Is bad accuracy of tracks normal?

Moderators: HAZ - Moderators, GPS - Admin

Linked Guides none
Linked Areas none
Post Reply
User avatar
paulshikleejr
Posts: 105
Joined: Feb 02 2010 4:37 pm
City, State: Scottsdale, AZ
Contact:

Is bad accuracy of tracks normal?

Post by paulshikleejr » Jun 05 2010 9:54 pm

Am I doing something wrong or not doing something? Is this as good as it gets? How do people make the really nice tracks I've seen here for other hikes?

I just bought a Garmin GPSmap 60CSx and have been using it to make tracks.

The first two tracks (http://hikearizona.com/location_g.php?GPS=9082, http://hikearizona.com/location_g.php?GPS=9087) I uploaded weren't all that accurate, but I thought it was because I was just learning how to use the thing (still learning, of course).

Then I did a few tracks this past THU (http://maps.google.com/maps/ms?hl=en&gl ... 1e8e34257d, http://maps.google.com/maps/ms?hl=en&gl ... 73f950ec80, http://maps.google.com/maps/ms?hl=en&gl ... aa2472bc80) and was appalled by the very bad accuracy: they were sufficiently bad that I didn't upload them here, only to Google Maps so others could see. (On Google Maps, they look pretty much the way they did when I started [but didn't finish] uploading them here.)

I e-mailed my tracks and the Google Maps URLs to Garmin and asked them if I bought a crappy model and what's going on with the "accuracy" (or lack thereof) of the tracks. They said "No the GPSMap 60CSx is a high end GPS. As I stated, this issue is caused by signal multipath error and not your GPS. When I compare the Google maps to the surrounding areas where the errors occur, I do see buildings and other obstructions that can cause this issue. For instance, the Scottdale Healthcare building is tall enough to block satellites on the horizon. Please remember that you are not only getting a signal directly above you, but you are also getting a GPS signal off the horizon. The other activities show you in a valley. These large mountains can cause this issue. This issue will occur no matter what GPS you have unless you purchase a very expensive survey grade GPS." Does this explanation hold water?

Before I read the e-mail containing the above explanation, I called Garmin and someone else said I should do the "hold down the 'zoom out' key while turning on the unit and keep the 'zoom out' key pressed until the satellite screen shows" reset procedure to refresh the satellite data in my unit ("because the data may have been corrupted by solar flares or something"), so I did that last night. This morning, I made this: http://maps.google.com/maps/ms?hl=en&gl ... e1f4d1160f. It's arguable more sucky than before (for both ways, I tread virtually the same ground [except for a few times when I moved to the right on the way back {since I was on the left to duplicate my track from start to end}]).

By the way, I'm laying down the dots every 5 seconds on the Conocido Park Perimeter and Camelback Walk tracks. For the Lost Dog, Old Jeep, Ringtail, Anazasi Loop (http://hikearizona.com/location_g.php?GPS=9082), I believe I was doing dots every 0.01 mile and for the Wagner Granite Bluff Loop HigherD (http://hikearizona.com/location_g.php?GPS=9087), I believe I was doing dots every 6 seconds.

Am I doing something wrong or not doing something? Is this as good as it gets? How do people make the really nice tracks I've seen here for other hikes?
Decisionmaking and action should only be informed by, not subject to, healthy, reasonable concern.
That being said, anything worth doing is worth doing to excess.

User avatar
joebartels
Posts: 7020
Joined: Nov 20 1996 12:00 pm
City, State: Phoenix, AZ

Re: Is bad accuracy of tracks normal?

Post by joebartels » Jun 05 2010 10:44 pm

The Camelback Walk one pretty much shows that when you get away from the buildings your accuracy increases from 80ft down to 5-15ft. You probably won't do much better than 5-10ft. Not sure if the 60CSx model has it but the eTrex gives you a constant updated "accuracy" measure.

Be sure it's clipped high on your shoulder strap or on top of your pack too. I just leave mine on normal tracking. Increasing frequency may increase AEG(may make it worse too...lol) but it will not increase accuracy of plot placement. You can see that the line didn't jump back and forth. Rather it was off by a margin. The margin slowly increased and decreased, hence more or less data would not move the line.

