Runpower is a "developer field" which is documented (FIT SDK). FIT files exported from Connect (" export original') from Garmin Connect do not include the runpower developer field though (disclaimer: I only tried it once) so that may also be a reason that not all software seems to support RP (if exported files are used instead of files directly from the watch USB storage. The RD pod ANT+ profile is still "secret" in the sense that it is only available to those who pay 1500 USD to the ANT+ alliance for early access membership. Otherwise it is waiting. Until the spring, I have been told.
Odd that the "original" file is not seemingly actually that. Will have to test that myself. Generally if am "messing" with FIT files would be the original but sometimes use Export Original from GC. Not noticed anything missing before.
"Export Original" from GC with CIQ fields in the FIT file fails to import into SportTracks (ST3). FIT files that include running dynamics but not the Power CIQ fields import just fine. I just use export to TCX from GC, which strips the CIQ fields, and use that for import into ST3. I believe if Power were to be 'native' to FIT, instead of CIQ, the file would import without error. Further, Garmin users wouldn't have the 2 field CIQ limitation per activity. Hopefully, Garmin will do this going forward as it has with other running dynamics (vert oscillation, GCT, etc.).
Some more quick feedback. I've been using RunPow for each of my runs and I can't say I'm finding it very useful with little variance between easy runs (around 375w) and TPace intervals (430w). Intuitively the easy run value seems way too high and the small variance makes it hard to "relate" to the data.
Initially I was pretty impressed with the wind factor but I'm finding that it exaggerates the calculation more than it should really. Yesterday I was doing some fast 200s with and against the wind at pretty much the same speed and got 430/600 values on average. Now there was some wind but the perceived exertion was certainly not that different ! GC shows a wind speed of 31kph and darksky a wind speed of 22kph that seems more realistic.
Based on the lack of feedback here I'm guessing everyone is rather perplexed by RunPow ;-)
A bit high perhaps if it was an easy run as it's upper end of zone2 (Calculated with FTP of 430W), but close enough I guess.
I'm still recovering from an injury but have slowly started running short distances again. Currently I'm testing Garmin running power while using Stryd as pace source. Stryd will also gather offline data so I can compare the data.
So far seems the two are actually quite similar comparing the graphs, but Garmin seems to give higher uphill power (compared to average). Downhill power so far seems at least on par with Stryd, so that's a good first sign.
As I'm still running granny paces it's to be expected that my uphill power is a bit higher as I'd practically have to walk to stay on target so will see if it will even out when I'm able to run faster paces again. Consistency is difficult to determine at this point as I'm currently in a phase where gains are to be expected until I reach my previous fitness levels.
1. Very consistent between two runs on the same course (and therefore same surface) at the same speed, cadence, and HR. So, the algorithm is internally reliable but on the other hand, speed/pace and HR might be just as good as intensity indicators. However, Power might just be the composite 'score' that can be used, especially on easy days, to keep intensity from creeping up to a moderate level (a recurring problem for me).
2. Power is not useful comparing runs on different surfaces. I ran a hilly, technical trail run the other day at high intensity and the Power score was about 30 W less than a moderate intensity road run a couple of days later. Perhaps no way to incorporate surface calculations in the algorithm unless there's some kind of check box like with wind.
I tend to agree that running power on the face of it does not appear to change much with pace.
On the other hand in the example of going from 375W "easy" to 430W "intervals" if I went up 55W on the bike in a steady state effort I would certainly know about it.
Indeed last night on the treadmill starting at 10 kph "very easy" and finishing at 13 kph "bordering on tempo" saw a rise from about 275W to 340W so a fairly similar rise but 3 kph faster is quite a lot.
However on a treadmill run pace and power are directly correlated it seems (assuming constant gradient) so not really telling me anything "new".
I did a beach run the other day and like the trail run the extra effort needed is unsurprisingly not modelled so think power is perhaps best for rolling road runs where the effect of the uphills can be harder to gauge.
I would hope Garmin is working on incorporating this into a firmware drop. It's silly that this takes up 1 of 2 CIQ...either that or bump the CIQ max to 3. Come on Garmin....
I tend to agree that running power on the face of it does not appear to change much with pace.
On the other hand in the example of going from 375W "easy" to 430W "intervals" if I went up 55W on the bike in a steady state effort I would certainly know about it.
Indeed last night on the treadmill starting at 10 kph "very easy" and finishing at 13 kph "bordering on tempo" saw a rise from about 275W to 340W so a fairly similar rise but 3 kph faster is quite a lot.
However on a treadmill run pace and power are directly correlated it seems (assuming constant gradient) so not really telling me anything "new".
I did a beach run the other day and like the trail run the extra effort needed is unsurprisingly not modelled so think power is perhaps best for rolling road runs where the effect of the uphills can be harder to gauge.
Indeed, you'd think that the "struggling" to run on the sand would be picked up by the accelerators in the strap !
I was going to comment that I've never seen such a low value as 275w, even when walking, but then I noticed that you didn't go very high at 13 kph, so it's probably just better running economy on your part ;-)
