This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion
pgreyover 4 years ago
I'm a bit new to the Garmin family, even though I had a VA3, for about 18 months, it never had any sort of "battery saving GPS mode", such as UltraTrac.
We do a yearly "mini tour", for the 4th of July, sort of just a quick loaded (touring bike) ride, it's only 20-25mi/way, but my bike has my "regular touring load", so it's pretty heavy, carrying almost all the "group gear" for the family (maybe 125lb or so, including the bike itself).
Anyway, part of my purchasing decision for the 945, was to "retire" my Suunto Ambit3, and still be able to use a "low-battery-drain" mode, such as UltraTrac (on the Suunto it's just a low-sampling-rate setting).
I figured with 60 hours, I could use it daily, when touring (typically not having access to a charger, sometimes for days in a row), much like I used my old Ambit, I could easily ride a week with it, with very good accuracy, still.
However, when I did my first two "test rides" this past week, with UltraTrac on, my actual measurements were basically "useless", it seems to have maybe a 30-35% "accuracy", and otherwise it simply thinks I'm not moving, at all, for very large parts of the ride (the link shows 7.43mi, but in reality, it's about 22-ish): connect.garmin.com/.../3816075064
Is this a bug, in the current (newish) firmware?
All I did, to create a "Low_GPS cycling mode", was to simply copy my existing cycling activity, and change the GPS to "UltraTrac".
The "base" cycling activity is very accurate, checked against my Ambit3, and my cyclometer, it's withing maybe 0.1 miles, say for a 30-40mi ride, I've recorded a few with the 945, in my first 10 days or so.
It looks like the mode/algorithm basically thinks that "because it's getting infrequent GPS updates, on UltraTrac", that I'm "not moving", when it's not taking a GPS reading(s), in this mode, which seems like a "bad assumption"?
Doesn't it seem like it should extrapolate here, so that I'm say 10% off, not 70%? I could live with the former, that's fine, but 70% is kind of ridiculous, not even worth recording, for tracking anything other than where I've been (no real point in the training data, at ALL, given that it's SO far off).
Did I not change a setting, or something, in the process of creating the activity with UltraTrac, are there other settings that it's dependent on?
I'm open to ideas, hopefully to make the watch usable, in this mode.
I really don't want to return it, and get a Suunto 9, instead, the HR accuracy is pretty darn good, for wrist, which otherwise makes it very handy, for this use/mode. But, if I'm going to get 25-35% of my loaded-touring rides recorded, in this mode, there just doesn't seem to be much of a point (even having/using the mode).
Thanks, for any tips/tricks/suggestions/whatever, here.
Even if this is a "known" issue with the "earlier-on" firmware in the 945 (2.50), perhaps?
- Cancel
Top Replies
- Thanks, but not really related to this question/topic right?If I see this correctly you just put your watch down in ultratac mode with a running activity and let it deplete?Interesting battery wise, but…
FlipStoneover 4 years agoin reply to Jan502+2
- UltraTrac accumulates distance by toggling between GPS and WDR (accelerometer) as it progresses through the satellite active cycle. While GPS is 'on' it will use that as the source for calculating distance…
deeter26_2over 4 years agoin reply to pgrey+2
- For curiosity sake - it would be interesting to see what a service like My GPS Files would come up with for distance if you dropped one of your test rides out there. My guess is that it would probably…
deeter26_2over 4 years agoin reply to pgrey+1
All Replies
0FlipStoneover 4 years ago
That's weird, the gps track is very decent but indeed... it seems to ignore the 'parts in between'. Shouldn't do that afaik (could it be that ultratac is more geared towards running and using sensors in a different way?)
I'll try (if I remember ;) ) an ultratac ride home from work... is only 3k or so but should at least give an idea if this is 'normal behavior'
*Edit*
Forgot to change mode from GPS to ultratac... will try agian later on
- Cancel
- Up0Down
- Cancel
0pgreyover 4 years agoin reply to FlipStone
Yep, @flipstone, that's exactly what it looks like, it sort of "marked" the location, for few seconds, and then turned the GPS off, and assumed I was just "sitting there", I guess.
Except that it did a decent job of actually tracking my route, as you mention, that's the weird part, and I'm sure my fitness metrics are "a mess", with regards to this, given that it sees me working like crazy (the loaded touring bike does that ;-]), and maybe averaging 4-5mph.
Not really expecting the fitness metrics to be accurate, anyway, I'd need to change my "physiology", or have multiple profiles for that, or something, to account for the extra 80-100lb or so (of panniers and gear), but having me "eeking" along at a fast walk, seems a bit slow ;-]
I think there's a bad bug in the algorithm, for Ultratrac GPS monitoring.
I forgot to mention, too, that it drains the battery much faster than just using "GPS Only", which sort of negates the whole purpose, anyway, until they fix this.
Is there a way to "file a bug/issues", that's official?
I'm curious to see, when you do this, if you get the same results as I did, thanks!
I created a "copied" activity (per my above), for touring ("Bike_LowGPS"), so I'd be more apt to "select" this one (with the only change being "Ultratrac GPS"), vs changing the existing "Bike Settings"...
