Pausing directly after start of a run

like i said, it wasn't the case (ok, once in a while, and especially when joining a race, grrr) on my 620.

when i'm in a race, starting from row 1, i start with speed 0, and press start.
that would mean i cannot use autopause anymore i guess.
or make sure to disable it when doing a race.

it would be better if it couldn't autopause directly after the start, but i don't know if thats possible programming the software.

Why not just let it auto-pause upon start, when workout Timer excludes pauses anyway?

when i'm in a race, starting from row 1, i start with speed 0, and press start.
that would mean i cannot use autopause anymore i guess.
or make sure to disable it when doing a race.

Why? Since you know that is how the watch behaves, why not press to start the Run on the watch while you're waiting for the race to begin, knowing that the tracking will auto-pause immediately? Pauses are excluded from the Timer metric and calculations of averages for a workout anyway. When the watch detects that you've started running, it will auto-resume.

Now, if you disagree with the watch's criterion or sensitivity for auto-resume, then that's a different thing altogether. Auto-anything means you surrender a degree of direct control, and if you're worried that the watch does not pause and/or resume in exactly the conditions you would want it to pause and/or resume, then yes, you ought not be using auto-pause.

If you're not the programmer, then as the user you have to work around the existing programming.

like i said, it wasn't the case (ok, once in a while, and especially when joining a race, grrr) on my 620.

when i'm in a race, starting from row 1, i start with speed 0, and press start.
that would mean i cannot use autopause anymore i guess.
or make sure to disable it when doing a race.

it would be better if it couldn't autopause directly after the start, but i don't know if thats possible programming the software.

Well, just to be sure you can just disable autopause for the race right?

You shouldn't need it anyway, and if you do the autopause at the start doesn't really matter either.

Auto Pause definitely seems broken to me....

Today I thought I would give auto-pause a go during one of my regular walks into town at lunch time. My previous limited experience was yes it would pause straight away and then start again. Today I was already walking when the GPS kicked in and I pressed the run button.
Auto pause worked Ok when I stopped but the timer seemed to take a while to get going again. Once it seemed to get stuck and I manually paused the activity and restarted to get it going again.
Then I went under an underpass and lost GPS. This caused the watch to pause. This also happened again when in a built up part of town, whereas previously I hadn't noticed any issues. Note, I was walking when this happened. GPS loss seems to trigger a pause.
Also the restarting of the timer is delayed by up to 30s when the GPS is detected again making the actual activity time incorrect which kind of defeats the object really. While the watch is looking for the GPS signal the pace is being recorded but no timer or distance, even though it knows you are moving.It also recorded less distance than usual due to being paused when I was actually walking.
Finally when you look at the All Day HR trace the activity normally overlays the period of increased HR activity. In this case the activity time was used as opposed to the elapsed time so it only overlaid part of the area of increased HR activity.

Behaviour without auto-pause being active is that if you lose the GPS then the timer keeps going and when the GPS kicks back in the distance is added on. E.g. if I had recorded an elapsed 1km and I walked through a 0.2 km underpass, when I get to the outside again and GPS is detected the watch will re-establish its position and add on the 0.2 km and show 1.2km. This to me is the correct behaviour.

So in my experience this seems to be a bug to me. As soon as movement is detected the timer should restart and the correct distance should be recorded. Instead it waits for a GPS signal and loses recorded distance. Seems like another feature that doesn't work too well. :-(

Note I was using the "other" activity and GPS+GLONASS was on.

Would be interested if anyone else has experience this.

Behaviour without auto-pause being active is that if you lose the GPS then the timer keeps going and when the GPS kicks back in the distance is added on. E.g. if I had recorded an elapsed 1km and I walked through a 0.2 km underpass, when I get to the outside again and GPS is detected the watch will re-establish its position and add on the 0.2 km and show 1.2km. This to me is the correct behaviour.

That is not what I observed. It appears that when the FR235 loses GPS signal, it will assume the user is continuing at the last known pace – with minor adjustments based on cadence from the accelerometer – until it is able to reacquire a GPS signal. That might not be so bad when you’re actually running at relatively stable pace, but in my ‘experiment’, every time I was standing in a queue (when cadence is effectively nil) waiting to be served, the watch assumed I was still walking at the last known pace. I spent about 20 minutes in an underground supermarket, and the watch kept telling me my pace was 11’48”/km, and tallied over 1km of distance travelled while I was queuing at the checkout.