A couple days out of the year I get real wacky tracks but that doesn't seem to be the issue in this case.
Hike Arizona it ROCKS!

User avatar
Grasshopper
Posts: 1651
Joined: Dec 28 2006 5:06 pm
City, State: Scottsdale, AZ

Re: Is bad accuracy of tracks normal?

Post by Grasshopper » Jun 05 2010 10:53 pm

paulshikleejr wrote:Am I doing something wrong or not doing something? Is this as good as it gets? How do people make the really nice tracks I've seen here for other hikes?
Paul, a number of us here on HAZ are actively using a Garmin 60CSx and I think we are generally satisfied the way it is working for us. When you say you are having an "accuracy" problem, accuracy in regards to what areas of measurement are you finding incorrect compared to what baseline(s) you are comparing too?

Maybe if you could hike an HAZ listed hike description that someone has already posted a default route for, then you post your GPS Route to this same hike description, then we can open both yours and the default route and see what the differences look like on our TrailDEX maps. Also, I know you know that there are a ton of of 60CSx menu setting options to select from in the units main menu and other places in the unit. I assume you have become familiar with most all these various settings and have chosen the setting options that best fit your required needs? Many of these may not effect "accuracy", but some may if not set correctly. I'm certainly no expert on all this, but it has taken a long time for me to tweak all the menu options in a way so my unit seems to deliver accuracy ~correctly for my hiking tracks.. One also needs to learn how to edit their tracks to, for example, edit out(track erase) bogus track points that the unit sometimes puts down on a hike.
paulshikleejr wrote:Does this "Garmin" explanation hold water?
I can't comment on this one, maybe someone else can. I can only say that this 60CSx has a very good antenna and if you have your beep signal options turned on and set up correctly within your 60CSx, you should get a beep if you ever lose signal while hiking and if you loose signal it will tell you this on the screen and it will also cut you track at this point and start a new track, so when you get home and upload your hiking track, you may end up with more than one and if so, you need to determine if you want to keep the additional track portions, and if so, you need to learn how to do the edit track function to "join the track(s)".. enough detail. Hope others who have a 60CSx will comment too..
(Outside.. "there is No Place Like It!!")

User avatar
Grasshopper
Posts: 1651
Joined: Dec 28 2006 5:06 pm
City, State: Scottsdale, AZ

Re: Is bad accuracy of tracks normal?

Post by Grasshopper » Jun 05 2010 10:59 pm

joe bartels wrote:Not sure if the 60CSx model has it but the eTrex gives you a constant updated "accuracy" measure.
Yes, it is always located on the upper left of the satellite screen, measured in +/- feet of accuracy. The more satellites, the better the accuracy!
joe bartels wrote:I just leave mine on normal tracking.
I prefer "more often", but that don't necessarily make it better. gpsjoe uses a 60CSx and he prefers normal.
(Outside.. "there is No Place Like It!!")

User avatar
joebartels
Posts: 7020
Joined: Nov 20 1996 12:00 pm
City, State: Phoenix, AZ

Re: Is bad accuracy of tracks normal?

Post by joebartels » Jun 05 2010 11:08 pm

Grasshopper wrote:Maybe if you could hike an HAZ listed hike description that someone has already posted a default route for, then you post your GPS Route to this same hike description, then we can open both yours and the default route and see what the differences look like on our TrailDEX maps.
Paul has one you can compare, his is the black #1 line
http://hikearizona.com/location_g.php?L ... =9087,7475
looks pretty good to me
Hike Arizona it ROCKS!

User avatar
big_load
Posts: 4368
Joined: Oct 28 2003 11:20 am
City, State: Andover, NJ

Re: Is bad accuracy of tracks normal?

Post by big_load » Jun 05 2010 11:28 pm

I zoomed in on each of the posted tracks and consider them typical of what you should expect. If you find them appalling, you should have seen what unedited tracks from a older-generation GPS look like. I used to have to hand-edit a lot of wild points where the GPS was starting to lose or regain lock, which I doubt has occurred on any of your tracks. Even after hand editing, most tracks posted at HAZ (or anywhere else) are probably no more accurate than yours, but that won't be readily apparent because they aren't superimposed on an obvious "true" path.

