Garmin GPS watch review

[Explorer Editor]

Thinking of buying a GPS watch? You might be interested in my review of the Garmin Forerunner 405.

Click here to read it.

 

[DougPaul]

A general comment on GPS watches.

The wrist is a poor place to carry a GPS because one tends to swing it back and forth. This movement, which is large enough and fast enough to be picked up by the GPS, adds a significant amount of noise to the GPS position and velocity measurements which decreases its accuracy. GPSes also use smoothing over several measurements (typ one per second in a consumer GPS) to increase accuracy. Such movement tends to defeat the smoothing.

In addition, one's wrist often has a poor skyview due to blocking by one's body.

I have compared carrying a GPS in my hand held stationary with respect to my body and while swinging my arms in a natural way. While I haven't compared tracks, the speed numbers are much steadier in the stationary carry.

Generally, carrying a GPS on the upper arm will be better due to less swinging and a better skyview. And on a pack strap (particularly on the top of the shoulder), in the top of one's pack, or in/on one's hat are even better.

Doug

 

[DougPaul]

Thinking of buying a GPS watch? You might be interested in my review of the Garmin Forerunner 405.

Click here to read it.

On your comment about the total ascent: This GPS does not have a barometric sensor and therefore uses GPS altitude. GPS altitude has errors on the order of 2--3 times the horizontal position errors. And if you add in a few particularly bad altitudes, it isn't hard to see how it could accumulate a pretty significant total ascent when you are moving on the level. It will most likely also accumulate a fairly significant amount of ascent if you simply leave it stationary for an hour or so.

Doug

 

[Kevin Rooney]

I have a mapping GPS (Garmin 60CSx, and before that a Vista), and I use them - on average - a couple of times a week. The most important function for me is the actual map, showing my location on that map.

For a few years before the GPS became widely available with some degree of reliability and accuracy, I used an altimeter watch (Casio) which was useful, especially bushwacking. However, the Garmin models I use have that capability, so an altimeter watch is redundant.

And - all things being equal - I like the lightest, thinnest watch I can find. Right now a basic Timex watch fits that bill. If I need the date, my phone has it.

 

[DougPaul]

And - all things being equal - I like the lightest, thinnest watch I can find. Right now a basic Timex watch fits that bill. If I need the date, my phone has it.

Your phone and GPS can also give you the time...

In fact, the GPS knows the time to ~10 nanoseconds internally. (The time display is probably within a second or two.)


I also like the maps on a GPS, but the Forerunner is aimed more at recording a run than navigation.

Doug

 

[max]

The wrist is a poor place to carry a GPS because one tends to swing it back and forth. This movement, which is large enough and fast enough to be picked up by the GPS, adds a significant amount of noise to the GPS position and velocity measurements which decreases its accuracy. GPSes also use smoothing over several measurements (typ one per second in a consumer GPS) to increase accuracy. Such movement tends to defeat the smoothing.

In addition, one's wrist often has a poor skyview due to blocking by one's body.

Do you have anything to back up either of these statements? Personally, I think they both are a bit far-fetched. This guy did a battery of tests on wrist gps accuracy, and his 98.9% rate while running is pretty good.

http://www.dcrainmaker.com/2010/11/sport-device-gps-accuracy-in-depth-part.html

Hiking in Grand Canyon with my GPS never seems to affect its ability to hold a satellite signal, even while hiking close to vertical cliffs. I can't imagine the small bit of body shielding amounting to a hill of beans here.

 

[DougPaul]

Do you have anything to back up either of these statements?

Of course I do...

1. Signal blocking by the human body
The human body is basically a bag of salt water at GPS frequencies--it absorbs, diffracts, and reflects the signal. Signal degradation due to reduced skyview has more to do with the proportion of the sky that is blocked than the size of the object--what is important is which satellites are blocked and how badly, is there multipath (reflected and delayed signals), etc.

I ran a set if experiments (similar to those published in http://www.vftt.org/forums/showthread.php?t=14406) evaluating the cross track difference between the outgoing track and the incoming track of an out-and-back walk using a pair of 60CSxes:
* antenna on top of the head (ie clear of my body): 1.2 meter avg difference
* antenna in a pack on my back (partially shielded by my body): 3.6 meter avg difference
* The experiments reported in http://www.vftt.org/forums/showthread.php?t=14406 also show an advantage for antennas located on one's head rather than near one's body.

