So I sanity checked the Wired RealAxiom trainer against a powermeter today and the Axiom seems to read 40-100w high consistently using the information Elite has supplied to us. A quick search showed a few people on Elite’s forums reporting the same thing.
The question I have is, are you all who own a wired Real axiom aware the power is high, and is there a common workaround for this? The problem also seems to extend to the wireless Real Axiom.
Yes, sort of. There are three numbers which RA tech had given me which dealt with adjust the feel (and therefore, power) which was being reported. These three numbers related to how flat, downhill and uphill power readings. There was a way to adjust these numbers somewhere either in the program or in program file. The best I was able to get was to get the RA to report 200 watts and 200 at the Quarq. As you went up or down, the RA would begin to get off target +/- depending on which way you went. Since most of my interval work lived around that number, that’s how I adjusted the numbers on the RA program. You could tweak the number to get RA to match a lower watt, or vice versa, but you couldnt get it to follow the same numbers of Quarq throughout the range. Its been a while since I ran the RA program since I used to run it on Parallels and no longer have Parallels installed.
Hope that helps.
Jon - dug through my archives of an old Yahoo group that was for RA owners and found this email which deals with those three numbers: P1, P2, and P3 are the adjustable numbers. I don’t know where they live in the software - maybe someone here remembers.
“I went ahead and tried to calibrate the unit using my Power Tap. The figures I entered for P3 were based on 40Km/Hr average readings but use judgement. I rode for several miles at that speed on a flat area but the wind varies as you all know so the average for P3 was 220 Watts.P1 and P2 are not so obvious, I am not sure what they mean when they say high slope low speed at 20km/hr so I rode up Mt. Baldy and it was hard for me to maintain 20km/hr going up a 15% grade but I was able to get a figure for about 10 second each so the average was 380 watts after several tries and legs on fire!.P2 I rode up a 6% grade and at 100 rpm cadence I was able to get an average of 165 Watts. BTW I am 1.66 meter tall 59kg is my weight and my FTP (Functional Threshold Power)is 180 Watts. Hope this helps!”
Jon, don’t know if you saw my post but yes was aware. RA does allow for recalibration but you need to have a real power meter to accurately dial it in and even then, most RA models aren’t going to simulate more resistance over an 8% grade. But, having the resistance change with slope is great, and for those of us with real power meters that zwift can use to calculate speed, problem solved. For those that don’t, and given the other ways to cheat (reported weight), why not just identify/categorize//segment those not using real power?
Great progress so far, thanks for looking into RA wired.
I spent quite soem time trying to calibrate my RA with a Stages powermeter using the REAL software and adjusting the P1, P2 and P3 parameters. Main findings:
It’s pretty accurate now around 200 to 280 Watts (Quite some effort to calibrate it, lots of Watt-pyramids up an down)
Above 300 Watts its way off, hard intverals at 450 Watts show only 350W on RA
Hysteresis is a bit of an issue. On a upward step test (20 W increment) it looks alright, but downslope it’s a bit off (about 10%)
Unit needs to warm up for 5-10min to get stable measurements.
Each RA is a bit different, sharing and applying the 3 P-factors is no guarantee that your unit shows the same watts than mine.
Hope this helps a bit. But keep in mind, the RA measures the power not via strain gages but via electric resistance. It was not designed to be as precise as a real powermeter
That’s great news Thanks Jon
So a BETA version of support will ship for USB RealTour, RealPower, and RealAxiom in the next game update (soon), but it will be for Windows PC’s only at the moment. The buttons on the hand control also let you choose your route at intersections, as well as skip workout sections and dial in your resistance in workout mode.
A few things to note:
It’s beta. We’ve tested it on 2 units at Zwift HQ but that is a far cry from hundreds of people testing it.
There’s no calibration for power. It’s an estimation using the formulas Elite has provided. I’d still recommend using a power meter with it if you want accurate power data.
It’s Windows only. The mac version seems like it’ll work eventually but it’s not working right now.
Wait for the update.
Congratulations on your efforts to support Zwift users!
When will the update be released?
Thanks for keeping us informed.
Do you need people to do some beta testing?
BETA support for the wired Elite trainers is live now for Windows users only. Just run Zwift and the latest version will automatically come down. OSX support will hopefully come soon.
