Gone is the warm weather of Australia and Singapore, and now it’s back to the mostly dreary weather of Amsterdam. Though honestly, it wasn’t that bad this weekend. In fact, it felt unseasonably warm for the first weekend of February. I guess I can’t really complain. In any event, here’s what I was up to.
1) A New DCR Studio Table
As anyone who has ever watched any one of my videos since moving into the Amsterdam Cave about a year ago knows, the table I used in front of the work-bench for unboxings and such was a bit…umm…wobbly. The table is actually awesome in real-life, but not super ideal when a ceiling-mounted camera is zoomed in on a small watch, making even a 1-2mm movement look massive.
We originally bought the table for the Bertie’s Cake Studio in Paris, actually picking it up on a weekend trip to Dijon (yes, the mustard place). It was on clearance at a small furniture shop. We had our bikes with us that trip and managed to disassemble the table and get it fit into the van. Woot! It was used as a display and photography table for cakes for years. I slowly started doing product photography on it too for a while, though that was cumbersome removing all the display cake stuff each time. Actually, this older post manages to capture both those two things one after another!
Given that, this fall I was determined to pick up a more stable table. I found one on a Dutch auction site and had it delivered back in December, but didn’t have time to get it all re-assembled till Friday. Had it just been simple assembly it’d have been a 3-minute task. But I wanted to make the table taller – about 10-15cm or so ideally (3-5”). For some odd reason, all the tables we’ve got in the Netherlands are disturbingly short.
I got it unpackaged:
Then I went and grabbed some wood from the wood dumpster at our building. I ended up cutting it in half. Regrettably the electric saw and regular hand-saw were at home for some reason, so I had to use very much not the right tool for the job. Still, faster than going home and back.
A boatload of giant-ass screws later, and I had made it taller:
And, done:
It doesn’t wobble. Mostly because I can place it directly against the wall. Not because it’s any less wobbly after my ‘height increase’. Standalone it’s wobbly AF due to that.
But again, no wobble against the wall now.
2) Gathering more Peloton Power Accuracy Data
I don’t really have a timeline for when I’ll publish a Peloton Bike In-Depth review (after buying one used back in early January). Maybe March. However, I have been getting in plenty of rides. All of which I’ve got Garmin Vector 3’s on there for power accuracy comparison. At some point I’ll switch to Favero Assiomas for a few weeks, merely to demonstrate it’s not just N+1. Maybe even a pair of PowerTap pedals too. Either way, two more Peloton workouts in this weekend.
The one thing I continue to be somewhat intrigued upon is how ‘misunderstood’ Peloton is within the endurance sports community. For example, if I were to show you the structured workouts in an app-agnostic way, my bet is that some workouts would be indistinguishable between Peloton and TrainerRoad. Especially those that are power-zone focused. Yup, just like TrainerRoad they’ve got precise power targets based on FTP as well as ancillary targets like cadence. Of course, there’s also plenty of other differences. But the more I use it the more I see the similarities.
In any case, here’s a few more power accuracy examples. With about 10 rides on it now, the trend seems to be pretty clear: It drifts as it warms up, with the most significant drift happening after about the 20ish minute marker. Here’s an example from Friday:
You’ll see that the first while it’s incredibly close, and then it starts to meander. While one could try and argue that it’s a Vector 3 drift issue, I’ve got 2.5 years and probably hundreds of rides of evidence against dozens of other power meters to indicate Vector 3 simply doesn’t drift. It might have plenty of other reasons to be upset about, but drift isn’t one of them (also, why I’ll swap to another pedal power meter or two for fun).
I haven’t done any non-instructor workouts yet to try and narrow down exactly where it’s most susceptible to drift. Meaning, does higher cadence drive more drift, or higher power, or something else? But time certainly does, which I think is heat-driven.
Why do I think that? Well, aside from things typically warming up – on Saturday the radiator directly next to the bike decided to turn on right at the start of the workout. This is the first time this has done it, but given it’s less than 30cm (1 foot) from the bike, it has an immediate impact. And in doing so, the drift started at the 8-10 minute marker, rather than later in the workout.
Also, for the fun of it, if we look back at the test numbers I did before I disassembled the bike and relocated it thousands of miles away, the drift was there too. As it has been in the three other locations around home that I’ve used it in.