Conclusion: carrying the GPS in a spot where its skyview is partially blocked by the human body degrades accuracy. (FWIW, any knowledgeable GPS user knows to keep the antenna clear of his body if he wants to maximize the accuracy.)

2. Degradation due to irregular motion of the GPS.
The phase-locked loops tracking the GPS signals in any GPS are sensitive to the carrier phase with resolution down to ~1% of a cycle. Since the wavelength of the L1 frequency used by consumer GPSes is ~19cm this means that a GPS can resolve to movements as small as ~2mm* and must respond to movements of at most a few cm to maintain a good lock on the incoming signal. Steady motion of the GPS results in an additional steady component of the doppler shift of the incoming signal and thus can be tracked without difficulty. Unsteady motion of the GPS (eg from attaching to one's wrist and swinging one's arm) requires the phase locked loops to respond to the changing doppler shift which reduces their accuracy.

* The carrier phase is used by survey-grade GPSes to obtain sub-centimeter accuracies. This phase information is not used by consumer GPSes which makes them significantly less accurate.

The basic GPS outputs a raw 7D vector: 3D position, 3D velocity, and 1D time (PVT). Consumer GPSes use Kalman filtering to smooth the PV data based upon the current and past raw PV values. Even if swinging the GPS back and forth doesn't damage the accuracy of the P values, it will introduce randomness into the V values and ultimately result in poorer smoothed values for both P and V. I have observed this randomization of the indicated V values while walking by comparing a GPS held steady in my hand to one swung back and forth in my hand.

As further conformation of this issue, the Garmin 60CSx manual gives the following accuracy spec: "Velocity: 0.05 meter/sec steady state" (emphasis added).

Furthermore, as you swing a GPS on your wrist (or in your hand), the portion of the sky that is blocked by your body will change. This can cause some satellites to be blocked during part of the swing causing the GPS to jump back-and-forth between sets of satellites. It is well known that different subsets of the available satellites will yield slightly different locations so this will cause the indicated position to jump back and forth.

Personally, I think they both are a bit far-fetched.

Only if you don't understand how a GPS really works. It is a triumph of modern GPS receiver technology that that they continue to work well when placed in such degraded operating conditions. Such was not always the case...

This guy did a battery of tests on wrist gps accuracy, and his 98.9% rate while running is pretty good.

http://www.dcrainmaker.com/2010/11/sport-device-gps-accuracy-in-depth-part.html

The only thing this guy tested is the odometer function, not position accuracy as would be required for navigational use.
* He carried the GPSes on a stick out away from his body, so he didn't test them under realistic conditions of skyview blocking and unsteady motion due to arm swing.
* He compared them to a device whose accuracy was not established--thus his absolute numbers are meaningless. (Comparisons to the other GPSes might still have been meaningful as comparisons, but not as absolute accuracy numbers.)

His results do little to nothing to establish the accuracy of wrist-GPSes for navigational purposes.

Hiking in Grand Canyon with my GPS never seems to affect its ability to hold a satellite signal, even while hiking close to vertical cliffs. I can't imagine the small bit of body shielding amounting to a hill of beans here.

As stated earlier, the signal shielding is much more a function of how much and what part of the sky (ie which satellites are blocked) than how physically big the blocking object may be. Both flesh and rock absorb, diffract, and reflect signals. The modern high-sensitivity GPSes are much better than the older non-high-sensitivity GPSes at giving the user a fairly accurate location under degraded signal conditions, but the accuracy is still reduced. The average user will simply not notice that the average error might be 4 meters in the canyon or near your body when the average error might be 2 meters up on the rim with the GPS held steady and clear of your body. See http://www.vftt.org/forums/showthread.php?t=14406 for a set of tests where older GPSes completely lost track under conditions where a newer GPS was able to maintain a reasonably accurate track.

Doug

 

[Oldmanwinter]

It is a triumph of modern GPS receiver technology that that they continue to work well when placed in such degraded operating conditions. Such was not always the case...

