Different distance with GPS/Ultratrack

It's look normally. I have tested running GPS vs Ultratrack. Test result is same as you.

So ultratrack is measuring short or gps long? I'm guessing ultratrack short.

Utratrack is short, and there are further problems with the speed coming in bursts, and average speed (as a result of short distance) being far too high. The track is shorter and different on the Ultratrack recording, but not enough to account for that much error.

It is my experience with Utratrack also, it simply doesn't work as advertised. You should see how poor it is on hikes. It looks like Harold and the Purple Crayon on crack. I just stick with GPS as it works.

It would be really nice to have Ultratrack actually function as advertised to get really long battery life out of long activities like a long 14 hour hike and still have enough left for the next day without a re-charge.

It's seems to be working perfectly as intended if I understand the issue being discussed. I get the feeling the intention of Ultratrac is misunderstood. It's a battery saving feature, not an extra high precision feature. Think ultra long battery, not ultra precision. Weird I know but that's what they named it.

Ultratrac saves battery because it's turning the GPS off, it's the only way that can work. Using the GPS uses battery at a much higher rate than just doing it's thing as a watch/HRM. If GPS stays on 100% you get much more accurate results, but you pay for that in battery consumption.

Ultratrac is for when you want to save battery but still want an idea of where you've been, so it automatically "wakes up" the GPS periodically to grab a position and then turns it off again. It's a compromise.

If you are on a short hike, there is no real need to save battery. An Instinct will run nearly a day in GPS on mode. Plus it's easy enough to carry along a charging cable and USB battery pack if you so desire. But, if you have a situation where you can't easily maintain charging the Instinct, say a multi day backpacking trip and you forgot your charging cable, well that feature would let you trade detail for duration.

Airmapper is absolutely correct. If you want a record of anything more detailed than what county you were in, then you should use either Smart or Every Second Data Recording - definitely NOT Ultratrac.

For example, here are two short bike rides I recorded at the same time - one with Every Second Data Recording on my tactix Bravo, the other with Ultratrac Data Recording on my fenix 3:

Every Second Data Recording
[IMG2=JSON]{"data-align":"none","data-size":"full","height":"786","width":"1000","src":"http:\/\/i219.photobucket.com\/albums\/cc18\/gaijinnv\/GPSComptactixBravo_11JUN16_zpsb3g109eo.jpg"}[/IMG2]

Ultratrac Data Recording
[IMG2=JSON]{"data-align":"none","data-size":"full","src":"http:\/\/i219.photobucket.com\/albums\/cc18\/gaijinnv\/GPSCompf3_11JUN16_zpsfdzrpcig.jpg"}[/IMG2]

Here are the two tracks overlayed with the Every Second track in BLUE and the Ultratrac track in RED:

[IMG2=JSON]{"data-align":"none","data-size":"full","src":"http:\/\/i219.photobucket.com\/albums\/cc18\/gaijinnv\/GPSCompGE_11JUN16_zpsrmtqtfai.jpg"}[/IMG2]

Obviously, it is not possible to record accurate Speed/Distance data using Ultratrac.

HTH

IMHO This isn't tracking at all. It is supposed to be using the sensors apart from the GPS for dead-reckoning, and it clearly is not doing that properly. It is simply connecting the dots between GPS fixes with straight lines, which is useless in all but moving in a straight line. No activity have I ever done except for driving in a car, is a straight line. If mileage is off by 50% or more, than it is simply not functional. It should not be called "tracking" but be called "auto-way point"

For the record, I have the Fenix 5 and it behaves the same way.

How it should be working is: Start be getting a good GPS fix, then shutting GPS off. Then using the non-GPS sensors to record: "I am measuring steps, and this many in this direction, estimate a fix here, this many steps in that direction, another estimated fix, get a new GPS fix after several minutes and shift.rotate all points since last GPS fix to match up." If biking, the GPS should attempt a fix at every detected change in direction, and work things a tad differently but basically the same to use GPS fixes to rotate the non GPS sensor data into position and back-calculate measured data to make it work. Lastly, if in Ultratrack, speed calculations should be smoothed and averaged between fixes, not big humps of brief speed.

I don't think Garmin has spent any effort on getting Ultratrack to be useful since the vast majority of users are using GPS. It is rare to even see someone chime in on Ultratrack.
The question is if it is worth trying to get working when battery life with GPS on is already acceptable? I tried Ultratrack twice, and moved on.

IMHO This isn't tracking at all. It is supposed to be using the sensors apart from the GPS for dead-reckoning, and it clearly is not doing that properly. It is simply connecting the dots between GPS fixes with straight lines, which is useless in all but moving in a straight line. No activity have I ever done except for driving in a car, is a straight line. If mileage is off by 50% or more, than it is simply not functional. It should not be called "tracking" but be called "auto-way point"

For the record, I have the Fenix 5 and it behaves the same way.

How it should be working is: Start be getting a good GPS fix, then shutting GPS off. Then using the non-GPS sensors to record: "I am measuring steps, and this many in this direction, estimate a fix here, this many steps in that direction, another estimated fix, get a new GPS fix after several minutes and shift.rotate all points since last GPS fix to match up." If biking, the GPS should attempt a fix at every detected change in direction, and work things a tad differently but basically the same to use GPS fixes to rotate the non GPS sensor data into position and back-calculate measured data to make it work. Lastly, if in Ultratrack, speed calculations should be smoothed and averaged between fixes, not big humps of brief speed.

I don't think Garmin has spent any effort on getting Ultratrack to be useful since the vast majority of users are using GPS. It is rare to even see someone chime in on Ultratrack.
The question is if it is worth trying to get working when battery life with GPS on is already acceptable? I tried Ultratrack twice, and moved on.

On the contrary, UltraTrac is performing exactly as designed. From the Instinct Manual:

UltraTrac
The UltraTrac feature is a GPS setting that records track points and
sensor data less frequently. Enabling the UltraTrac feature increases
battery life but decreases the quality of recorded activities.You should
use the UltraTrac feature for activities that demand longer battery life
and for which frequent sensor data updates are less important.

The Instinct does not have some kind of inertial guidance system, or "dead reckoning," that would allow it to extrapolate track/speed/distance info in the absence of a GPS signal.

Your dream of "How it should be working..." is a nice dream, but not based in fact.

HTH

Holding the handrails while walking or running on a treadmill may decrease the accuracy of the recorded speed and distance. I'm not sure if that is dream or fact based but seems to fall into the category of common sense.

What do folks make of the elevation differences in both of your tests? why do they provided different numbers there?