Bought mine from these guys on eBay. Pretty hard to beat the price. Mine was the Podium with rechargeable battery but some people prefer coin cell.
FWIW I had a Stages Gen3 previously and they both work well. I only switched because I got a new bike with Shimano cranks.
I’d like to add another thing in favor for 4iiii. It’s scale factor. Stages lack the possibility to change the slope and that means it’s worthless for all of us that have perfectly calibrated trainers at home. I own both Stages, 4iiii and Favero Assioma Duo and I’d probably get Quarq if I had an all Eagle-setup on any bike. If not, then 4iiii.
Another happy 4iiii (precision, 105) user here. Got it on sale last fall during the DCRainmaker 20% off deal. No real complaints.
I own the spesh branded 4iii and battery works better than stages it seems. But no complaints on either. I bought both from friends so price wasn’t an issue. 4iii is on my road bike and the stages is on my cx if that makes a difference.
I have purchased a 4iiii power meter the beginning of this year, and for the most part it is great. I have only run into two problems: despite following the installation instructions to the letter (installation is really easy, but you need a torque wrench and a special tool to unscrew the plastic nut on the crankset), my power meter read fantastically high (about 430 W average on a mellow ride), despite calibrating it before every ride. Then I went on a trip for about a month and when I came back the power numbers were accurate. I am not sure what happened.
The other niggle I have is that the battery indicator is not reliable: earlier this month I went to participate in my first race. So the day before I checked my bike and the 4iii’s battery was at 70 %. When I warmed up, the battery was all of a sudden at 1 %. Fortunately, a riding buddy had a replacement, but this is something to consider. Now I will just make sure to have a spare.
For the price, I think the 4iii is hard to beat. I would have paid 3-5 times as much in Japan for a crank-based, single-sided power meter from Rotor. (Of course, I would have gotten a free crank, but on the other hand, I already have one.) So if you are on a budget, the 4iiii is a great choice.
4iii had some firmware upgrade issues recently. Other than that, I’ve been happy with it. They also were the only option for my cranks.
Just out of interest, What constitutes a “perfectly calibrated trainer”? How do you know it’s perfect? Against what?
Back on topic, I’ve had both 4iii and stages. Both are great, 4iii is cheaper. Go for 4iiii
the key is "Metrological traceability*
Metrological traceability is defined as “the property of a measurement result whereby the result can be related to a reference through a documented unbroken chain of calibrations, each contributing to the measurement uncertainty”.
https://www.wmo.int/pages/prog/www/IMOP/publications/Flyers/Traceability_flyer.pdf
I also just bought a 4iii and will calibrate it against my powertap hub. The hub is calibrated by the powertap service. And there instruments are, hopefully, calibrated as well.
But the main point for me is that the measured data are consistent between 4iii and hub.
Well. In my case the perfectly calibrated trainer is the one that I use with all my indoor training. And the fact that the Neo is highly accurate both for DC Rainmaker, Shane Miller and a bunch of reviewers pin it as go-to-equipment for power readings is sufficient for me to classify it as perfectly calibrated. 
I could of course be wrong but since I compare all my other powermeters to the Neo. But other than Tacx stating that it’s perfectly calibrated and doesn’t need re-calibration I can’t say for sure that it’s perfectly calibrated.
And yet, Shane’s own testing shows that the Neo can suffer from accuracy issues when used at higher gearing/flywheel speeds (I might be able to dig up the FB post, but he shared some interesting data and observations). There are real cases where the Neo is not the most accurate power data, and not “perfect”.
This appears to be a common trait with most wheel-off trainers as a result of how they measure/estimate power. Notably, most of them do not feature an actual strain gauge that is present in a regular power meter. They use advanced measurement of the trainer speed and electrical properties measured in the system to make estimates (quite accurate in most cases, but estimates just the same) of the power values.
The exception is the Elite Drivo & Direto, that use and Optical Torque Sensor (OTS) that is a lot closer to an actual strain gauge. It measures the time gap between two sides of a shaft that have these segmented wheels installed. The shaft deflects via torsion when under load, and that can be very accurately measured to calculate power. It is more accurate in most cases when compared to the other trainers, especially on the Drivo that uses 2 lasers for a very accurate measurement of the torque displacement.
I’ve been using a 4iii for nearly 3 years inside and outside faultlessly. And this…
I’d like to add another thing in favor for 4iiii. It’s scale factor. is a very good reason to get one - you can adjust this to equalise the 4iiii output to a another source - a Direto in my case - so that training indoors and outdoors shows the same watts.
Love my 4iiiii
Okay. Thanks for digging it up. Not a real concern for me though because I always run the Neo in ERG and 34/17T.
I have drop outs with my 4iiii pm and my Fenix 5 worn on my left wrist. when the watch is on my left wrist, or I’m using mt 520 on my stem is is fine. I have read that the Fenix 5 doesn’t have a great ant+ antenna, so it could be the watch rather than the 4iiii, but I don’t recall the stages G2 it replaced having the same problem, so the 4iiii might have a slightly weaker signal. Haven’t tried BTLE though.
Apart from that my 4iiii is solid. I am considering getting another for my other mtb
I tested 4iiii against a powertap hub and the results are disappointing. In short here are results of 3x8 min intervals. I calibrated the 4iiii just before the series:
PT,4iiii
316, 300
318, 304
323, 297
For em this means the 4iiii is useless for pacing 
Fortunately I locate in EU can send the 4iiii back.
That to me just shows that one PM is about 15-20W different from other…which isn’t unusual. Both power meters had a range of 7w between highest and lowest interval. All within usual % error. Data looks reasonable to me, just 4iiii records around 15 watts lower which may be due to leg imbalance…maybe slightly more so on last interval as you got tired.
.
In the first interval ratio is roundly 1.05 , in the last 1.09. This mean the fluctuation of the 4iiii between 1. and 3. interval 11 satts. If I would try to ride 11 watts harder , this would lead to failure. This is simply not sustainable.
I have no stages to make the same experiment. Possible that the difference in the last in due to increasing imbalance caused by fatigue. In this case this problem is expected to occur with all left side only PM. But this is speculation.
Also note there’s a very real mechanical difference between how the hub is receiving power through the drivetrain and how the 4iiii is through the crank.
You could easily adjust the output of the 4iiii in the app to compensate an provide equal results.
This was my hope and thats why i bought one.
But this is not possible, because the ratio changes. The situation is even worse then the intervals show. During the warm up the 4iiii show even higher power then the powertap. And a the end of the ride on a climb out-of saddle was exactly the same.
Overall is unpredictable how the 4iiii differ from the PT.