Doug

So true, this is the the most significant advance in Civilian GPS technology ever for automotive and mapping GPSr's, Ever!

 

[Remix]

The wrist mount GPS's are optimized for training and exercise. I don't think anyone would want to take a run holding a handheld GPS. Nor will a handheld unit take the vibration on a bicycle handlebars. I had a friend who tried this--AA cells have alot of mass--the battery contacts became intermittent and then something broke on printed circuit board.

And similarly, a wrist mounted GPS probably won't do as well as a handheld for hiking...sensitivity is limited by antenna size and power. Size limits expandable memory for mapping and limits sensor options.

So it really depends on what you want to use the GPS for.

As far as accuracy goes...everything depends on where the satellites are at any given moment as well as receiver design. Its not uncommon to lose all signals in a notch or ravine.

 

[DougPaul]

The wrist mount GPS's are optimized for training and exercise. I don't think anyone would want to take a run holding a handheld GPS.

I tend to view the wrist GPSes as fancy pedometers... (I often throw a GPS in my pack while taking a walk, hike, or ski as an even fancier pedometer.)

Nor will a handheld unit take the vibration on a bicycle handlebars. I had a friend who tried this--AA cells have alot of mass--the battery contacts became intermittent and then something broke on printed circuit board.

A number of people have observed this loose battery problem--the battery holders on some of the older GPSes (eg the original eTrex Vista which used real springs) are much more resistant to this failure mode. One fix is to wrap the batteries with tape so they lightly jam in the battery holder to prevent the vibration. Or one can do the same by putting the tape in the battery holder (but different batteries may have different diameters so one might need to adjust the thickness of the tape).

(FWIW, lighter lithium cells might also minimize this problem.)

And similarly, a wrist mounted GPS probably won't do as well as a handheld for hiking...sensitivity is limited by antenna size and power. Size limits expandable memory for mapping and limits sensor options.

Compromises must be made if you want to minimize the size and weight...

As far as accuracy goes...everything depends on where the satellites are at any given moment as well as receiver design. Its not uncommon to lose all signals in a notch or ravine.

The accuracy depends on a number of factors, of which the satellite constellation is only one.
From http://edu-observatory.org/gps/gps_accuracy.html:

Table 2   Standard error model - L1 C/A (no SA)

                                One-sigma error, m
Error source      		Bias 	Random 	Total   DGPS
------------------------------------------------------------
Ephemeris data 			2.1 	0.0 	2.1	0.0
Satellite clock 		2.0 	0.7 	2.1     0.0
Ionosphere 			4.0 	0.5 	4.0     0.4
Troposphere 			0.5 	0.5 	0.7     0.2
Multipath 			1.0 	1.0 	1.4     1.4
Receiver measurement 		0.5 	0.2  	0.5     0.5
------------------------------------------------------------
User equivalent range 
  error (UERE), rms* 	 	5.1 	1.4 	5.3     1.6
Filtered UERE, rms 		5.1 	0.4  	5.1     1.5
------------------------------------------------------------

Vertical one-sigma errors--VDOP= 2.5           12.8     3.9
Horizontal one-sigma errors--HDOP= 2.0         10.2     3.1

Ephemeris data=error in the orbital parameters
Satellite clock=errors in the satellite clocks
Ionosphere=the ionosphere adds a variable delay
Troposphere=the speed of light is affected by atmospheric properties
Multipath=the signal can arrive by different paths with different delays
Receiver measurement=noise etc in the receiver
UERE=user equivalent range error (combination of the above errors)
VDOP=vertical dilution of precision, typ >2.5
HDOP=horizontal dilution of precision, typ 2-4, can be 10

EPE=estimated position error (~95% probability)
EPE=2*HDOP*UERE

The UERE is the total of the various errors affecting the measurement of the range to each satellite. The VDOP and HDOP are factors due to the geometry of the satellite constellation (and the only factors directly affected by the constellation).

The satellite positions indirectly affect the ionospheric, tropopheric, and multipath contributions to the UERE.

The above error analysis assumes a perfect skyview. Blocking the direct-path signals from any the the satellites can increase the VDOP, HDOP, multipath, and receiver measurement errors. (Remember, satellites below the horizon are always blocked--this is why HDOP is >2 and the VDOP is usually greater than the HDOP.)