Again, more data to collect here. Also, I haven’t contacted Peloton to see if there’s any tricks to mitigating drift or such. Typically speaking if it were just an accuracy/calibration issue, it would be consistently wrong across the entire workout. But it’s going from +/- 0.5% earlier in the workout to +/- 5-7% later in the workout Of course, before folks get all outraged, it’s probably important to point out that there’s far more to a Peloton bike than the power accuracy. That’s sorta why they’ve sold approximately 500 *times* more bikes than Tacx, and probably double that again for Wahoo, all with attach/renewal rates that Zwift would dream of. But we’ll get into that later.
3) Under the cargo bike
Ok, two long chunks above. So here’s a quickie! I mounted a GoPro under the Urban Arrow coming home from the office on Saturday. We had brought the kids down there to practice riding bikes inside (it’s a massive undercover warehouse basically). I’ve been wanting to mount a camera under there for a while. Initially I mounted the camera a wee bit close to the ground, and made it approximately 5 meters from the front door before I clipped the concrete. A short bit of repositioning later and I was good to go.
It’s a fun view. I put the video up on Instagram here in the story archives – just tap ‘Cargo Bike’.
4) Track Run
I headed out Sunday for a quick track workout. I had about an hour to spare once the kids went down for nap, so by the time I got all my watches ready and outside that was about 50 minutes of usable time. I figured I’d go to the track, do 800’s till my clock ran out, and then run back.
It was crazy windy. The bar for wind in the Netherlands is reasonably high. But the way I can tell how windy it is, isn’t by going outside. Rather, by what type of aircraft are landing over the top of our home, and about 10 seconds later, the track. Typically this runway is just used for business jets and nearby European flights (narrow-body aircraft).
But, when it gets crazy windy it becomes widebodies all day long. A380’s, 747’s, 777’s – you name it. And that was the case then. The running track is on short final to runway 04-22, though every once in a while I’ve seen some cargo 747’s do some crazy approaches to 09-27 making a hard right bank at the last minute in.
In any case, my 800’s were nice. Nothing crazy. I was doing them at a ~4:00/KM (6:20ish/mile) pace, depending on my luck with the wind. Some were closer to 3:50, and some were 4:05/KM. I’d do 200m walk/jog in between, and the direction of the wind kept changing slightly, just enough to go from ‘shrug’ to ‘!@#$#@’.
Post-run, I thought this shot was pretty:
I’ll probably use it on my Vivoactive 4 review. Might even finally do that this week. I’ve got 6hrs and 18 mins left on this flight’s worth of time to kill.
5) Fresh Pasta with the Kiddos
After getting back from the track it was time to make fresh pasta with the kids. They helped with some of the mixing of ingredients (just flour, water, eggs, and salt), and then kneading of the dough.
The entire process doesn’t take that long – perhaps 30 minutes from start to finish, including letting the dough rest a bit. I bought the KitchenAid stand-mixer pasta maker attachment kit over a decade ago, which includes a flat sheet roller, and then two cutters. One for fettuccine, and the other for spaghetti. I personally prefer the fettuccini one, but the kids want spaghetti. So…spaghetti it was.
Afterwards they selected pesto sauce, their favorite. But…I’ve got no pics of that. Sorry! With that, the weekend is a wrap!
Thanks for reading and have a good week ahead!
FOUND THIS POST USEFUL? SUPPORT THE SITE!
Hopefully, you found this post useful. The website is really a labor of love, so please consider becoming a DC RAINMAKER Supporter. This gets you an ad-free experience, and access to our (mostly) bi-monthly behind-the-scenes video series of “Shed Talkin’”.
Support DCRainMaker - Shop on Amazon
Otherwise, perhaps consider using the below link if shopping on Amazon. As an Amazon Associate, I earn from qualifying purchases. It doesn’t cost you anything extra, but your purchases help support this website a lot. It could simply be buying toilet paper, or this pizza oven we use and love.
Is that the Vivoactive 4 on the track pix?
Yes it is.
I’m using vector 3s on my 2 year old Peloton. Looking forward to your analysis and happy to cross-check anything for you if that’s useful.
Nice! I’m definitely up for gathering data sets from a larger group of people. Based on what I’ve seen thus far, there’s a bit of variation in Peloton accuracy from bike to bike. With newer bikes generally having a higher QA bar and being more accurate. My bike was originally bought in Feb 2019, so a bit newer in the grand scheme of things.