Anyway, the explanation you received from Garmin is accurate. The zoom out thing on power up isn't going to help you, because you almost certainly have lock on plenty of satellites at all times and are receiving all the necessary data updates. If you want high accuracy for an individual point, you can use the averaging capability, but that won't work for a track, since it depends on you holding still during the averaging.
joe bartels wrote:Not sure if the 60CSx model has it but the eTrex gives you a constant updated "accuracy" measure.
Yes, it has an uncertainty circle, but you don't normally see it because you have to be zoomed in to a reference scale of 120 feet or so for it to come into play. (At least that's how big the uncertainty is in my living room, which is at bad as it gets, and where five years ago anyone would have called me a liar for claiming reception).

User avatar
Grasshopper
Posts: 1651
Joined: Dec 28 2006 5:06 pm
City, State: Scottsdale, AZ

Re: Is bad accuracy of tracks normal?

Post by Grasshopper » Jun 06 2010 8:33 am

joe bartels wrote:Paul has one you can compare, his is the black #1 line
http://hikearizona.com/location_g.php?L ... =9087,7475
looks pretty good to me
Looks good to me..close enough for government work! ;) . The difference in mileage 5.16 vrs Paul's 5.24 could be due to Paul's unit set to put down twice the # of route points than the other guys unit (which should still be ok). I prefer more points set down when the trail has a lot of twists and turns which, for me makes the route more accurate to follow if ever I need to download again and follow it.
(Outside.. "there is No Place Like It!!")

User avatar
big_load
Posts: 4368
Joined: Oct 28 2003 11:20 am
City, State: Andover, NJ

Re: Is bad accuracy of tracks normal?

Post by big_load » Jun 06 2010 2:04 pm

Grasshopper wrote:The difference in mileage 5.16 vrs Paul's 5.24 could be due to Paul's unit set to put down twice the # of route points than the other guys unit
Yes, that's as good as you can expect, especially with lower a lower sample rate. Rounding off corners on a circuitous urban running route once gave me a 6.1 measured vs. a 6.4 actual.

User avatar
nonot
Posts: 1925
Joined: Nov 18 2005 11:52 pm
City, State: Phoenix, AZ

Re: Is bad accuracy of tracks normal?

Post by nonot » Jun 06 2010 10:26 pm

For the most accurate tracks, have WAAS turned on and stay away from cliffs and canyon walls and tall buildings :)
http://hikearizona.com/garmin_maps.php

Hike Arizona it is full of sharp, pointy, ankle-twisting, HAZmaster crushing ROCKS!!
Hike Arizona it is full of sharp, pointy, shin-stabbing, skin-shredding plants!
Hike Arizona it is full of striking, biting, stabbing, venomous wildlife!

User avatar
Jeffshadows
Posts: 2590
Joined: Jan 30 2008 8:46 am
City, State: Old Pueblo

Re: Is bad accuracy of tracks normal?

Post by Jeffshadows » Jun 07 2010 8:04 am

nonot wrote:For the most accurate tracks, have WAAS turned on and stay away from cliffs and canyon walls and tall buildings :)
This, and the model of Garmin you buy still does make a difference. You needn't go out and buy a surveyor model but the higher-end ones tend to give better results. I took a $99 REI special out once along with my expensive model to compare - there was a big difference even on a jeep trail...
AD-AVGVSTA-PER-ANGVSTA

User avatar
Tough_Boots
Posts: 1756
Joined: Mar 28 2008 7:08 pm
City, State: Phoenix, AZ
Contact:

Re: Is bad accuracy of tracks normal?

