Hiking with 67i, 66sr & 66i

[GPSrChive]

The .gpx files are labled "original."

But are they?

The GPX files in the original post are exports from the host website, not from the devices themselves.

I would like to see the actual non edited untouched GPX files from each device.

[jlg2]

I appreciate your posting the tracks comparing those devices and the care you took to make the comparisons. If you have time, I'd love to also see track files of similar side-by-side comparisons recorded with devices in expedition mode.

[Matt16598]

As you have guessed, "Original" means unedited. I copied each track directly into BaseCamp, renamed it immediately (so there would be no question as to where it came from), and then exported it directly to GPX. As noted, I did edit the 66i tracks from previous months. Here are the original files. Unless CalTopo changed the data somehow (it shouldn't have, since I didn't edit anything there), these should be identical to what you already have.

Hike 1 Larch Mtn

Hike 1 Larch Mtn 6-May-23.zip

Hike 2 Eagle Creek

Hike 2 Eagle Creek 6-May-23.zip

Hike 3 Eagle Creek

Hike 3 Eagle Creek 20-May-23.zip

I too was surprised at how close the elevation values were (when checking them in the field). On the other hand, these all use barometric altimeters for tracking elevation, after the initial auto-calibration from GNSS elevation at the beginning of the track. Therefore they should be very close if the hardware is working correctly. I've previously noted virtually identical elevation gain figures from my friends' eTrex or Oregon units, which also have barometric altimeters. For example, on a hike from Shorty's Well to Telescope Peak, with around 11,800 ft. actual gain, the 66i and a different unit with altimeter (Oregon series?) both gave overall gain within 100 feet of each other (I think the trip meter showed 10,500 while the track was higher). For the gain to be almost identical, the elevation value at any given time must also be very consistent between units.

[Matt16598]

I agree that it would be interesting to see the results with a point recorded every second. It would give a better of idea of the raw accuracy of the unit. However, this isn't how I use my equipment in real life. I don't want to have 3600 points per hour to clean up afterwards. Some of my day hikes have been more than 20 hours long.
Also, Auto mode should still help us to see how consistent between passes the units are, and if there are any large deviations from the actual location.

[GPSrChive]

Unfortunately, the GPX files shared above have been edited and are not in fact original copies as taken from the device.

Added for clarity:

GPSMAP 67 Track, Not Original

GPSMAP 67 Track - NOT Original.png

GPSMAP 67 Track, Original

GPSMAP 67 Track - Original.png

[Przekątny]

There was a similar situation in another thread, no one paid attention to it.

Spoiler

https://www.gpsrchive.com/Discussion/vi ... 005#p11005

The track was from the "CompeGPS" software.

[MarkHL]

Unfortunately, the GPX files shared above have been edited and are not in fact original copies as taken from the device.

This is an interesting point... I often export GPX data from Garmin Connect or Basecamp since the activity is saved by default on the device in Garmin priority file format (*.fit). I'm going to enable the option to additionally save to a GPX format so I can work with the raw data from the device.

[jlg2]

Unfortunately, the GPX files shared above have been edited and are not in fact original copies as taken from the device.

Please expand on that? Why do you think the files were edited? Who or what edited them?

A later post by Matt16598 makes me think Matt16598 believes the editing you refer to is the file format conversion from .fit to .gpx. Is that what you meant? I don't think that would change anything pertinent to this thread. Please correct me if I'm wrong. I'd like to know. I believe both formats store precision beyond the capability of the devices in question. So there should be no difference in precision between formats. I think the only difference in the formats is that Garmin's .fit is a binary format that needs much less storage and cannot easily be read by humans. If Garmin is using an algorithm that does excessive rounding in .gpx that loses precision provided by the device, I'd like to know.

Matt16598 also thinks saving files in .gpx format in the device might make a difference. Again, please correct me if I'm wrong. I believe Basecamp converts .fit to .gpx when it imports a .fit file; and I always assumed it uses the same translation algorithm that the device uses. If true and if Matt16598 is always running files through Basecamp, it wouldn't matter how he stored the files in the device.

[Matt16598]

Yes, the original tracks were saved in FIT format. However, I did not manually edit anything, except for the titles and color. Any "editing" would be some kind of automatic process performed by BaseCamp in converting the tracks. I assume that the points and elevations remain very close to the originals, i.e., any differences would be due to rounding error from the differences in the formats. Does anyone know how FIT and GPX actually differ? Does it matter for the comparisons we're trying to make? (I'm certain that FIT can represent fairly precise locations, in any case, based on my experience.)
Garmin says that FIT is better for totals and GPX is better for precise locations. Maybe someone can shed some light on this.

GPX vs FIT

https://support.garmin.com/en-US/?faq=qn6PDmIjvs2TzPBDHULTD6

[Matt16598]

I agree that we are pretty far off topic, although it's been interesting.

I've opened a new thread to discuss the differences between GPX and FIT and what happens when and how GPX is exported from FIT. Perhaps we can get back to analyzing the differences between track logs from actual hikes with these three units.