Side By Side

[+geojibby]

I recently got a Garmin GPSMap 60Cx to replace my Garmin eTrex H. Tonight I went for a walk to test them out side by side. My old eTrex H was registering 2m of accuracy while the 60Cx was jumping between 4 and 3m accuracy. Now with the huge hardware upgrade in the 60Cx, shouldn't it have better accuracy than the little eTrex H?!??! Both are high sensitivity with WAAS (enabled on the 60Cx at time of test). Both were in identical conditions in bald prairie (no cover overhead).

 

Any idea about the difference in registered accuracy?!

[+dakboy]

There's a world of difference between estimated accuracy calculations and actual accuracy. You can't compare the ± 2m on one unit to ±4m on another because they may be doing the calculations completely differently. The eTrex may be overly optimistic about its calculations, and the 60 more realistic.

 

Find a known good reference point (horizontally adjusted benchmarks are good for this, find one nearby - do NOT use a "scaled" benchmark) and check the coordinates that each device registers at that location, preferably allowing it to average the waypoint over several readings.

[MtnHermit]

Two thoughts:

 

1. You're dealing with EPE, not accuracy.

2. Forum members are obsessive compulsive about accuracy, to what purpose?

 

A few days ago I walked the center of my driveway, ~1000 feet, to record a track. I used both the Colorado and eTrex Venture Cx. The two tracks were offset ~20 feet, but more important to me the Venture's track was on top of itself going and coming, the Colorado was not. New tech would seem not to be better tech.

[+dakboy]

2. Forum members are obsessive compulsive about accuracy, to what purpose?

See also: "mine's bigger than yours", "my dad can beat up your dad".

 

A few days ago I walked the center of my driveway, ~1000 feet, to record a track. I used both the Colorado and eTrex Venture Cx. The two tracks were offset ~20 feet, but more important to me the Venture's track was on top of itself going and coming, the Colorado was not. New tech would seem not to be better tech.

Repeatability is good. I'd rather have a unit which gives consistent results than one that shows a 30' difference from one day to the next.

[MtnHermit]

2. Forum members are obsessive compulsive about accuracy, to what purpose?

See also: "mine's bigger than yours", "my dad can beat up your dad".

So true.

[+Cache'n Hounds]

Two thoughts:

 

1. You're dealing with EPE, not accuracy.

2. Forum members are obsessive compulsive about accuracy, to what purpose?

 

A few days ago I walked the center of my driveway, ~1000 feet, to record a track. I used both the Colorado and eTrex Venture Cx. The two tracks were offset ~20 feet, but more important to me the Venture's track was on top of itself going and coming, the Colorado was not. New tech would seem not to be better tech.

 

Correct me if i am wrong but aren't EPE (Estimated Position error) and accuracy the same thing??? Just curious not trying to sound condescending or anything. Just thought it would be helpful to know the difference.

Thanks

[+dakboy]

Correct me if i am wrong but aren't EPE (Estimated Position error) and accuracy the same thing??? Just curious not trying to sound condescending or anything. Just thought it would be helpful to know the difference.

EPE is "you are here, give or take this many feet"

 

Accuracy is "You are here. But the GPS says you're over there."

 

To determine accuracy, one needs to know what the correct answer is (when you google "define accuracy" the first result is the quality of being near to the true value). The GPS alone doesn't know that - it can only estimate its own confidence in its calculations. Which is why, to determine how "accurate" your GPS is, you need to go to a known location and then compare the GPS readings to that location.

 

Not related to Garmin devices, but here are the lengths some people have gone to in demonstrating/testing the accuracy and repeatability of their DeLorme PN-40s. http://forum.delorme.com/viewtopic.php?t=17523

Edited March 24, 2009 by dakboy

[MtnHermit]

Correct me if i am wrong but aren't EPE (Estimated Position error) and accuracy the same thing???

The key word in EPE is estimated. The estimate is a mathematical model of what that unit thinks it's circle of confusion is. Is the probability 100% that you are inside that circle? Unlikely. Okay is it 50% certainty, if so who cares? So each unit has its own model and probability.