Post by Tough_Boots » Jun 10 2010 12:38 pm

When inside any type of canyon area, I assume my track is going to be a mess. I figured this out after seeing differences in mileage in and out of trails. For example, at Barnhardt, when I do the switchbacks up the canyon wall the track is a total train wreck. I also remember being at the bottom of Reavis Falls and watching my little arrow move all over the place when I was standing still. I've learned in certain areas not to rely on my GPS but to use it as a more general guide. For example, down at Reavis Falls if you drop a waypoint to show yourself where to exit the creek back onto the trail, you're screwed. Your GPS, tracks, and waypoints do not give you an excuse to go into any situation with blinders on. I have one of the cheaper models (Garmin Etrex Venture HC) but I would go about it this way no matter how much I spent on one.
"there is no love where there is no bramble."
--bill callahan

User avatar
PLC92084
Posts: 723
Joined: Dec 22 2009 8:46 am
City, State: Vista, CA

Re: Is bad accuracy of tracks normal?

Post by PLC92084 » Jan 14 2013 9:58 am

Looking over the GPS-related threads, this looks like the appropriate place to post this inquiry...

Since acquiring my 60CX, I've noticed an annoying discrepancy between mileage shown on my unit's summary screen versus what is being reported in my Track properties. Most-recent example is the Aqua Tibia-Wildhorse Loop over New Years... Screen display recorded 21.8 miles (approx); Track Data showed 15.8 miles (approx) after downloading to mapping software. Even for a government worker, that's a BIG discrepancy! This didn't happen when using my old Etrex Legend. Doesn't seem to matter what application I load the track into (HAZ Route Mgr, Basecamp, Mapsource).

All firmware and software is updated; all settings (that I'm aware of...) have been set to reduce error (hiking vs ATV for example). Waas is On. The only thing I've noticed, that seems odd, is the low number of points being generated during a long hike (seems like there should be more). Looking at the track overlaid on the trail, there seems to be a fair amount of truncation going on (corners being cut, curves bypassed, etc.). The trail I'm using as an example had no tree coverage to speak of, no canyons or rock formations, nothing that traditionally influences accuracy/coverage in a bad way.

OK... climbing off my soapbox now...

User avatar
Tough_Boots
Posts: 1756
Joined: Mar 28 2008 7:08 pm
City, State: Phoenix, AZ
Contact:

Re: Is bad accuracy of tracks normal?

Post by Tough_Boots » Jan 14 2013 10:13 am

I have the Garmin 60csx and don't have that problem. Your summary screen's odometer is a different function then how mileage is figured out through your track data. If you're getting a low number of track points then that is for sure the problem-- the fewer track points the less accurate your mileage will be. If you're looking at the track you saved at the end of your hike-- your unit will always save to 500 points. If you pull the "active log" off of your unit when you get home from your computer, that should be a much larger track and much closer to your true mileage.
"there is no love where there is no bramble."
--bill callahan

User avatar
chumley
Posts: 7042
Joined: Sep 18 2002 8:59 am
City, State: Tempe, AZ

Re: Is bad accuracy of tracks normal?

Post by chumley » Jan 14 2013 10:26 am

I've always been frustrated with the way Garmin "saves" tracks. Not only does it truncate, but it also eliminates the time data. (Perhaps this has changed on newer models?) For me the key has always been to keep the Active Log going and use that for my uploads. I split it into separate hikes once it's on the computer, and don't delete the Active Log until I have the file uploaded.
Profound observer

User avatar
big_load
Posts: 4368
Joined: Oct 28 2003 11:20 am
City, State: Andover, NJ

Re: Is bad accuracy of tracks normal?

Post by big_load » Jan 14 2013 10:40 am

chumley wrote:I've always been frustrated with the way Garmin "saves" tracks. Not only does it truncate, but it also eliminates the time data. (Perhaps this has changed on newer models?) For me the key has always been to keep the Active Log going and use that for my uploads. I split it into separate hikes once it's on the computer, and don't delete the Active Log until I have the file uploaded.
Yes, I keep having to relearn that.

User avatar
tibber
Posts: 671
Joined: Feb 26 2004 1:27 pm
City, State: Phoenix, AZ

Re: Is bad accuracy of tracks normal?

