GLONASS

[Boyd]

I got a Garmin GLO last week and have been putting it through its paces. If you aren't familiar with the GLO, it's a very small GLONASS-enabled bluetooth GPS receiver that sends position data to applications running on a laptop, tablet or phone.



While this specific device may not have broad appeal, the GLONASS performance is remarkable. Here's a test I did this morning side by side with my 60csx and Montana sitting stationary at a known location. You can see that the GLO is a completely different kind of device. It appears to have created less points than the other devices, but in fact they're all there, just stacked on top of each other. I don't think I will buy another handheld unless it also has GLONASS. 

[popej]

What you observe here could be "static navigation", it is artificial fixing of position if measured speed has low value. Static navigation is mostly harmless for car navigation but can't be tolerated in outdoor GPS. That's why 60 and Montana show quit big drift of position.

Other thing is that for high sensitivity GPS static measurements don't correlate with movement tracking precision. You should take all this equipment for a walk and then compare registered tracks. This could revers your opinion

[Boyd]

That's a good point. I have looked at tracklogs while driving slowly (15 mph) and they look very good.  Have not tried walking, and really only got this unit for the car.  I have not seen much in the way of tests of the eTrex x0 units with GLONASS. Are they having problems?

Even if the GLO was "artificially fixing" the static position, it was a pretty good guess. 

[popej]

I have seen some tracklogs comparisons for new eTrex, but I don't remember where it was. There was no problems but if I remember correctly, there was no gain from Glonass either.

[Boyd]

You should take all this equipment for a walk and then compare registered tracks. This could revers your opinion

Took your suggestion and went for a walk with my equipment. It did not reverse my opinion. 

First I used Globalmapper to track my position along with the NJ 2007 Orthoimagery shown below. It tracked me perfectly. Then on the return trip, I used OziExplorer to track. Now Ozi does a poor job of displaying your position in real time, but it's the only software I have that can record the full 10 points per second that the GLO is capable of. I recorded for about 4 minutes and ended up with over 2400 track points.

This is about the cleanest track log I've ever seen. I walked around in a circle at the North end of the track. That is just a big sandy intersection and I made a rough circle; I did not re-trace my exact steps the second time around. I would say this track is extremely accurate, probably within a meter. No way that any other device I've used could create a log like this.

Look at the enlarged view of the circular track at the scale of 1:100. The orthoimagery is at the resolution of 1 foot per pixel, so those pixels you see in that image are 1 foot x 1 foot. Wow.

For a $90 GPS, it's pretty impressive in my book. Add some kind of a smartphone or small tablet and you have a really accurate tracking device (assuming there are apps available).

[popej]

This looks good, 10Hz is impressive. Can I really see your steps in this track?
Would be interesting to compare simultaneous tracks form other devices.

[Boyd]

Can I really see your steps in this track?

Haha, it almost looks like that. But I would say the points are probably accurate in the range of +/- one meter. Looking at the track in such extreme detail raises some questions about the accuracy of the orthoimagery. Here's the metadata:

https://njgin.state.nj.us/NJ_NJGINExplorer/full_metadata.jsp

Horizontal Positional Accuracy Report: Orthophotography has a +/- 4.0 ft. horizontal accuracy at 95% confidence level, National Standard for Spatial Data Accuracy (NSSDA), for a 1.0 foot Ground Resolution Distance (GRD).

So without an accurate survey to compare to the track, it may be hard to tell exactly what's going on when viewing it at a scale of 1:100. Nevertheless, I think it's pretty damn good! 

I will try to do some track comparisons with the Montana and 60csx soon. But I know from experience, they will look really crude compared to the GLO. I think I need a break from this right now. And I don't want to be considered guilty of what a friend of mine calls "measurebating". 

[Indrid Cold]

I will try to do some track comparisons with the Montana and 60csx soon. But I know from experience, they will look really crude compared to the GLO. I think I need a break from this right now. And I don't want to be considered guilty of what a friend of mine calls "measurebating". 

What are you using with the GLO, your laptop ? For some reason I'm seeing you trekking around with your PC running Global Mapper.

[Boyd]

I am using an HP Slate 500 which has a netbook class CPU with 2GB RAM, 64GB SSD, a 9" 1024x600 touchscreen and Windows 7 Professional. Got it about two years ago but have not used it much until recently when I built a better auto mount so I could use Mobile PC. Runs Globalmapper surprisingly well, I would have expected it to choke. The SSD probably helps. But it's awkward to run any "regular" Windows program with tiny menus, a stylus and virtual keyboard. 

For the test above, I just carried the Slate around and put the GLO in my shirt pocket.

[Boyd]

This morning's test was disappointing. Notice the gaps in recorded points - I suspect that OziExplorer was having trouble keeping up with the 10hz data stream since this was a longer hike than the previous test (total tracklog was over 12000 points). But it also could have been a problem with the GLO itself?

Will do another test and record the GLO with Globalmapper since that seemed to work well, although it maxes out at 1hz.

[Seldom]

Took my Etrex30 up to Observation Point, Zion NP, on Sunday.  This is one of my GPSr challenge hikes because of the severely restricted view of the sky.  The Etrex performed better than either my 62s or 60CSX.  The errors in the center of the image were caused by loss of signal, when I was below an overhang in a manmade cave.  Don't know if the GLONASS gets the credit, or a fortunate set of satellites.  Elevation change from bottom left to top right is about 1500 feet.  Total hike elevation change is 2100 feet.  Imagery is NAIP 1 meter.  Track is up and back.  Track recording interval set to Normal, recording method set to Auto. 

[Boyd]

Wow, that must have been a cool place to hike.

[Seldom]

Shorter, and waaay prettier than the Grand Canyon.

[Boyd]

This will be my final test with the GLO; I think I have a good grasp of its capabilities now. My initial excitement has been dampened a bit... looks like performance is very similar to the 60csx and Montana. I think Popej's summary of the new eTrex devices probably also applies to the GLO:

There was no problems but if I remember correctly, there was no gain from Glonass either.

The static tests I did earlier still impress me though, and I may take it out to my property survey monuments and let it run for awhile to see where it locates them (having previously done this with a  60csx and Oregon).

The GLO does a great job for my primary purpose - as a receiver for Garmin Mobile PC running on a Windows 7 tablet in the car. But I will continue to use my Montana on the trail. 

[Indrid Cold]

Which GLO package did you get, the standard one or the aviation package?