That is not what I observed. It appears that when the FR235 loses GPS signal, it will assume the user is continuing at the last known pace – with minor adjustments based on cadence from the accelerometer – until it is able to reacquire a GPS signal. That might not be so bad when you’re actually running at relatively stable pace, but in my ‘experiment’, every time I was standing in a queue (when cadence is effectively nil) waiting to be served, the watch assumed I was still walking at the last known pace. I spent about 20 minutes in an underground supermarket, and the watch kept telling me my pace was 11’48”/km, and tallied over 1km of distance travelled while I was queuing at the checkout.

I presume in the above scenario Auto-pause was off?

So are we saying that if auto-pause is off, in a scenario where you lose the GPS and are relatively still, the watch will over-estimate distance and when auto-pause is on it will under-estimate, as it has to wait for a GPS lock?
I will have to check out your scenario as my assumption was that when I has auto-pause off and I lost GPS, it was adding on the distance accurately when the lock was regained as it was effectively drawing a line between the two points. Now you have got me thinking about whether it is just using an average pace as you state.

Why? Since you know that is how the watch behaves, why not press to start the Run on the watch while you're waiting for the race to begin, knowing that the tracking will auto-pause immediately? Pauses are excluded from the Timer metric and calculations of averages for a workout anyway. When the watch detects that you've started running, it will auto-resume.

Now, if you disagree with the watch's criterion or sensitivity for auto-resume, then that's a different thing altogether. Auto-anything means you surrender a degree of direct control, and if you're worried that the watch does not pause and/or resume in exactly the conditions you would want it to pause and/or resume, then yes, you ought not be using auto-pause.

If you're not the programmer, then as the user you have to work around the existing programming.

OK, I tried the start before running technique a few times, but i'm not happy how it behaves.
it pauses directly, thats good, but when i start moving it takes about 5 (wild guess) seconds before it starts recording my run.

i think for me it's better to switch autopause off and during my runs where i (have to) stop from time to time, i have to be aware to pause manually.

i'm glad autopause is working fine (almost instantly) on my edge 520 with speedsensor. but on the forerunner 235 it's not doing what i expect.

OK, I tried the start before running technique a few times, but i'm not happy how it behaves.
it pauses directly, thats good, but when i start moving it takes about 5 (wild guess) seconds before it starts recording my run.

i think for me it's better to switch autopause off and during my runs where i (have to) stop from time to time, i have to be aware to pause manually.

i'm glad autopause is working fine (almost instantly) on my edge 520 with speedsensor. but on the forerunner 235 it's not doing what i expect.

Was it waiting for a GPS lock after pausing?

Must say, I don't get Auto-pause. When I'm out on a training run and I'n crossing the road I don't worry about 10 secs delay. It's a training run and I'm not going flat out so why worry about a slight delay, you're still getting the miles in. If you are racing a set course in training and you have to pause have you considered that pausing at lights gives you a break which lowers your heartrate so auto-pause is giving you artificial results, by making it look like you didn't pause and therefore allowing you to run quicker on average than you actually could have. On a race you don't need it as you shouldn't be stopping!

Anyone else agree?

CW

Must say, I don't get Auto-pause. When I'm out on a training run and I'n crossing the road I don't worry about 10 secs delay. It's a training run and I'm not going flat out so why worry about a slight delay, you're still getting the miles in. If you are racing a set course in training and you have to pause have you considered that pausing at lights gives you a break which lowers your heartrate so auto-pause is giving you artificial results, by making it look like you didn't pause and therefore allowing you to run quicker on average than you actually could have. On a race you don't need it as you shouldn't be stopping!

Anyone else agree?

CW

CW you make some good points. For me it's not a big deal and I won't be using auto-pause. I do use it on my Edge 1000 as I like to see an average cycling speed without stops which can be significant if you stop for a coffee for example and don't want to mess around with manually pausing the activity. On the Edge it works perfectly.
I guess where the watch is concerned people may want to use the feature. I thought it was useful however to point out the odd (?) behaviour that I was experiencing with delay in timer restart when GPS lock is lost which can be significant in terms of some of the recorded data such as time and distance. In my view Garmin should look at this but not put it at the top of their list of things to fix!

If you are thinking of using auto-pause I would test it out first.