RChung

As you say, it depends on what you're focusing on. I don't use a Garmin so I'm not quite sure what "Autopause" does but I *think* it mostly affects what's displayed on the head unit's screen in terms of averages but it doesn't affect the data that are being stored and then downloaded for analysis. -->If that is so<--, when you download and analyze the data with Golden Cheetah, the TSS doesn't get inflated by stopping time (so your PMC ought to be unaffected). I don't use TrainingPeaks so I'm not sure what it does but if I did I guess I'd check it.

DaveLeeNC

While I have never checked the data, this is what Garmin says that they do.

dave

You can use Auto Pause to pause the timer automatically when you stop moving or when your pace or speed drops below a specified value. This feature is helpful if your activity includes stop lights or other places where you need to slow down or stop.
NOTE: History is not recorded while the timer is stopped or paused.

PepeM

I'm pretty sure the Garmin stops recording when autopause is on. Affects some stuff more than other. Things like W' get really screwed. Things like Average power and normalized power are also affected too, as well as heart rate. Average speed too I guess but I don't look at that one much.

RChung

Hmmm. Could you look at the data downloaded from one of your rides to see whether power and speed are recorded as zeros when you stop at a stoplight? I've seen data files collected from Garmins where after a couple of seconds the data stream stops, then when you start up again the data stream continues (with a timestamp, so you can tell how long the gap was).

Seattle Forrest

^^^ In my Garmin data when I used auto-pause I would not see a long string of zeros in the file. It would stop with a time stamp and then resume with a time stamp, and those would tell you how much time had elapsed. It's expressing the same thing, just more economically.

If you're using a Garmin, make sure you're recording a data point every second and not set to "smart recording."

PepeM

Just recalled that I was using another head unit when I had autopause on. With the Garmin it has always been off so I am not sure how the data would differ with it on.

DaveLeeNC

I don't have a simple way to investigate (directly) a .fit file. But in looking at Golden Cheetah's export of .fit files to .csv files (and their direct .fit editor), I see that there is no data (beyond a time/date stamp at each end) during a pause. But just to be sure I generated two more .fit files in my backyard today.

1) About 20 minutes of (mostly) autopause with about 13 seconds of actual recording with the timer running
2) 13 seconds with the timer running (a very short recording session)

These two files were the same (minimal) size at 2K. So I am pretty sure that a Garmin 800 records no data during an autopause (but there is time/date info available in the file).

dave

RChung

Very nice. Thanks. What happens when you turn autopause off and stop for, say, a minute? Do you get a minute's worth of zero power and speed?

DaveLeeNC

I can't recall the last time that I did a ride with Auto Pause disabled. But I went out in the back yard and walked around for 10 seconds and then stopped for 50 seconds (Auto Pause off). In looking at the file there are a bunch of zero entries (covering the last 50 seconds).

If I were on a bike with power, HRM, wheel sensor, etc. being detected, I can't imagine it being any different.

dave

Drew Eckhardt 

With autopause off you get a minute of zero power and speed.

The behavior is the same regardless heart rate sensor presence.

revchuck 

I have it turned off. Strava automatically eliminates the pauses, and my speeds are higher on Strava than on Training Peaks.

RChung

That's interesting. I don't use either a Garmin or TrainingPeaks so I didn't know that. I'll have to think about what the consequences of that are.

Seattle Forrest

Yes.

A Garmin (at least an Edge 800 and a Fenix 3) provides exactly the same information in the fit file whether you're using auto pause or not. There's either a zero every second until you resume moving, or the recording stops but the time stamps make the gap obvious.

I guess technically the auto-pause recording will be slightly different because it takes about a second for a Garmin to realize you've started/stopped moving and pause or resume the recording. Maybe that auto-pause lag under states your efforts if it ignores some of the power you generate to start moving from stops at lights and whatnot?

PepeM

Another difference I can think of is on heart rate recording, if you're into that.

RChung

Good point. I haven't used a HR belt for years so I tend to forget about it. From the point of view of power and speed, as long as you have the time stamp (so you can identify gaps) recording the zeros seems like a waste -- but not if you still pay attention to HR.

gregf83 

The information is similar but results in different metrics being calculated by most training software, at least it does for Golden Cheetah.

I leave auto-pause on mostly to eliminate coffee shop stops. On group rides we'll finish at a coffee shop that's about 10 min from my house. If we do a 3 hr ride and stop for an hour at the coffee shop I don't want any calculations based on a 4:10 ride.

RChung

Well, that's interesting. My a priori guess was that as long as I wasn't stopping for coffee, the effect of traffic lights and stop signs would be relatively small but I thought I should check to be sure before I opened my mouth. I just repeated a short 40 minute ride I did yesterday, once with Autopause on and once with Autopause off. Elapsed ride time was almost the same, average speed was almost the same, NP was almost the same so IF was almost the same, TSS was 10 points (=about 20%) different.

DaveLeeNC

Could you expand on this as it doesn't seem possible. If IF and duration is the same, then TSS is the same. Additionally if you intended to do 'the same ride' with Autopause On and Off, then average speed would have to be less in the Autopause Off case.

I think I am confused about what you did here.

Thanks (and Happy New Year).

dave

RChung

Happy New Year to you, too!

Both rides, same route.

[TABLE="width: 500"]
[TR]
[TD]Autopause[/TD]
[TD]TotalTime[/TD]
[TD]MovingTime[/TD]
[TD]Mean Kph[/TD]
[TD]NP[/TD]
[TD]IF[/TD]
[TD]TSS[/TD]
[/TR]
[TR]
[TD]On[/TD]
[TD]47:52[/TD]
[TD]34:50[/TD]
[TD]20.0[/TD]
[TD]184[/TD]
[TD]0.877[/TD]
[TD]45[/TD]
[/TR]
[TR]
[TD]Off[/TD]
[TD]42:28[/TD]
[TD]36:15[/TD]
[TD]19.3[/TD]
[TD]184[/TD]
[TD]0.878[/TD]
[TD]55[/TD]
[/TR]
[/TABLE]

Evidently, Golden Cheetah used "moving time" for the first ride but "Elapsed time" for the 2nd in calculating TSS, but "moving time" for both for calculating average speed.

I need to do more investigation, but those TSS's are consistent with that.

DaveLeeNC

Interesting - thanks.

I would assume that Golden Cheetah would include/exclude the same time intervals when calculating NP and in calculating TSS. But you really can't tell from just the summary data.

dave

RChung

Prompted by a comment in another thread I re-visited this thread. I think there's some good though subtle information here.