GPS Accuracy

alciston romaine

So this morning I walked the dog (least ideal situation for the F5, walking and lot of standing still). Wore the F5 on the left wrist, 735xt on the right. Waited for 'press start' button on the F5 to start which is already a bit beyond the green GPS signal.

Actual distance is I think little over 1.25k. Both devices set to GPS+GLONASS and 1sec recording... No high buildings, pretty open.

http://www.mygpsfiles.com/app/#ueuCXXUH

So yeah, looks pretty awful for the F5 and although the 735xt also struggles here it atleast manages to keep me on the path and out of the water.

Elevation is way off for the F5 as well, actual elevation is 1-5m around here. So will re-calibrate that before the run (again, sucks if something isn't working and you have to manually tune the sensor each time you plan on using it).

Gonna do a small run in a bit so we'll see how it does with running then :/

ps. Noticed a huge difference in distance reported in mygpsfiles and gc so here the gc links as well:

Fenix F5:
https://connect.garmin.com/modern/activity/1649737147

735XT:
https://connect.garmin.com/modern/activity/1649735164

out of interest I tried my 735XT with 1 sec recording and turned smart off, the track is loads worse IMHO, I've had smart on since I had it and it's been much better

Wed run with 1sec recording - https://connect.garmin.com/modern/activity/1645041782

and a city centr run with smart recording from end of last year https://connect.garmin.com/modern/activity/1436422259

Elevation is way off for the F5 as well, actual elevation is 1-5m around here. So will re-calibrate that before the run (again, sucks if something isn't working and you have to manually tune the sensor each time you plan on using it).

Sounds like the F5 didn't get a good lock before starting; if the GPS reception was poor at the start of the activity, then the rest of the track is likely to be pants too. Assuming you have the Altimeter set to Auto Cal.>On then a good indication of whether you had a half decent GPS lock at the start of the activity is if the start Elevation was correct (basically the watch calibrates to GPS elevation when you start the activity if Auto Cal is On). The rest of the activity is then relative altitude to that start point using the barometer (unlike the 735 that only uses GPS elevation and is corrected to a topo map when uploaded to GC). As getting a good GPS lock before starting seems directly linked to better tracks (from viewing hundreds of posts on the subject on the F2, F3 and now F5 forums over the past 3 years) I've found it worth adding a datafield with elevation on the screen whilst waiting to start. Once the elevation number has stabilised (no drift or big fluctuations up/down) that's a pretty good indication of a reasonable GPS lock and you're ready to start.

BLEAK Detox

Did a little 10k run last night....

[URL="https://www.strava.com/activities/921162917?utm_content=7287560&utm_medium=referral&utm_source=twitter #strava"]www.strava.com/.../921162917 #strava[/URL]

Pace was all over the place, from 7min to 12min, even though I know I was doing around 8:16min average. The tracks are far from perfect and although not terrible, they do wander a lot and go a little crazy at times.

Also, I noticed that my wife Forerunner 235 had an the elevation at: 81ft , where my Fenix 5 had it at: 164ft!! ....exact same run, she was beside me!! ...bit of a difference there.

I had 1 sec recording on and GPS+GONASS.

I think Im going to set it back to Smart recording as it didn't really seem to help much. Plus Im going to turn off GONASS just to see what the tracks are like too.

Im hoping things improve a little otherwise I feel like I have gone back and now have a more inferior watch to my old Forerunner 630. Shame, because I love my Fenix 5 otherwise.

Did a little 10k run last night....


I think Im going to set it back to Smart recording as it didn't really seem to help much. Plus Im going to turn off GONASS just to see what the tracks are like too.

I came to same conclusion my F5 provides best results with GPS only and SMART recording. With 1 sec instant pace is just mess with smart recording it seemmed to be fine for me. Also GPS maps looks much more precise with smart recording

Sounds like the F5 didn't get a good lock before starting; if the GPS reception was poor at the start of the activity, then the rest of the track is likely to be pants too. Assuming you have the Altimeter set to Auto Cal.>On then a good indication of whether you had a half decent GPS lock at the start of the activity is if the start Elevation was correct (basically the watch calibrates to GPS elevation when you start the activity if Auto Cal is On). The rest of the activity is then relative altitude to that start point using the barometer (unlike the 735 that only uses GPS elevation and is corrected to a topo map when uploaded to GC). As getting a good GPS lock before starting seems directly linked to better tracks (from viewing hundreds of posts on the subject on the F2, F3 and now F5 forums over the past 3 years) I've found it worth adding a datafield with elevation on the screen whilst waiting to start. Once the elevation number has stabilised (no drift or big fluctuations up/down) that's a pretty good indication of a reasonable GPS lock and you're ready to start.

I tried the trick with vertical pace before and that sets to 0 pretty fast. Now I also waited a few minutes (after turning green) before going. Really don't want to go back to the days where I have to put my watch in the window 15 minutes before going out to get a decent lock (and I feel that's not really the case here anyway). Hell, in that case I'd prefer to tbe able to disable the barometer or whatever is used completely, if it needs a gps-lock to get set properly anyway.

Anyway, now also ran a bit. 2 times ~2.5k at a fairly steady pace... Again gps+glonass on both, 1sec rec

http://www.mygpsfiles.com/app/#uev6Xc4i

GC tracks:

735 towards:
https://connect.garmin.com/modern/activity/1649839811

F5 towards:
https://connect.garmin.com/modern/activity/1649841530

735 back:
https://connect.garmin.com/modern/activity/1649839824

F5 back:
https://connect.garmin.com/modern/activity/1649841549

Again pace all over the place for the F5, not sure about the gps tracks though... Indeed increased pace at least helps.

ps. Will try gps only (and maybe smart) when I go walk the dog in a bit to see what that does ;)

Agreed on not standing for 5 minutes to get a good lock, it's not 2003/FR201 time anymore, the game changed with the FR205/SiRFStarIII in 2005 ;-)

Anyway these tracks look pretty good to me not knowing where you ran exactly. Did you have autolap activated ? That's how I noticed that all was not well with the F5 compared to my FR235 and started to look at tracks more closely to see what could be going on. It turns out the distance measurements are inaccurate (as measured on Google Earth), sometimes too short, sometimes too long. See my previous posts for the details.

Oh and I saw some comments about "smart recording" being "better", it's not, it has zero impact on internal calculations (distance, pace, etc...) it just "smooths" the tracks, so yes they might look "better", but it's a bad idea if you're comparing tracks to see what's going on.

Yeah, have to analyze the tracks better myself to see which one tracks better... Autolap was indeed on.

And I did wait about 3 minutes at least before setting off after acquiring the green 'ok-go' circle.

THis was running through the city as well, which means adjusting more to traffic and bikes so harder to check the exact route. Will take another peek later today.

Also, I can run a fairly straight road for a few k as well from my house, will give that a test as well later (as in, later next week... Not running anymore till the HM this sunday and after that I'll take some days off as well ;) ).

Still not sure if I'm gonna take the F5 or the 735xt (or both, but don't want to be annoyed by 2 watches during this... aiming for a new personal best) for the HM.

Good luck with the race! Looks like you're looking at a pretty fast time based on your pace and TE ;-)

Did you notice the autolap coming in differently/inconsistently between the 735 and the F5 ?