I think everyone has answered your question. A real power meter works in a range of +/- 1.5% accuracy +/- 15% not useful information and how do you use that when you are actually riding?
If you just want to see some #'s use something like Strava it will give that to you free.
Go spend that $8 on something else.

it would be more fun and worthwhile to spend 7.99 on scrap wood, dousing it with gasoline then lighting it on fire.

You don't think that developers will race against time to integrate the $8 app that doesn't work into the $200 earbuds that nobody wants?

I believe the posters are doing what they consider fun.

It's not clear from OP's initial post whether the +/- 15% is the accuracy or the precision. Actually I suspect it is the accuracy. I imagine it doesn't move around a lot during use, but the approximations evident in entering your weight and "type" of bike as well as your personal drag coefficient could easily cause a relatively FIXED 15% error. Yes that would be a problem the first time you used it. But if you could possibly calibrate your output under standard conditions to a power meter as compared to the app (more or less), you could transfer that calibration to the app and determine the fixed error you are dealing with. Then just make the necessary correction to the results.

Self delusion?

Lots of people donated

It may not be accurate for instantaneous readings, but I think if the errors average out over a long ride, then it shouldn't be completely worthless.

Just assuming, I admit , but I don't think that is how it works. It is not precision error that averages out. It is accuracy. It gives the wrong answer. Besides power is an instantaneous measurement. What is the point of averaging it over time?

Are they selling the ibike for $8 now?

I'm trying to figure out why this thread is still going. I mean, from the point you read "$8" and "app", I was done. Completely useless.

oh wait it's an app? I missed that part. nevermind, it's s**t.

The description states that it's typically within 15% *over a given ride* - so, no, it's not averaging out errors over a long ride. Even if it did, frankly, I still don't see what value that would provide.

I just pee'd myself a little bit!
Pirk

I don't think it's even that good. Heck, it's not even as good as that PowerCal thing.

Mind you, only the first one is free. If you think you have a training increase the retest will cost you.

Neither does an SRM/Quarq/Powertap/Stages etc...

Power is the product of force and velocity which are both measured directly by power meters like the SRM/Quarq/Powertap/Stages/etc. Those power meter measure force via strain gauges. They multiply measured RPM by 2 pi times the strain gauges' distance to their center of rotation to get velocity. One multiplication returns instantaneous power.

This smart phone application can measure position (via GPS) and acceleration (via accelerometer) from which it can derive velocity.

The big forces on cyclists are the component of gravitational force parallel to the road (zero where it's flat), aerodynamic, and rolling resistance. While the smart phone knows what direction you're headed and where gravity is due to the accelerometer so it can arrive at that force based on your weight, it can't measure anything to do with aerodynamic forces or rolling resistance so it can't come up with an accurate total force and therefore power.

Given actual knowledge of CdA at a variety of angles, Crr, wind speed, and wind direction it could calculate those but it's limited to approximations which aren't enough.

They are making a direct measurement, not trying to guess.

So the general consensus is that +-15% is junk compared to normal powermeter accuracy. Is the innards of a powermeter really such that a $1000 price tag is warranted? One poster brought up that older PowerTaps can be found for $200 or so. Asides from obvious technical improvements that result in greater accuracy, is there really much significant difference between those older models at that price and the newer fancier models?

The older wired powertap hubs are very good. They do not have more recent "bells & whistles". Brand new Powertaps are well under $1000.
If you really want a power meter, then first decision is wheel based.or crank based.

Best to start a new thread if you decide to go for it.

Fixed - it doesn't even need to be compared to a real power meter.

a) ~$200 is the used price
b) those are the wired units - you need a sensor on the seat stay that is wired to the computer/head unit. This locks you in to that head unit and means that the wheel is not compatible with a Garmin or the like
c) significantly heavier than the new units

IMHO, sounds like it costs $8 more than it's worth. I'd guess most cyclists could estimate their power +/- 15% just based on perceived exertion. Calling a 30% range, inaccurate, is being generous.

I don't see how this could be remotely accurate. I recently did a ride with a small group and one of us had a power meter in the pedals. We went North at 23 to 23.5 mph and even leading this pace was not too hard thanks to a nice tail wind. While toward the back of the line, this was pretty easy.

BUT we then turned around into the wind and we all worked a lot hard to maintain about 18 mph. Even though Strava said I was using less power going South into the wind, I assure you I was working a lot harder. The actual power meter on the other bike said he was generating more then twice the wattage then Strava said I was using. But on the Northbound tailwind part of the ride, Strava said I was at twice the power output then the actual power meter was.
Ok, sure, were talking about two different riders on two different road bikes, but we were in the same pace line so it should have been pretty close, a lot closer then +/- 100%