Hi - I’m grappling with the best way to do the syncing, and duplication handling - but I’m also wondering how important some of it even is:
Right now, when I do a virtual ride, Strava gets the ride with simulated elevation and grade info, which Garmin does not because it’s getting the ride from my Edge 520.
So if both Strava and Garmin are syncing with intervals- my understanding is the data from Garmin will “win” but name changes and such will come over from Strava.
In that case though, I still have the ride without elevation.
If I import the virtual ride into Intervals, there is a bit of a difference in load - somewhat counterintuitive because the Virtual ride, which doesn’t include warm up/cool down has a higher load value than the Garmin data which includes those periods.
Another difference is that I have the Edge set to record every second, but that isn’t necessarily true for all the Virtual apps (eg Bkool appears to be very 2 seconds).
Before I go farther down this rabbit hole…is one of these inherently more accurate? Does incorporating elevation/gradient in the file change anything significant in the analysis?
I can figure out workarounds to get the Virtual version into Intervals - but first question is - “should I?”. It “feels” better to have all the data captured here - but maybe that’s more about my subjective experience than about the analysis.
I personally just like the D+ cos it makes me happy that I did x amount of D+ (even tho for Zwift, if you have the “trainer resistance” set at 50% for eg: Did you really climb 1000m?)
As someone who also dual records with a garmin, I miss having elevation from Zwift via Strava. Regardless of how real it may or may not be, effort can increase going uphill, through sprint points, etc… and I wish I could tie that in.
That is absolutely true on a trainer. There is no difference between 200W an a virtual flat section and 200W on a steep virtual hill. Outdoors, there is a different feeling but not on a trainer. You can’t realy simulate the force that is dragging you down on a hill and influences your power production throughout an entire pedal stroke. A pedal stroke on a steep hill is much more jaggy then one with the same power on a flat section.
All parameters like distance, speed, elevation and gradient on a trainer are backwards calculated from the power you produce. Those calculations make assumptions regarding wind-speed, rolling resistance, your frontal surface… You can easily cheat by reducing your height and weight to reduce frontal surface for wind drag and weight for hill drag.
I know that a lot of people will disagree, but those numbers mean absolutely nothing on a virtual ride.
Everybody is free to have their own opinion, but I’m not using those non-acccurate numbers. If you like to have them, who am I to hold you back?
PS: using Strava as a source means that you will not have L/R balance data because Strava isn’t storing that data. This is an accurate metric but there is very little you can do with it unless you are tracking recovery of an injury that caused a weaker side. You can’t train to make your balance 50%.
oops… you caught me there… This is a ride on my BreakAway Trainer App, whereby i tweaked the aero profile (cda) to be super duper more aero than superman. (i was the ride was like 80-100w, nursing a not so great achilles / hamstring soreness and yet still wanted to try to complete the Strava / Rapha Festive 500 - which in the end, I ended up w/ only completing 244/500km )
It’s actually easier outside vs on the trainer to hold say 120% of FTP for 1m vs indoors. In fact, I managed to do like 5x 1m@125% FTP outdoors this morning and it was much easier than holding that power on my trainer. ¯_(ツ)_/¯
I actually use these numbers (mostly the distance) for me to calculate the wear and tear (and when to lube) my chain. (I just use a rough estimate of 1hr = 30km and manually enter them into Garmin)
exactly what i’ve read etc, but still ended up getting the assioma duos and then realising that I really didn’t have much use for the data.
I tot that now Zwift can sync directly into Garmin, so you will have the data, unless you’re like the OP where you don’t get the warmup/cooldown part of it? (i don’t zwift - too much $$)
OP here…so differences in calculated load between the two sources would presumably be because of…saving data every 2 seconds (BKOOL Platform) vs every 1 second (Edge 520)? Or different calibration? (Is the calibration of power specific to the recording device/platform or inherent in the trainer itself?).
I did the spin down calibration with the wahoo app and with the Edge 520. It wasn’t clear to me if that was redundant or if they each need their own calibration.
I guess the other reason it’s nice to see the simulated elevation and gradient is being able to look at things like cadence, HR, power in relation to those . Since resistance of the trainer is not captured anywhere (at least not in simulation mode), without seeing elevation/gradient, it’s not obvious why my cadence dropped or my HR went up…
how big is the Load differrence?
Load is calc
Load = (sec x NP® x IF®)/(FTP x 3600) x 100
So, NP calc may be affected if your head unit /app isnt getting the same number. (it happens) but if it’s off by a few points, it personally doesn’t seems much diff.
Calibration is inherent on the trainer. The result of the spin-down calibration offsets the speed/power curve of the trainer. The software or head unit only triggers the trainer to measure the spin down time and adjust the curve offset.
Got it, thanks. Makes sense.
So the difference (which was around 5 TSS units) is most likely due to the different resolution of the data. So I think I won’t worry about it.
w/ regard to trainer calibration, somehow (my non-technical explanation - despite me reading ANT+ specs etc and looking at calibration etc) I still recommend ppl to use their own trainer software/app to do the calibration rather than some 3rd party apps etc.
I know for eg: Wahoo trainers does some “unconventional” things to their trainer firmware and thus theres sometimes some trainers which “behaves” differently from the official ANT+ specs.
)