Give it a try and let me know how it goes. The feedback from the early users of this feature will help us iron out any remaining issues.
PS, the wired Elite hand control will let you turn left/right at intersections in Watopia, as well as let you skip workout blocks and increase/decrease resistance within workouts.
Thanks so much, Jon. Will try it out asap and let you know.
i tried it this morning. it works great! i have the wired realaxiom. i only rode for about 10 minutes as i didn’t have a fan and i was getting hot. ha! it did feel much smoother than the ‘official’ realaxiom videos. as in the resistance seemed to gradually change. kind of like a sine wave vs. a square wave. the realaxiom videos being square waves.
Jealous! Keep us posted on OSX. Can’t wait to try it. Thanks for all your help Jon!
I tried today to drive 20 minutes after riding on the street in the snow. I have wired RA and coded polar HR, all worked and showed fine.
If this is true Watts I am very weak in fact, the difference with the elite video ~100 watts or more. It was very hard to go to 250 watts is about as hard as I’m going on Alp d’Huez at 370-390 watts per elite video.
Who has real power meter, tell me, is really such a big difference?
i don’t have a power meter but my wattage looked about right compared to what strava estimates on outdoor rides. within say 10-30 watts.
I rode my first training ride ever in Watopia this morning and nearly used up my free kilometers. Ride lasted 2h05. My visual live observation was that Zwift was showing about 60-70W lower power numbers than my Rotor power meter, both on the flat and hilly sections. After the ride I checked the averages for the whole ride and they differed 50W (power meter 135W and Zwift 85W). Being that slow naturally meant that everybody was passing me by on flats, uphills and downhills and I was among the slowest 5 of about 200 riders on the Watopia. My ride was easy steady endurance ride, so I cannot say what type of workouts others were doing. All in all, the power difference in my case is substantial.
However, the with the Elite Realaxiom software it was the opposite with power readings, Elite was about 50W above the power meter levels. I most probably have not tweaked the P1/P2/P3 as suggested for more accurate readings in RA, but I cannot remember that for sure. Anyway, my observations are similar with Hansmer Sergei’s about the rough 100W power difference between RA and Zwft.
I did not wear a compatible HR belt this morning, So, I cannot say anything about that.
I was in “just riding” -mode and the wired Elite hand control allowed me to choose which way to go on the intersection (the arrow button). I tried other buttons as well, but they did not seem to do anything.
Hi, pretty new to Zwift but have enjoyed it so far. I went ahead and bought a speed/cadence sensor and ant+ dongle waiting for this but have been hoping for the USB connection so I can get the change in feed back and was very excited to see it in Beta last night and went for a ride today. First the good news, it worked well changing my resistance, though it seamed maybe a bit less of a change than I would have expect and get on the video courses. I am ok with that part though, it gave me enough feedback to make me shift.
The watts seem WAY off however, even from what I get in RA itself. My first two rides in Zwift using the sensors and ant+ I setup a RA course I made that was just a 0% grade and ran that while I rode Zwift and compared results afterwards. Pretty good.
1/14 Zwift Watts: 217 RA Watts:235 Speed 20.9 HR: 159
1/17 Zwift Watts: 212 RA Watts:208 Speed 18.9 HR 157
This has Zwift at -18W and +4W and those speeds look pretty close to my real speed at that intensity. I was pretty impressed.
1/19 Zwift Watts 122 RA Watts:?? Speed 14.1 HR 85
This was today’s ride with Zwift controlling the RA. The HR is misleading as it kept going wonky. Would spend most of the time in the mid to high 150’s, but continually dropped to the 30-50 range for some reason which is what makes it so low, I assure you that intensity was just as hard. Cadence was also funny. It would just hang at about 91-94 even when I was not in that range, until the last ten minutes where it was in the 40’s even though I was still spinning in the 80-95 range with a few high speed test spins, no change.
So, at the same intensity, I was about 100W less and 6 mph slower. Something it not right there. I think the RA is giving accurate readings, so somewhere this is getting lost in translation. Is it still being pumped through a virtual curve somewhere?
Thanks for your time!
Just tried the RA wired beta with the power controlled by my stages power meter and it was awesome! Thanks so much Jon!!
Not sure I understand why those on this thread that are trying to tweak their RA power with a real power meter wouldn’t just use their real power meter?