Fenix 6 Series - 17.72 Public Beta

It's easy to always complain. Now something positive. Looks like the track when running has massively improved with this firmware. If this remains as is it's goodbye to my bowl of spaghetti runs

Now:

Previously:

Well this is good but proves very little. You should compare two Fenixes, one with 17.72 the other with an older firmware, run few times with those and then swap firmwares (upload 17.72 to the one with the older one and revert to the older one on the one with 17.72) and run few more times.

Of course, but I don't have two. A cycle ride on Saturday with this firmware also looked very smooth. All earlier rides and runs were of the spaghetti bowl type. Thus this is new. Were these two outliers? We'll see.

Of course I know, don't worry.

I would just like to thank the anonymous who downvoted my reply, which is just the correct scientific method to assess the validity of a statement. Congratulations, that says a lot.

I'm still waiting for the next hike under dense tree cover, but my last runs with 17.72 and 17.73 seem to have improved a little the GPS track, at least in those segments along narrow road with building on both sides.

It may be just my impression but it seems satellite fix got better, too. I hiked yesterday in a narrow canyon and at a certain point I stopped for lunch, pausing the Fenix. When I resumed, I though it would take a long time for the watch to get a fix (if any), but as a matter of fact it took only one minute or so. 

Last time I hiked with 16.60, in similar conditions, when I resumed the watch got a fix 20 minutes later.

Note for self: never stop recording when you're in a canyon...

Don't worry about downvotes. Might just have been an unintentional mis-fingering with a touch screen. Anyway, long run tonight with stretches of poor and good signal. Lets see what i get out of it.

I just had a look at a short walk last night. This one looks a lot better as well. Of course it's still jiggly as the mode of movement is a lot slower than running and more datapoints are added over a shorter distance. But there's very little crossing road multiple times or cutting through buildings. I'm very carefully optimistic.

Look i am not the guy who downvoted, i actually never vote... However... Ok the scientific approach, it's just bit too strict! Even if with same device, the comparison between old and new firmware looks already good enough... Actually it is very significative, will tell people what to expect with the firmware change on their own device, before and after! That's it

Frankysour said:

Ok the scientific approach, it's just bit too strict!

Well, in my opinion you can't compare two runs taken in different days. Maybe one day you had 30 satellites in view, maybe the other day you have only 12. As for the switch between firmwares, it is just to rule out that one of the watches is inherently better than the other. Since I work with sensors, I know more than well than two sensors built in the same way, even in the same batch, can behave very differently when subject to the same environmental conditions.

Finally, as for the several runs, you need that to build up a little more statistics to try and establish a rule. Otherwise it's like you just observe a picture on the wall that falls down and conclude that all pictures fall down.