Post by tibber » Jan 14 2013 10:50 am

chumley wrote:I've always been frustrated with the way Garmin "saves" tracks. Not only does it truncate, but it also eliminates the time data. (Perhaps this has changed on newer models?) For me the key has always been to keep the Active Log going and use that for my uploads. I split it into separate hikes once it's on the computer, and don't delete the Active Log until I have the file uploaded.
I agree. That's frustrating. I can't figure out how to keep all the tracks saved when I go on multi-day hikes in a row and can't upload to my computer. In 2011 I had all these tracks in Glacier Park and I saved them but then when I put them on the computer, all the time data was gone. So this year I didn't save the track thinking I would make one long track but apparently that's more info than my Garmin 60 can handle so it only saved 2 of the 5 tracks. The only thing I can think of is to change the SD card after two days hoping that the info will be on the card.
:?
For me, sometimes it's just as much about the journey as the destination.
Oh, and once in awhile, don't forget to look back at the trail you've traveled.

User avatar
PLC92084
Posts: 723
Joined: Dec 22 2009 8:46 am
City, State: Vista, CA

Re: Is bad accuracy of tracks normal?

Post by PLC92084 » Jan 14 2013 10:54 am

Could be a learning curve; just doesn't seem intuitive (like renaming Tracks with more than 500 points to Active Log) - Who thinks of that stuff when designing the hardware/firmware!?

I usually clear all track data before starting a hike so everything is, presumably, starting at zero. This doesn't include the Track (Active Log) I plan to use for that day... Chumley - if you're retaining the ActiveLog file, do you clear the accumulated mileage on the summary screen so you know how far you've gone? Or do you mentally subtract mileage you've done previously and just "guesstimate" mileage until you get back to your computer? If the latter, that method doesn't seem "optimal"...

User avatar
chumley
Posts: 7042
Joined: Sep 18 2002 8:59 am
City, State: Tempe, AZ

Re: Is bad accuracy of tracks normal?

Post by chumley » Jan 14 2013 11:36 am

I clear all the data ... except the Active Log at the start of every hike. I have also gotten in the habit of taking a photo of the GPS screen at the start, finish, and any notable waypoints along my trip. So even if I lose the track somehow, I will still have the relevant distance, time, speed, etc. to reference later if I need to.

But with the pathetic amount of space that a track takes up (compared to maps), I still don't understand why Garmin doesn't at least give you the option of saving a track with as many points as you want, and keeping the time data intact. I can't imagine it would be that difficult to program that way. ](*,)
Profound observer

User avatar
paulshikleejr
Posts: 105
Joined: Feb 02 2010 4:37 pm
City, State: Scottsdale, AZ
Contact:

Re: Is bad accuracy of tracks normal?

Post by paulshikleejr » Jan 14 2013 12:16 pm

Actually, it's only recently that I've been told that I'm supposed to "Save" tracks.

I've had my 60CSx for more than two years now and, up until a week or two ago, I only worked with "ACTIVE LOG"s--I didn't know what the "Save" thing was (OK, I admit that I didn't read the manual).

When I thought my unit was glitching recently, I called Garmin and they said, "Breaks in the ACTIVE LOG are a normal thing [actually, I'd never seen breaks in the ACTIVE LOG before now after over two years of use and that's why I called]. What you're supposed to do is 'Save' tracks when you're done hiking."

I tried that [actually, only last night] and was dismayed to find that not only did it concatenate all active logs into one track, but it looked like it truncated the point count to 500 for all! Luckily, I'd saved the original data so I wasn't SOL. Also, that track now occupies one of the few slots available for tracks.

My recommendation is don't "Save," work with active logs.

Also, if you put a 60CSx in "flash drive" mode (I'm not sure what this mode is actually called in the manual, but when you're in this mode, you can see the installed flash card as a removable drive) or if you read the removed flash card as a removable drive using your Windows PC, there is a folder that has *.gpx files by day. If you haven't downloaded from that folder before (especially if you're getting "card full" messages), then you have in that folder *.gpx file for everyday you've turned on your 60CSx. By the way, if you "clear" track data from the tracks screen, you DON'T clear the daily *.gpx files.

Generally speaking, I download and clear all active logs after one to three days of hiking (depending on how full the unit says the card is) and I download the daily files every three or four months.
Decisionmaking and action should only be informed by, not subject to, healthy, reasonable concern.
That being said, anything worth doing is worth doing to excess.

Post Reply

Return to “GPS & Electronic”


cron