This discussion has been locked.

You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Pace, distance, and stride length

Former Member over 9 years ago

I've been trying to understand how stride length is determined by the 735xt for use in calculating pace and distance when GPS is off. This shows my current understanding after a fair amount of searching through forums and posts:

In normal activity tracking mode, using device accelerometer (not in a formal activity)
The stride length is set automatically, dependent on your height. You can set a custom stride length manually in GC web or GC mobile. In GC mobile there are settings for Walking stride length and Running stride length, in GC web only for walking. I don't know if setting a running stride length in GC mobile does anything, or how the 735xt would determine when to use it.

In a Run activity without foot pod, using device accelerometer
The stride length is set automatically based on the last several runs with GPS enabled. A calibration table is constructed of stride length vs three activity intensities (low, medium, high) which are based on arm swing. The calibration range takes care of pace ranges from slow walks to fast sprints. It doesn't work well for interval workouts, better for steady runs. A master reset clears the table.

In a Run activity using foot pod
The stride length is set automatically based on GPS. I don't know how it is set from the run (e.g. last part of run? First part? Average? Moving average during run?)? You can also set the foot pod stride length manually to a fixed value. The foot pod only has one stride length, not a table like the accelerometer does. Despite this, the foot pod is generally considered to be more accurate than using the device accelerometer, due to the fact it is directly measuring foot movement. When GPS is on, the foot pod can be set to override GPS for the "instant pace" metric because it is more responsive to pace changes than GPS, especially in situations where the GPS signal is weak.

Does this agree with others' understanding? Can anyone clarify or provide more info?

0 timgrose over 9 years ago

I think in only pure activity tracking mode where only steps are counted and not speed/distance is stride length a fixed metric as obviously you can compute an approximate distance if you know the number of steps and stride length.

To calibrate the internal accelerometer, I believe it basically models your rate of arm swing with pace as observed through GPS measurements. So without GPS it can compute an approximate pace and in turn can compute a stride length as the HRM-Run (or the watch if not using one) measures cadence as well.

Again the foot pod has no concept of a stored stride length. You seem to be confusing that with a single stored calibration factor. I am not sure how exactly it works (a quick Google throw up some suggestions - but none I found were that great - any suggestions?) but it uses the accelerometer in it to effectively dynamically compute an approximate speed/distance/stride length in combination with the calibration factor you give it.
For sure if it ought to be the "best" method without GPS as long as you give it a decent calibration figure. The 735 tries to compute this calibration factor for you if you run with it when GPS is on.

0 Crispin_Ellisdon over 9 years ago

Just posted more in the foot pod thread, but tech detail of how foot pods work is here http://www.garmin.sk/img.asp?attid=8735

0 Former Member over 9 years ago

Thanks for the great feedback! Tim of course you are right about the foot pod calibration not being stride length, for a moment (senior one?) I was thinking of the calibration factor as being distance in cm.

What I don't understand is why the 735 uses a fixed stride length for normal (pure) activity tracking and a table-based technique for tracking Run activities without GPS or foot pod.

Has anyone tested Run activities without GPS or foot pod, using just the device accelerometer, to see how well the 735 tracks speed/distance at different paces?

0 timgrose over 9 years ago

Don't forget that the activity tracking is in "low power mode" so I imagine if it tried to be "clever" it would eat through the battery much quicker.

DC Rainmaker has done some of these tests and from personal experience it tends to work reasonably at a steady pace but less so otherwise. Of course it also depends on consistent arm swing movements. If you keep stopping or say taking a drink it is going to be all over the place.

0 Former Member over 9 years ago

...
I don't know if setting a running stride length in GC mobile does anything, or how the 735xt would determine when to use it.
...

I tried setting the running stride length in the GC mobile app and it didn't have any effect on running distance while in normal activity tracking mode.

My walking stride is set at 2.66 ft and this is being used by the watch when walking. When I run, the watch is using a stride length of 4.15 ft. It should be 3.39 ft but setting the running stride length to that didn't change anything. :(

0 Former Member over 9 years ago

...
Again the foot pod has no concept of a stored stride length. You seem to be confusing that with a single stored calibration factor.
...
.

Tim, I'm not so sure about that. My calibration factor on the foot pod is 104.9; converting that from cm to feet yields 3.44 which is quite close to my running stride length that I calculated today of 3.39. The difference could be accounted for due to different shoes or different terrain.

0 timgrose over 9 years ago

Those calibration factors are actually percentages so 100 means 100% i.e. that whatever the foot pod computes the pace that is the pace displayed on the watch.

If it is under 100% the pace shown on the watch is "slowed down" from the raw value and in your case would be sped up.