 

From my perspective, the only way to gauge accuracy is from a surveyed benchmark. The EPE is simply a marketing toy. But, I use it too, so who's perfect.

[+julianh]

Correct me if i am wrong but aren't EPE (Estimated Position error) and accuracy the same thing??? Just curious not trying to sound condescending or anything. Just thought it would be helpful to know the difference.

Thanks

No, they are NOT the same thing; however, EPE can be loosely taken to be the unit's estimate of the current probable accuracy and precision. (It might sound pedantic in "simple English", but if we are going to have a discussion about GPSr "accuracy", we need to be speaking a common language!)

 

Firstly, you need to understand the difference between "accuracy" and "precision". This wikipedia article http://en.wikipedia.org/wiki/Accuracy_and_precision expresses it better than I can, but in summary:

 

"Accuracy" means "proximity to the true value", or how close does the average instrument reading compare to the true value. (I.e. is the scatter in readings random in all directions, but averaged on the true value, or is there a general "bias" away from the true value.)

 

"Precision" means how "consistent" and "repeatable" are the readings. (Small scatter = high repeatability = high precision.)

 

For example, an analogue watch which keeps very good time and which is set by the owner every morning at 8:00 am would be both "accurate" and "precise" (for normal consumer purposes anyway, but maybe not to an atomic physicist!) If you neglect to reset it regularly, it will gradually acquire an error, gaining or losing a few seconds each day. In this case, its accuracy is reducing over time (e.g. it might be out by a minute or more after a month), but its precision is still very good (day to day readings are consistent within a second or so). However, if after resetting it, the minute and second hands fall off, it would still be very accurate but it would have low precision, as you can only tell the time of day to about plus or minus 10 minutes or so.

 

With a GPSr, if you go back to the same spot (with known true coordinates) many times, and record the displayed coordinates on each occasion, you can get some indication of your unit's accuracy and precision (at least inasmuch as they can be measured at that particular location - accuracy and precision at other locations, or at other times of day, with different reception conditions may be different!)

 

If your coordinates are generally very tightly clustered, this indicates high precision. If they are loosely scattered, this would indicate relatively low precision. If the mean location (the centre of the scattered group) is dead on target, this would indicate high accuracy. If the mean location is biased away from the true location, this indicates relatively lower accuracy.

 

Finally, "EPE" is Garmin's terminology for "Estimate of Position Error". It is no more than an estimate, and it is based upon the instrument's self-diagnosis of the quality of the signal it currently has. It is based upon a proprietary algorithm (Garmin don't state exactly what it is reporting), and it is NOT a guarantee of accuracy or precision. It is generally taken as being the 50% CEP (Circular Error of Probability), meaning that the instrument estimates that you have a roughly 50% chance of being within the displayed EPE of your true location, and a 50% chance of being further away. There is no theoretical upper limit as to how far away you might be at any given time. (I.e. the true error can be much, much higher than the displayed EPE, although most of the time, the displayed EPE is a reasonable estimate of the current probable accuracy and precision.)

 

Unfortunately, seeing the displayed EPE doesn't tell you what the real current error is, it only provides a guide.

 

Hope this helps!

[+geojibby]

That does help, thanks. I guess I need to find a surveyed benchmark to test it out. In actuality though, we are talking about the difference of a meter – even if looking for a micro, a meter isn't that bad! heheheh..... Thanks for your comments!

[+kallt_kaffe]

I recently got a Garmin GPSMap 60Cx to replace my Garmin eTrex H. Tonight I went for a walk to test them out side by side. My old eTrex H was registering 2m of accuracy while the 60Cx was jumping between 4 and 3m accuracy. Now with the huge hardware upgrade in the 60Cx, shouldn't it have better accuracy than the little eTrex H?!??! Both are high sensitivity with WAAS (enabled on the 60Cx at time of test). Both were in identical conditions in bald prairie (no cover overhead).

 

Any idea about the difference in registered accuracy?!

 

I'm not so sure that the Etrex H is actually older than the 60Cx.