- Cancel
- Up0Down
- Cancel
0FlipStoneover 4 years agoin reply to pgrey
Just gave it a try with what I know was a ~11k route and it tracked about 4.4 ;)
https://connect.garmin.com/modern/activity/3825537968
So indeed the same, I guess it's more for running/walking (more sensors to use) but it should at least be able to calculate the differences between the measured gps points right.
The tracking is very decent but indeed, it just ignores the spaces in between. Not that I ever used ultratac, but this would be a reason to not use it with biking at least.
Battery usage indeed wasn't as low as I'd expect, but usually the longer the activity the more precise the battery usage is so in the end it might be a bit better and it IS a bit lower then gps only.
- Cancel
- Up0Down
- Cancel
0Jan502over 4 years ago
I did test 56hours max from 100% to 0%
https://connect.garmin.com/modern/activity/3823292823
regarding configuration:
https://www8.garmin.com/manuals/webhelp/forerunner945/PL-PL/GUID-1500E73F-F386-49AF-A542-25D4B1655A08.html
- Cancel
- Up-1Down
- Cancel
0Jan502over 4 years agoin reply to Jan502
- Cancel
- Up0Down
- Cancel
0FlipStoneover 4 years agoin reply to Jan502
Thanks, but not really related to this question/topic right?If I see this correctly you just put your watch down in ultratac mode with a running activity and let it deplete?
Interesting battery wise, but not really answering the question here ;)
- Cancel
- Up+2Down
- Cancel
0isaziover 4 years agoin reply to pgrey
Looks similar to what is happening in "open water" mode. Have you tried the new GPS beta? If (as it seems) the track is good but the metrics are not, it is probably a software bug that could be fixed by Garmin. Give a try to the beta firmware and get in touch with the engineers.
- Cancel
- Up0Down
- Cancel
0FlipStoneover 4 years agoin reply to isazi
This is with the gps beta... But I now I happened to see this post:
Which makes it kinda, meh... As it seems the only thing one kinda could use ultratac for is hiking.
I might give it a try for a run soon, just to see how it behaves then, but doesn't seem to promising.
- Cancel
- Up0Down
- Cancel
0pgreyover 4 years agoin reply to FlipStone
Yeah, this was my thinking, as well, why is it just "skipping", the distances, between the coordinates, is that even useful?
I bought the watch under this premise, using it part of the year for loaded-touring, but w/o the UltraTrac, it's pretty useless for this mode (I'd have to carry a charger and charge it every day or two, whereas my old Ambit could easily go a week, or more, and was really pretty-darned-accurate, in low-GPS mode).It seems to me like it's MUCH worse than the 935 though, where the "accuracy" seems to be around 70-80%, the 945 is maybe 20-30% (accuracy), not even sure how the released it, like this, someone clearly ignored a test-result, if you ask me...
I would assume this (bug) is (hopefully) a separate issue, the fact it seems to just "drop off", and think that no movement happened, in-between the GPS tracks; even for hiking I don't think this would be very useful...
If this is "devolving", into a "hike only" type of mode, it seems like they should come clean, and just say that, so people can shop by feature, accordingly.
I had seen that same link, looking around for this a couple of days ago, but I also found this, where it was pretty darn accurate (in terms of distance and speed):https://www.strava.com/activities/601063088/laps
It sure looks to me like someone tried to fix something, with UltraTrac, and broke something else, big-time, in the process...
I'm fine with going on the beta GPS s/w, to test, but does it revert, after you move to the latest firmware revision? I'll do some digging and possibly test this, I'm going out for a "quick 25-30" this afternoon, unless it rains ;-]
- Cancel
- Up0Down
- Cancel
0pgreyover 4 years agoin reply to Jan502
Yeah, I don't think this really applies, the watch is "going nowhere", so it's tracking a "zero averaged" location, and even then it has a "strange track", if you ask me, why is it off by several blocks, in numerous instances?
This sure looks like a bug. If it were "remotely accurate", it should record the same spot, within a pretty small margin, for each GPS-point?
This *could* be the same issue, it's sort of tough to say, without looking at some traces, and understanding the mechanics of the algorithm here, IME...I'm starting to get the feeling that Garmin has a very inconsistent quality s/w release cycle, to put it nicely. I had basically zero issues with my Ambit2 and 3, even on low-GPS and "fast activities" like Alpine-Skiing (I'm an ex-racer, and chasing my daughter who just quit racing U14's last year, so my ski speeds are "quick"), it did a pretty decent job of my distances, and even my speeds, in most cases.
This references the *exact* issue at-hand, most likely, or very close (speaking more to the "general" issue, but I think this specific swim-tracking likely applies to the GPS issue) , if not exact:https://www.dcrainmaker.com/2019/06/competitor-software-instability.html
I'd be okay, if they said there were some "problems being worked out, with GPS-UltraTrac", or they noted that it was now relegated to "very slow activities only - such as trekking", BEFORE they marketed it, implying more or less the opposite (better battery life, a bit worse GPS tracking/accuracy).
- Cancel
- Up0Down
- Cancel