For more detail, see http://edu-observatory.org/gps/gps_accuracy.html

A minimum of 4 satellites are required to get a position fix, however it is possible to use more which reduces the error. If you have 5 or more satellites, it is possible to check for the presence of a bad satellite and 6 or more will enable one to determine which satellite is bad.

The vast majority of the performance improvement seen in the newer high-sensitivity GPS receivers is due improved signal processing in the GPS chipsets. For instance, the SiRFstarIII chipset in the 60CSx has over 200K correlators. (It takes two to track a signal.) This allows very fast signal acquisition and reacquisition and potentially allows separate tracking of the individual multipath signals. (Some of the earliest consumer GPSes could only track 4 satellites and might take a half hour to get a fix under good conditions.) My tests have shown that not only is the 60CSx able to acquire a fix under degraded conditions, that fix is more accurate than one from an older unit (eg the 60CS) which might only find fewer satellites under the same conditions.

Doug

 

[Remix]

I tend to view the wrist GPSes as fancy pedometers... (I often throw a GPS in my pack while taking a walk, hike, or ski as an even fancier pedometer.)

The wrist units have some features such as 1) racing against a virtual partner, 2) detecting that you have completed a lap and showing/logging time, 3) pace or pedal cadence too slow or too fast, 4) all the heart rate stuff.

You really need the display on your wrist or handlebars and not in your pack to take advantage of these features. Again, you won't see all of these features on a handheld. Each has a different purpose.

 

[Kevin Rooney]

This is a bit tangential to the topic of the thread, so my apologies -

My current GPS is a 60CSx, and I use it constantly, typically twice a week. I bought it in 2007, and it does everything I want it to. Since Garmin has announced a revised model (the 62 series) the price has been dropping, and places like Cabela's have even run deals. A friend recently bought a new one for $200 which included a larger microSD chip and the 100K maps. I played with it for a few minutes and was impressed that it grapped satellite lock even faster than mine. Amazon has them for about $200.

In any case, I decided to keep my eye on prices, and if they dropped even more, to buy a replacement as a backup. A few days ago, I noticed that the 76CSx has dropped to about $230.

My question is - is the 76CSx essentially a 60CSx, and floats and is waterproof? Or is aimed at marine navigation as opposed to land? It weighs about 1oz more, but other specs seem the same. Also, the specs listed on the Amazon website says "water resistant to IEC 60529 IPX7 standards (can be submerged in one meter of water for 30 minutes)" means that for hiking purposes the 60CSx (at least the new ones) are waterproof enough for hiking purposes.

Any thoughts? Stick with the 60CSx?

 

[DougPaul]

This is a bit tangential to the topic of the thread, so my apologies -

You could have started a new thread or tacked it on to one of the choosing a GPS threads...

My current GPS is a 60CSx, and I use it constantly, typically twice a week. I bought it in 2007, and it does everything I want it to. Since Garmin has announced a revised model (the 62 series) the price has been dropping, and places like Cabela's have even run deals. A friend recently bought a new one for $200 which included a larger microSD chip and the 100K maps. I played with it for a few minutes and was impressed that it grapped satellite lock even faster than mine. Amazon has them for about $200.

In any case, I decided to keep my eye on prices, and if they dropped even more, to buy a replacement as a backup. A few days ago, I noticed that the 76CSx has dropped to about $230.

My question is - is the 76CSx essentially a 60CSx, and floats and is waterproof? Or is aimed at marine navigation as opposed to land? It weighs about 1oz more, but other specs seem the same. Also, the specs listed on the Amazon website says "water resistant to IEC 60529 IPX7 standards (can be submerged in one meter of water for 30 minutes)" means that for hiking purposes the 60CSx (at least the new ones) are waterproof enough for hiking purposes.

Any thoughts? Stick with the 60CSx?