 

Isn't the eTrex H series using the a patch antenna with high sensitivity GPS chip while the 60Cx has a quad helix antenna and a "standard" SIRFstar chip?

 

In other words, your 60Cx has a better antenna but an older (but well reputated) chip, while the eTrex H has a simpler antenna with a more sensitive chip. There are likely cons and pros with both combination and they will likely perform differently under different conditions. I think the 60Cx will be less likely to get confused by reflected signals and may be more trustworthy in the long run.

[+dakboy]

Finally, "EPE" is Garmin's terminology for "Estimate of Position Error". It is no more than an estimate, and it is based upon the instrument's self-diagnosis of the quality of the signal it currently has. It is based upon a proprietary algorithm (Garmin don't state exactly what it is reporting), and it is NOT a guarantee of accuracy or precision.

EPE is not a term proprietary to Garmin. My Magellan MeriGold (firmware from 2001) reports EPE (and labels it as such) as well.

[roybassist]

Isn't the eTrex H series using the a patch antenna with high sensitivity GPS chip while the 60Cx has a quad helix antenna and a "standard" SIRFstar chip?

If by “standard” you mean not high-sensitivity, then no, that is not correct. The SIRF chips in the 60/76C(S)x series are high-sensitivity receivers. Earlier, non-x 60/76 models were not. Maybe that’s what you were thinking of.

 

When Garmin started putting other (non-SIRF) high-sensitivity chips in their H models, I saw several reviews comparing the performance of the new H receivers to the 60/76C(S)x series. I don’t have time to look for them now, but you can probably still find them with Google. As I recall, they pretty much unanimously concluded that if there was any difference in performance at all, it was very slight. In some cases the H units would do better than the SIRF-powered units, in other cases worse.

 

So although the high-sensitivity chips that Garmin is putting in its newer models may be newer, they aren’t necessarily better. There has been much speculation that Garmin switched the type of chips, not because of any difference in performance, but rather for cost savings. Many who have experienced the “drift” problem with Garmin’s newer receivers have expressed the wish that Garmin had stayed with the SIRF chips, which have never, to my knowledge, exhibited that problem. This may be part of the reason for the almost fanatical loyalty that many have for the 60/76C(S)x series.

[+kallt_kaffe]

Isn't the eTrex H series using the a patch antenna with high sensitivity GPS chip while the 60Cx has a quad helix antenna and a "standard" SIRFstar chip?

If by “standard” you mean not high-sensitivity, then no, that is not correct. The SIRF chips in the 60/76C(S)x series are high-sensitivity receivers. Earlier, non-x 60/76 models were not. Maybe that’s what you were thinking of.

 

When Garmin started putting other (non-SIRF) high-sensitivity chips in their H models, I saw several reviews comparing the performance of the new H receivers to the 60/76C(S)x series. I don’t have time to look for them now, but you can probably still find them with Google. As I recall, they pretty much unanimously concluded that if there was any difference in performance at all, it was very slight. In some cases the H units would do better than the SIRF-powered units, in other cases worse.

 

So although the high-sensitivity chips that Garmin is putting in its newer models may be newer, they aren’t necessarily better. There has been much speculation that Garmin switched the type of chips, not because of any difference in performance, but rather for cost savings. Many who have experienced the “drift” problem with Garmin’s newer receivers have expressed the wish that Garmin had stayed with the SIRF chips, which have never, to my knowledge, exhibited that problem. This may be part of the reason for the almost fanatical loyalty that many have for the 60/76C(S)x series.

 

I'm guessing, but as I understand a quad helix antenna is supposed to be better than a patch antenna so my guess was that the higher sensitivity in eTrex chip compensates for the lack of a quad helix antenna. Reversing this line of thought would mean that the SIRFstar in the Cx/CSx logically would be a bit less sensitive as they perform more or less equal to the eTrex H series.

 

I've also seen comparisons where the eTrex HCx are comparied to Cx/CSx and what I remember from it seems to be same as you. Performs more or less equal just like you say.

