It would be nice if you could store crank length in the Garmin profile and then create a profile for each bike. I am constantly swapping my power pedals between my gravel and xc bikes and manually changing crank length each time.
I think this is true for me as well.
I run XT and XTR pedals on all my gravel and MTB frames, it seems like the clip in/out resistance increases dramatically with any hint of dirt near the Favero, whereas the Shimano seems a lot more resilient in that sense. But perhaps the interfaces just need some polish from use and it will improve over time?
This has happened to me several times and lead me to doubting all my training over the winter on them.
My solution has been to use my Garmin with the crank length set to verify power readings.
Check double check.. this is the solution… thx…
fyi… I got the pedal back from Assioma after their testing, no issues observed:
“After conducting several tests, no anomalies were found in the product. Additionally, to further investigate and ensure the proper functioning of your pedals, the latter were tested directly on a bike by our technicians, comparing the data with another power meter. During this test, the values transmitted by your Assioma were identical to those of the other power meter. For this reason, we are confident that your pedals are working correctly and that the situation you experienced may be due to an external factor unrelated to Assioma.”
Detail Question for people using these. Has anyone done any more detailed power testing and comparison against other power meters? I.e. - really looking at Average Power and NP as compared to another power meter while dual recording? I have been…
I’ve had a Quarq on my MTB that has never given me a single issue, ever, in my 3rd season with it with the slope staying the same. Never felt like it’s varied or been off, don’t have any data to support that it has when I compare it to other meters or my trainer, it’s my reference right now.
But, it seems like every time I compare my Assiomas to this Quarq, they read anywhere from dead on, to 5-10 watts low. And this is after I went through an initial calibration on the trainer a month or two ago to adjust the Assiomas to match. Not talking out of the box readings here. Crank length is correct and the same. Unless there’s an even newer Firmware than the one discussed above, these have been updated
Sometimes average power is the same and it’s NP that’s off. Sometimes it’s average power that’s low. At first I thought it was just outdoor and NP that was different with the Quarq reading higher peaks or more spikiness - but that isn’t it. Just did a controlled environment erg mode trainer workout, where average power was 5-10W low for Sweet Spot. I’ve also had outdoor workouts where the Assiomas read Average Power HIGHER than NP for long steady state intervals. I’m pretty good at holding power - but not that good.
TLDR, beginning to suspect mine aren’t as consistent as the original ones I had where I’d had to do the conversion, and while they’re really fine and close enough, 5-10 watts does matter when I’m trying to do a long 100% threshold interval. Could it be the Quarq? Maybe, but it doesn’t seem that way.
Anyone else?
@FergalK - Any more testing on your side or did you ever figure out the issue you were having?
That’s kinda interesting. As I understand it: PMs just report power and AVG and NP are calculated by whatever app is analyzing the data. Mathematically, for the same data, NP can’t be less than AVG. I wonder if there is something going on with conditional factors like dropouts, zero averaging, and/or pauses…
So, I think that NP can be lower. At least I do see it in certain scenarios in my Garmin Head Units (1040 and an 840). Basically - hit the lap button, do 400W for 30s, then back off to 280, but consciously trying to be as steady state as possible. You can do it on the road, or you’ll see it if you hit the lap button on a trainer in Erg mode. Basically - Average Power Spikes because you’ve got no pauses or drops, but NP Lags. Sort of the Inverse of an NP Buster.
But my thinking is - I’d expect to see that reverse, especially on the road outside, even if I’m doing my best to make it a steady state 20 minute average power interval…
I’m also not seeing any drops in the power data feeds - graphing them out and checking. I do get periodic cadence dropouts from the Assioma’s though.
You are correct. You are taking advantage of the start-up effect of the NP calculation. No NP is calculated for the first 30 seconds until the 30 second rolling window is filled up. Then the NP calculation starts but those 400W samples keep getting shifted out of the window to be replaced by 280W samples. This lowers the NP compared to the average calculation where those 400W samples get equal weight with the 280W samples.
So it’s more correct to say: excepting start-up window effects NP cannot be lower than average power. 
This reminds me a bit of power curves where you’d think they’d be monotonically decreasing as time increases. Yet you often see a trough and then a new peak implying that you are capable of a higher power for a longer period of time
What about it? I didn’t see anything that stood out.
You asked " Has anyone done any more detailed power testing and comparison against other power meters? I.e. - really looking at Average Power and NP as compared to another power meter while dual recording? "
DCRainmaker has comparisons in his review
Yes - but this was about a specific scenario, and there’s nothing in his review that shows an inconsistency like this. And, he also doesn’t have any NP testing at all, it’s all AP.
Took a look at the video - it looks like they were pretty consistent against each other at first glance.
But, the comment doesn’t resonate for me (And, a comment like that might not be remotely credible either.) I’ve been doing comparisons against other power meters with this specific Quarq for over two years now and I’ve never seen this behavior against another power meter. I’ve now had it happen outdoors, but indoors too in controlled temps with the AC running and always with a calibration at the start.
The tricky part is two things - I’ve had it be dead on for one ride, and then AP on and NP off for another, and then both of them off for a third. It’s never been a drift mid-ride for me. Maybe something going on during the initial calibration? Maybe it is the Quarq, but the other thing is the NP from the Assiomas just doesn’t sanity check for me in some of the intervals I’ve done. Commented this above - but having it be lower than Average Power for an entire 20 minute threshold interval? I know for a fact I’m not holding those intervals on the road erg mode perfect and there’s some surging…
Maybe I just need to keep seeing if I can get more data. I’ve got two more long rides this weekend I can dual record.
~19m GPLama did have some issues with them not matching up. Maybe a thread to pull on?
So, I did a bottom bracket service today and cleaned everything up, just replaced batteries, I’ll reinstall the Assiomas and charge them, check all the settings, then dual record two more rides this weekend and see where things land. I’ve got a bunch of steady state tempo and threshold across a long ride tomorrow…
I use MX-2 to dual record nearly every Zwift race I do (vs. Tacx Neo 2T), and I often dual-record workouts for curiosity. I also sometimes dual-record Zwift races with Assioma Duos. Both pedals tend to be within a couple watts of the trainer for most time intervals, though there tends to be significant deviation in how quickly pedals respond to a sudden increase of high-force (ie - sprinting). The trainer takes a second or two longer than my pedals to hit peak power reading in a sprint. Here’s a typical race comparison (with NP and Average for this race different by 2.9W and 1.6W, respectively):
@bcm I sent the pedal back to asioma for validation and they found no problems at all with it