Auto-pause is rubbish

you need to understand how gps works.

It doesnt give you a x,y,z location. The location it gives you varies on every single reading and you get a location within a circle of error, typically a few meters radius.

So when stand still the gps locations bounce around this circle and not only does it increase you apparent distance it may also affect your cumulative ascent and descent.

This is called GPS nesting.

You can see it in your track logs on GE.

So you need auto pause or a manual pause.

I agree for slow hiking etc auto pause is pretty crap.

Also for cycling compared to my old 520+ edge bike computer. Both are just using gps on one bike, but the bike computer is SO MUCH FASTER with auto pause/auto start than my fenix. Fenix often misses stops on the traffic light, but the computer is really accurate and fast for auto pause. it does not even make a difference when i am having a speed sensor on my MTB. Fenix with speed sensor is way slower than the bike computer with gps. Something is different in the algorithm from the edge to the fenix

thats an interesting point.

A watch has a very constrained size so antenna and batteries are very small.

The edge has much more real estate to play with, so better longer antenna, much heftier batteries.

Maybe the Edge has a faster CPU than the watch ?.

Yes the algorithms could be different, but movement detection from a corodinate point is fairly generic so I would have thought they would share common generic code, but maybe not.

minty95 said:

perfect for faster sports, running or biking

It might be fine for open spaces but if you trail run or mountain bike, or exercise with lots of overhead cover it’s far from perfect. I’ve not had autopause enabled since the first days of my FR305. 

Autopause will rarely work exactly as you want it to every time. It's just the way it is. It is as good as it's ever going to get, it's not likely to ever improve - not because Garmin didn't implement it well, rather because that's just the nature of autopause on a wearable device.

It works fine on a bike computer, for example, because there is an absolute and clear threshold - Unpause if the wheel is spinning, pause if it isn't. The thresholds are not nearly as clear and absolute when it comes to a wearable GPS.

Besides the thresholds issue, it's also potentially going to unpause and pause when you don't want it to, for example if you stopped for lunch and then walk 20 feet away to pee behind a bush, do you want it to unpause and record your pee walk? Because it will.

My best advice is to get in the habit of manually pausing and unpausing; that's the only way to guarantee it works exactly the way you want it to 100% of the time.

thats exactly what I do.

C.sco said:

My best advice is to get in the habit of manually pausing and unpausing; that's the only way to guarantee it works exactly the way you want it to 100% of the time.

Exactly right.

For what I want to use it for, auto-pause should be much better. I don’t really understand how it can be so bad with the GPS and accelerometer working together. The GPS accuracy with Multi-Band GNSS enabled should be about 4m. I had auto-pause enabled for being stationary. On Sunday, I paused for lunch during my hike and resumed my walk again 5 minutes later. I walked a good 100m along the trail down a steep hill, looked down at my watch and saw that there was no breadcrumb trail. The GPS was still off. I was on a coastal cliff path so had a clear view of the sky and 1 second recording. Didn’t bother walking back up the hill just to capture lost data so I pressed start stop a few times to wake up the GPS.

Really poor. I’m a forgetful type of person so auto-pause should be useful for me. I would have left the GPS on but that screws up your average speed.

Cycling isn’t much better, it still takes about 5 seconds to realise that you have stopped moving when you stop at a set of traffic lights.

I was/am getting exactly the same, where mine would simply not bother restarting.