Looking at the left/right graph, the Peloton is tracking right side power very good, and the drift may be a user imbalance. Does the Peloton only measure right side power?
It only measures total power. What you’re seeing there is for sole-power sources, the DCR Analyzer simply splits it in half, giving you a virtual left/right. It can be useful in some analysis cases, primarily for left-only units. I’ll disable it on those data states to reduce confusion.
For a non-native speaker, what is “app-agonist”? Agnostic?
Right, app-agnostic. Thanks!
But, yeah, I’m basically saying there that if I were to show the outputs of a workout on, say, TrainingPeaks, from both Peloton and TrainerRoad – I bet I can find a few that actually overlap exactly in terms of structure. Primarily the power zone workouts.
Which isn’t to say either is copying each other of course, but merely to point out that a lot of Peloton workouts are actually way more structured than people realize. And some of them even have legit ‘science’ behind them in terms of spending time in certain zones. Of course, some of them are also definitely not, and have no basis in terms of optimal training methods. They might just give you a good workout, but not be optimally structured.
Peloton has some workouts that actually list the target cadence/power profiles for them. This screenshot isn’t the best example of a super structured one, but rather just one I came across on Friday. I’ve gotta dig into what, if any, pattern there is on when they do or don’t enable this. It seems super random. However, as always if looking at aerobic on non-interactive bike, it’s a combination of both cadence and resistance applied that drives power output.
Minor quibble: should be “Assiomas” – no apostrophe.
It’s funny, my spell-check flagged that. And my brain paused and was like ‘Huh, that’s weird – it’s never flagged it before.’
Which, is the usual sign that something is amiss (since Assiomas is already in my spellcheck…). Thanks!
Cool to see someone looked at this. While the Peloton PM is off, it seems absolutely fine for that equipment for that customer. Obviously, that can be solved by some software compensation. Better than needed even. I’d like you to do the same test on a few common Stages spin bike PMs. For most customers, works “good enough” will yield the same result as works to “scientifically accurate”.
taking this a step further, I’d like to see a yearly ‘value’ power meter shootout. I’d argue that the world needs a ‘recreational level’ PM at a lower cost, that sacrifices some accuracy to get more PMs out there. I’m not sure if there’s an opportunity to shave $100 – 200 off a pedal PM by manufacturing one less accurately or inferring power data by some other means, but accepting compromised power is better than paying for full accuracy for most folks. The powerpod is too finicky, there’s that tire valve one that doesn’t work, and the Avio one that also doesn’t work well- these are steps in the right direction IMO.
Low accuracy power meters already exist for lower cost, and the same in the trainer world (Tacx Vortex for example, or the Flux 2). Ultimately accuracy is used as a selling point and so better is better and we’ll gradually see accuracy improve accross the board and high end trainers will compete on things like Neo road feel.
In the case of Peloton, although it’s high end and expensive, the power accuracy isn’t one of the selling points so there’s less focus there. Peloton have the money and resources to fix this in the future, and once a light is shined on that I’m sure they will. Ray’s review may well be the catalyst there and we might see a v.next in 2020 with accurate power and temperature compensation. Some Peloton users might upgrade but the majority will shrug because it doesn’t affect their workout at all.
We need to remember that for the vast majority of humans on this planet, just exercising more than once a week is the improvement. I’d say even most cyclists buying Neos and Kickrs benefit more from more time in the saddle than accuracy of power. User engagement is hard to measure consistently and objectively as a metric for reviews though, and if Ray tried to do so the comments would go crazy! He does usually comment on it for his own experience though, which is nice.
Are you considering Technogym’s Skillbike for a review or comparison at any point in the future?
I’m a bit confused about the power data it reports to Strava. It only shows average workout power and it seems high.
link to technogym.com
2 interesting points with the Peloton you might want to know. 1) I have had a firmware update drastically change my bike and how difficult it is. 2) Through the re-calibration process you can modify their algorithm substantially so bike to bike can have a very high degree of variability. When trying to fix the accuracy of my bike I was re calibrating a bunch against a power meter and had it so that a resistance of 45 and cadence of 78 was 288 watts which it shouldn’t be. If you want to get into playing with calibration they will send you the little kit. The tighter you go against the spacer on the flywheel the harder the bike becomes. Back off from that piece and it gets easier so you can make it a “beast bike” or “extra hot” to post those huge numbers if you want.