 

I would rather have a 60Cx or CSx than the Legend HCx that I recently bought but the price of the Legend HCx was almost 50% of what I would have had to pay for a GPSMap 60Cx. I haven't had any drift issues yet though (knock on wood).

 

However my original point was that if the eTrex H uses the same chip and antenna as the Legend/Vista HCx then the upgrade to a Gpsmap 60Cx propably won't mean better "estimated" accurancy. Which chip is the most sensitive was just a guess but is a very interesting question.

 

And that Garmin did not base there new line (Colorado/Oregon) on the well performing 60/76 Cx/CSx is a shame. That had something that worked really well. They could just have added some bells and whistles to that (touch screen, paperless geocaching) and they would have had a winner. Now you have to choose if you want bells and whistles OR a well performing GPS.

Edited March 25, 2009 by kallt_kaffe

[+Jeff in Pa]

Correct me if i am wrong but aren't EPE (Estimated Position error) and accuracy the same thing??? Just curious not trying to sound condescending or anything. Just thought it would be helpful to know the difference.

Thanks

No, they are NOT the same thing; however, EPE can be loosely taken to be the unit's estimate of the current probable accuracy and precision. (It might sound pedantic in "simple English", but if we are going to have a discussion about GPSr "accuracy", we need to be speaking a common language!)

 

Firstly, you need to understand the difference between "accuracy" and "precision". This wikipedia article http://en.wikipedia.org/wiki/Accuracy_and_precision expresses it better than I can, but in summary:

 

"Accuracy" means "proximity to the true value", or how close does the average instrument reading compare to the true value. (I.e. is the scatter in readings random in all directions, but averaged on the true value, or is there a general "bias" away from the true value.)

 

"Precision" means how "consistent" and "repeatable" are the readings. (Small scatter = high repeatability = high precision.)

 

For example, an analogue watch which keeps very good time and which is set by the owner every morning at 8:00 am would be both "accurate" and "precise" (for normal consumer purposes anyway, but maybe not to an atomic physicist!) If you neglect to reset it regularly, it will gradually acquire an error, gaining or losing a few seconds each day. In this case, its accuracy is reducing over time (e.g. it might be out by a minute or more after a month), but its precision is still very good (day to day readings are consistent within a second or so). However, if after resetting it, the minute and second hands fall off, it would still be very accurate but it would have low precision, as you can only tell the time of day to about plus or minus 10 minutes or so.

 

With a GPSr, if you go back to the same spot (with known true coordinates) many times, and record the displayed coordinates on each occasion, you can get some indication of your unit's accuracy and precision (at least inasmuch as they can be measured at that particular location - accuracy and precision at other locations, or at other times of day, with different reception conditions may be different!)

 

If your coordinates are generally very tightly clustered, this indicates high precision. If they are loosely scattered, this would indicate relatively low precision. If the mean location (the centre of the scattered group) is dead on target, this would indicate high accuracy. If the mean location is biased away from the true location, this indicates relatively lower accuracy.

 

Finally, "EPE" is Garmin's terminology for "Estimate of Position Error". It is no more than an estimate, and it is based upon the instrument's self-diagnosis of the quality of the signal it currently has. It is based upon a proprietary algorithm (Garmin don't state exactly what it is reporting), and it is NOT a guarantee of accuracy or precision. It is generally taken as being the 50% CEP (Circular Error of Probability), meaning that the instrument estimates that you have a roughly 50% chance of being within the displayed EPE of your true location, and a 50% chance of being further away. There is no theoretical upper limit as to how far away you might be at any given time. (I.e. the true error can be much, much higher than the displayed EPE, although most of the time, the displayed EPE is a reasonable estimate of the current probable accuracy and precision.)

 

Unfortunately, seeing the displayed EPE doesn't tell you what the real current error is, it only provides a guide.

 

Hope this helps!

 

This is an Extremely good explanation!

Thank you for taking the time to post!

[+Jeff in Pa]

Correct me if i am wrong but aren't EPE (Estimated Position error) and accuracy the same thing??? Just curious not trying to sound condescending or anything. Just thought it would be helpful to know the difference.