The 76 series is the same as the 60 series except that it uses a different case. The 76 series case is larger which gives enough buoyancy for it to float. (I've seen reports that the 60 series floats with lithium batteries or the neoprene case.) The 76 series case is shaped for lying on a flat surface*, the 60 series case is shaped for sitting vertical in a pocket and has some rubber armor. The 76 series has buttons above the display, the 60 series has buttons below the display. IIRC, the 76 series comes with a larger memory card than the series 60, but it is still so small that IMO most users will need to replace it with a 2-8GB card.

* The antenna in the 60 series is oriented for vertical operation, I don't know the 76 series antenna orientation, but since the 60 series seems to work well in any orientation, I presume the 76 series will too.

I recently discovered some unadvertised upgrades (better battery life and use of lithium batteries) in the recent 60CSx. I don't know if the 76 series was similarly upgraded. http://www.vftt.org/forums/showthread.php?t=38355

IMO, the 60 series is a little better suited for hiking than the 76 series, but either will work. If my primary application was nautical navigation, then I'd probably favor the 76 series, but again, either would work.

IMO #2: While the 62 series has a few improvements over the 60 series, there are no show-stopping flaws in the 60 series (particularly the new 60CSx with the better battery lifetime and the ability to use lithiums).

Since I discovered the unannounced upgrades in my new 60CSx, it has become my primary outdoor unit. I haven't compared its TTFF (time to first fix) with the old one, but such comparisons are very difficult because it varies wildly (from ~1 sec to ~40 sec) based upon the information already stored from recent usage. The TTFF can also be increased by a degraded skyview and/or motion during satellite acquisition. (The 60CSx appears to be significantly more sensitive to motion during acquisition than does the 60CS.)

Doug

 

[Arthur Dent]

DougPaul-“IMO #2: While the 62 series has a few improvements over the 60 series, there are no show-stopping flaws in the 60 series (particularly the new 60CSx with the better battery lifetime and the ability to use lithiums).”

From having owned a 60CSx for years, and having worn out my first 60CSx, I agree that it is a great unit. However, there are features that the 62s has that, depending on your use, could make it a far better choice. The downside is the 2X price difference but it’s only money, right? ;-)

I have owned the 62s since Christmas when Santa brought me one. I use my GPS for geocaching and hiking as well as street routing. On the 62s I have the latest (2011.30) City Navigator NA loaded on the added 8Gb micro SD card and that map set takes up 1.4Gb of space. I have several different topo map sets loaded including the www.gpsfiledepot.com hi-res NH topos with 20’ contour lines. The 62s can support multiple map sets (.img files) as long as you give them different names and I have 8 loaded on my 8Gb card right now. I also just tried, and succeeded in creating a custom map (.kmz GoogleEarth overlay file, http://www.gpsfix.net/garmin-custom-maps-getting-started/ ) that has the trail system of an area I frequent. I set it up as an overlay that could exist on top of the NH topo map so I have the trails and well as the detailed resolution of the 20’ contours on the NH topo map. The limitation is that a custom map you make in this way is limited to about a 2mile X 2mile section and it will be pixilated when you zoom in past a certain point because the image you start with is in a JPEG format and as you blow up any photo it gets grainy. Where I have just the trails on the overlay and use the hi-res topo map for contour lines, it isn't that bad. Because the 62s is seen as a mass storage device by the PC it is easy to drag and drop or copy files using Windows Explorer the same as you’d do with any other storage media or you could use Garmin MapSource or BaseCamp.

The big advantage for serious geocachers is that the 62s works like a GPS/PDA combined and you have the cache description, hints, and recent logs to display on the 62s. The limit to the number of caches you can load is apparently 5000 so that shouldn’t be a problem. These caches can be transferred using GSAK or dragged or copied as well. They can be in multiple (.gpx) files as long as the names are unique. All files can be loaded to either the internal 1.7Gb memory or whatever micro SD card you add. Unless you need or choose to add huge map sets, the internal memory may be more than adequate for your needs.

Because everything is accessed a little differently on the 62s than the 60CSx, if you have owned the 60CSx it takes a while to “unlearn” those steps and learn the new ones. Also, there is a pretty steep learning curve and unless you use the unit a lot or are a whiz, it can be very frustrating to get it to do what you want it to. If you take the time and can grasp the many unique features and quirks of the unit, it is really great, and for me, a big improvement over the 60CSx.