That’s interesting info. I didn’t know there was a way to calibrate the bike at home. I have noticed that some people put up power numbers that just can’t be possible (at least not possible for non-pros). I would guess a resistance of 45 and cadence of 78 would result in a power of about 145 on my bike.
I have noticed that my power is down 10% recently. I’ll have to chalk it up to colder winter conditions and a firmware update as it can’t possibly be a falling fitness level 😉 Although, I took an in studio ride and my power was about 10% lower and my average heart rate 10% higher. That time I used the excuse that it was so hot in the studio that I ran out of gas halfway through the class. So too cold of a temp reduces my power but too high of a temp has the same effect.
I guess as long as the power stays consistent (which it won’t if firmware updates can impact it), then it is good enough for my training. Now Peloton just needs to schedule more 45 and 60 min Power Zone rides.
I’ve been using a power meter on a bicycle for 6+ years and my wife has a peloton. I re-calibrated her Peloton (using the kit) after the first time I rode it and realized it was harder than expected. It more or less lined up with their specs on cadence, resistance and wattage.
When I’ve ridden on it, I definitely noticed the drift. I didn’t have PM pedals to test it against, so it was just by feel.
The calibration basically involves setting the magnetic calipers a certain distance away from the wheel (with a tool) and this is your zero resistance set point. You then turn the resistance knob 90 degrees (with a tool) and accept basically all the way up to 100% resistance. Obviously, there is plenty of room here for manipulation to make power numbers better than WorldTour pros..
Interesting observation on the heat. I figure the resistance is calculated based on the distance the calipers were from the wheel based on calibration and then computed by the resistance curve/algorithm. Maybe there is an expansion of some part of the system that affects the caliper distance.
That pasta looks great!
Re table wobble, add some bracing or bracketing to the legs. 🙂
Yeah, that’s roughly what I started doing. Then I slid it against the wall and was like: Well..err..done. 🙂
Hi
Would you have a review or a comparison for the new garmin watches: Hero, Star Wars and Tactix?
Thanks
Nothing planned. The Star Wars one is simply a Vivoactive 4 (all the Marvel ones), and the Tactix is essentially a Fenix 6 variant of the Tactix Series. I’ve done occasional reviews in the past of those, but just run out of time for such little differences.
thanks…
I noticed that you have the Vivoactive 4 (?) on the track pics. I’m thinking about upgrading my Forerunner 35 to a Vivoactive, but can’t decide whether to get the VA3 or the VA4. I don’t really need it for anything spectacular, just gym time and running a few times a week. I noticed that in the 2019 Gear recommendations you suggested the VA3 over the VA4, is that still the case?
Btw, nothing better than homemade pasta! 🙂
Is the VA4 looks not swaying your decision?
Not really to be honest.
Fwiw, I should have up my Vivoactive 4 review tomorrow. My text is done, just doing accuracy charts now, and then photos in the morning.
That would be great! Thanks for the reply!
So, it’s a copy of the Venu review with comments about the older screen technology? 😉
I had honestly hoped it would be that easy, but in the end I re-wrote virtually everything from scratch – saving a few snippets here and there.
However unfortunately, I found the GPS & HR accuracy lacking substantially compared to the Venu. No idea why.
Glad I bought the venu then 🙂
Sorry one last note on the Peloton. I’ve had one for a couple of years now and belong to several groups where people have tested vs. power meters before. If you’re interested in data about Power Zone training, the graphs or power comparisons I can probably source some of that stuff for you.
Pesto is the favorite pasta sauce at my house too. Last summer we started to grow our own basil plants, which is dead simple. I followed the instructions in this YT video: link to youtube.com
“Basil, How to Grow More Than You Can Eat”. He uses artificial light, I did it in the window sill. Indeed, we ended up with more basil than we could ever eat but you can always give it away.
I can’t wait to see an iq2 visit in a similar update
Dont you mean “Pasta” Comment?
We’re you tempted by the Dutch Headwind Champs?
Noticing the drift as well using my assioma duos. Took a 60 minute ride today. Over the first ten minutes, the assiomas were 1.3% better than the Peloton. Over the last ten minutes, the assiomas were 10.5% worse than the Peloton.
I installed Vector 3 on my Peloton Bike+. Besides updating the crank length to 170, did you do any other configuration changes on the pedals ?
That’s the only modification I made for my Assiomas was crank length and then zero them out.