Thanks

No, they are NOT the same thing; however, EPE can be loosely taken to be the unit's estimate of the current probable accuracy and precision. (It might sound pedantic in "simple English", but if we are going to have a discussion about GPSr "accuracy", we need to be speaking a common language!)

 

Firstly, you need to understand the difference between "accuracy" and "precision". This wikipedia article http://en.wikipedia.org/wiki/Accuracy_and_precision expresses it better than I can, but in summary:

 

"Accuracy" means "proximity to the true value", or how close does the average instrument reading compare to the true value. (I.e. is the scatter in readings random in all directions, but averaged on the true value, or is there a general "bias" away from the true value.)

 

"Precision" means how "consistent" and "repeatable" are the readings. (Small scatter = high repeatability = high precision.)

 

For example, an analogue watch which keeps very good time and which is set by the owner every morning at 8:00 am would be both "accurate" and "precise" (for normal consumer purposes anyway, but maybe not to an atomic physicist!) If you neglect to reset it regularly, it will gradually acquire an error, gaining or losing a few seconds each day. In this case, its accuracy is reducing over time (e.g. it might be out by a minute or more after a month), but its precision is still very good (day to day readings are consistent within a second or so). However, if after resetting it, the minute and second hands fall off, it would still be very accurate but it would have low precision, as you can only tell the time of day to about plus or minus 10 minutes or so.

 

With a GPSr, if you go back to the same spot (with known true coordinates) many times, and record the displayed coordinates on each occasion, you can get some indication of your unit's accuracy and precision (at least inasmuch as they can be measured at that particular location - accuracy and precision at other locations, or at other times of day, with different reception conditions may be different!)

 

If your coordinates are generally very tightly clustered, this indicates high precision. If they are loosely scattered, this would indicate relatively low precision. If the mean location (the centre of the scattered group) is dead on target, this would indicate high accuracy. If the mean location is biased away from the true location, this indicates relatively lower accuracy.

 

Finally, "EPE" is Garmin's terminology for "Estimate of Position Error". It is no more than an estimate, and it is based upon the instrument's self-diagnosis of the quality of the signal it currently has. It is based upon a proprietary algorithm (Garmin don't state exactly what it is reporting), and it is NOT a guarantee of accuracy or precision. It is generally taken as being the 50% CEP (Circular Error of Probability), meaning that the instrument estimates that you have a roughly 50% chance of being within the displayed EPE of your true location, and a 50% chance of being further away. There is no theoretical upper limit as to how far away you might be at any given time. (I.e. the true error can be much, much higher than the displayed EPE, although most of the time, the displayed EPE is a reasonable estimate of the current probable accuracy and precision.)

 

Unfortunately, seeing the displayed EPE doesn't tell you what the real current error is, it only provides a guide.

 

Hope this helps!

 

This is an Extremely good explanation!

Thank you for taking the time to post!

[+Team CowboyPapa]

Two thoughts:

 

1. You're dealing with EPE, not accuracy.

2. Forum members are obsessive compulsive about accuracy, to what purpose?

 

A few days ago I walked the center of my driveway, ~1000 feet, to record a track. I used both the Colorado and eTrex Venture Cx. The two tracks were offset ~20 feet, but more important to me the Venture's track was on top of itself going and coming, the Colorado was not. New tech would seem not to be better tech.

Yeah, and overly obsessed with WAAS, too.

 

Now here is my a few days ago while drivng home I had EPE of ±6 ft and a WAAS indication and all blue bars.

Then about four miles from home the display changed to EPE of ±9 ft and a non WAAS indication and all green bars.

Now I had a decision to make, one of two courses of action based on this change.

1. Consider the information to be useless and continue on home, or

2. Consider the information pertinent, there for a purpose and therefore to be acted upon, so I:

2a. Pulled over an stopped, and

2b. Got on the cellphone and called for a taxi, and

2c. Cellphone again for a tow truck to get my Jeep home.

3. All of the above,

4. None of the above.

[+Team CowboyPapa]

 

Hope this helps!

