? Or did anyone tag along with him to a known benchmark to evaluate the unit?

Edited by lee_rimar
Link to comment
...Older Magellean units were infamous for the boomerang effect. They kept you moving in a particular direction of travel even after you stopped moving. Then after a few moment it would realize you had stopped and settle down to correct coordinates...
Okay, I'll concede that one as a mix of "hardware error" (the device is gonna read imcorrectly) and "user or process induced error" (a magellan user needed to know to let the device settle before setting a waypoint)

 

Except - it wouldn't explain ecandersons friends always getting errors in one direction, and if you were testing against a known, fixed location (benchmark or mailbox) on repeated visits it wouldn't show up as a consistent, repeatable wrong answer. Wou;dn't a Magellan boomerang get wrong answers that would always be different?

Edited by lee_rimar
Link to comment

....

Except - it wouldn't explain ecandersons friends always getting errors in one direction, and if you were testing against a known, fixed location (benchmark or mailbox) on repeated visits it wouldn't show up as a consistent, repeatable wrong answer. Wou;dn't a Magellan boomerang get wrong answers that would always be different?

Unless you ALWAYS approached a cache hide from (for example) the south - then you might have a "boomerang" error to the north.

 

 

okok - I know - very unlikely :unsure:

Link to comment
ECAnderson, I have to apply Occam's Razor to your most recent observation. User error, misreading, or "hurried/sloppy" work (or dare I say suggest, fudging the numbers just to mess with you?) are all simpler possibilities than a systematic hardware error consistently adding an arbitary offset to the reported position.
I'd buy the "hurried/sloppy", but given the consistent nature of the error, "misreading" is most unlikely. We're talking about someone with literally hundreds of hides, some 75~100 or so of those produced during the "problem period". What other "user error" do you envision that can get these results? Accidental use of NAD27 produces nearly no N/S error here, but produces a massive E/W error (about 0.035) in our area, and that would have shown up immediately and been understood for what it was.

 

The only "user induced" error would be a snap reading, and would require the GPS to have the sort of static bias noted under those conditions. I don't understand the innards of the SiRF Star III well enough to know if that is possible or even likely.

 

Repeating what I mentioned earlier -- are all of these observations based on posted coords from a cacher? Or did anyone tag along with him to a known benchmark to evaluate the unit?
As for the "25'N Syndrome": No, no one tagged along, but the community as a whole commented on the results in their online logs. New coordinates were frequently posted by those that feel comfortable doing so and would take the time, and most often requested an increase to the latitude by 0.004 ~ 0.005 minutes. If the hide was obvious enough (e.g., the imfamous LPC), it was usually ignored, but where there was a lot of "clutter" (more wrong places to look than one could begin to count), the comments would begin to appear early on in the logs. Edited by ecanderson
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...