 

This is an Extremely good explanation!

Thank you for taking the time to post!

I know this julianh. He is spot on.

[NordicMan]

Be careful if you are comparing 2 GPSr's, don't walk around holding them right beside each other (touching or *nearly* touching each other).. there has been discussions about how sometimes a "nearby GPSr" can affect the accuracy of the other one, something to do with radiowave transmissions.. Hold the 2 GPSr's at least a foot apart from each other.

 

Maybe that was more of a problem years ago than nowadays but, hey accuracy is accuracy

[+fizzymagic]

Now here is my a few days ago while drivng home I had EPE of ±6 ft and a WAAS indication and all blue bars.

Then about four miles from home the display changed to EPE of ±9 ft and a non WAAS indication and all green bars.

Now I had a decision to make, one of two courses of action based on this change.

1. Consider the information to be useless and continue on home, or

2. Consider the information pertinent, there for a purpose and therefore to be acted upon, so I:

Posted about it at least 4 times in various forums.

 

My problem is that I cannot understand why you deem it so important that you post it again and again and again...

[+fizzymagic]

forum burp...

 

The reason some of us care about accuracy is for recording the coordinates of caches we have hidden, not about finding caches. Since you seem inclined to listen to julian, maybe he can explain that in words you understand.

Edited March 25, 2009 by fizzymagic

[+Team CowboyPapa]

Now here is my a few days ago while drivng home I had EPE of ±6 ft and a WAAS indication and all blue bars.

Then about four miles from home the display changed to EPE of ±9 ft and a non WAAS indication and all green bars.

Now I had a decision to make, one of two courses of action based on this change.

1. Consider the information to be useless and continue on home, or

2. Consider the information pertinent, there for a purpose and therefore to be acted upon, so I:

Posted about it at least 4 times in various forums.

 

My problem is that I cannot understand why you deem it so important that you post it again and again and again...

To see if in the meantime, somebody has come up with a straight answer. Last time, the answer was that when their WAAS went away, they'd go in circles.

 

Perhaps I'll stop when I can get a straight answer on the usefulness of the WAAS indication.

 

You know, like when the little gas pump icon on your instrument panel lights up - one should take heed.

 

A similar for WAAS OFF or ON? Or just a frill?

 

Yes, I'll accept the offer and abandon this as no answer, just criticism, is forthcoming.

Edited March 25, 2009 by Team CowboyPapa

[+Team CowboyPapa]

Posted about it at least 4 times in various forums.

 

My problem is that I cannot understand why you deem it so important that you post it again and again and again...

Hey, I'm sorry. The realization and the gravity of the situation just came to me and I will stop posting the scenario and do apologize for the repetitiveness.

 

I now realize it's like politics; when the question can't be answered, question the questioner.

[+julianh]

 

Hope this helps!

 

This is an Extremely good explanation!

Thank you for taking the time to post!

I know this julianh. He is spot on.

Thanks TCP! (Not so long ago, I had thought we would never get to participate in a GPS forum thread where we were in agreement! )

 

If you're ever in Australia, drop me a line and I'll happily buy you a beer or three. (My favourite pub in Brisbane is the Belgian Beer Café; coordinates: 27° 28.238'S 153° 1.760'E. Yes, I know it's a bit sad when I have my favourite pubs stored as waypoints!)

[+geojibby]

I'm not so sure that the Etrex H is actually older than the 60Cx.

 

In other words, your 60Cx has a better antenna but an older (but well reputated) chip, while the eTrex H has a simpler antenna with a more sensitive chip. There are likely cons and pros with both combination and they will likely perform differently under different conditions. I think the 60Cx will be less likely to get confused by reflected signals and may be more trustworthy in the long run.

 

In point of fact, you are right. I do believe that my eTrex H unit is "newer" than the GPSMAP 60Cx that I just got. I bought them out of chronological release date time! However, I just assumed that because of the quad helix yada yada it would be a more accurate unit. I will have to go out and find a bench mark with both units and see what we come up with, other wise it's anybody's guess.