"The 33"-Road Bike Racing - iBike Sport Review

I did a quick review of the iBike Sport Gen3 and posted it on my blog; I figured some on here might find it interesting:

http://ft-atalay.blogspot.com/2011/04/ibike-sport-review.html

I am not sure exactly how accurate or up-to-date the Allen & Coggan race category table I refer to is, but I found it interesting that the numbers match my perceived abilities fairly well. There were some heated discussions a while back about the iBike's accuracy and/or usefulness, but based on my limited experience with it so far I think it will make for a very useful training tool (at least it seems consistent), especially given its price ($200). Let me know if there are any questions and I will do my best to answer.

P.S. I have never used the PT, SRM or Quarq power-meters before.

I don't even need to look. A device that measures everything but power is not worth my time and definitely not my money. Good luck with it. If it keeps you riding that's a good thing.

Sorry to be harsh, but if i'm spending $, the thing better be a measuring, not estimating device. Especially consider that you can get a real PM for a few hundred dollars more... The one that does aero measurements is $650.

This iBike sport model won't even allow you to d/l data, and the one that does is $500. I got my ptap wireless SL+ used for $560.

I don't even need to look. A device that measures everything but power is not worth my time and definitely not my money. Good luck with it. If it keeps you riding that's a good thing.

This argument is a red herring, but a "measuring power" categorization is up for debate. Also, don't discount the iAero's potential for drag measurement when used with a torque-based power meter.

This argument is a red herring, but a "measuring power" categorization is up for debate.

When the iBike actually measures Force I'll give it a second look.


Also, don't discount the iAero's potential for drag measurement when used with a torque-based power meter.

I actually like that aspect maybe coupled with a power meter. I hope Andy's Chung-On-A-Stick comes into existence. Cheers

When the iBike actually measures Force I'll give it a second look.

It has a very good accelerometer. F=ma you know.

The issue isn't whether the iBike "measures" power or not... it does as much as any other PM. The issue is how far it deviates from reality and how often. The way it determines power relies on an accurate measurement of several variables that are quite difficult to get right... especially if it's windy, you are going downhill (ie fast), or the road is rough. Errors tend to even out though, if it is properly calibrated and you use the post processing.

Another benefit of a real PM is that you can use it on the trainer.

The issue isn't whether the iBike "measures" power or not... it does as much as any other PM.

Incorrect.

It doesn't measure power. There are no strain gauges. It calculates power based on a series of assumptions.

For the price you'd be better served by a downloadable HR monitor.

I didn't mean this thread to be one along the lines of "iBike works just as good as PT/SRM/Quarq for a fraction of the price". I just wanted to share my experiences with it so far, and while it is not perfect, it seems to be very consistent under most conditions I have encountered so far. I think consistency is the key when trying to gauge improvement, so I am not too concerned if the readings are off a little bit as long as they are consistently so. I would actually love to put it next to a PT and see how the two compare (Velonews found them to compare very well, for what it is worth).

I bought the base version to give it a try and see if it was worth a damn, and so far I think it is worth the price. I am actually thinking of upgrading to the Pro version soon to have the ability to download and analyze the data. As it was pointed out, it gets somewhat expensive once you want to upgrade to the Pro version ($200 for the upgrade, on top of the $200 I paid for the base unit to begin with). But I think other riders/racers on a budget will find even the base version (which I think is reasonably priced, although other's idea of what is reasonable may be different) a very useful traning tool.

With that said, I should note that I haven't owned the unit long enough to assess its durability. Everything I just said will be meaningless if it doesn't hold up, and from my experience Velocomp's (iBike's maker) customer service was OK but not great. So I am hoping I won't have to use them again :)

It has a very good accelerometer. F=ma you know.

The issue isn't whether the iBike "measures" power or not... it does as much as any other PM.

Uhh, nope.

I didn't mean this thread to be one along the lines of "iBike works just as good as PT/SRM/Quarq for a fraction of the price". I just wanted to share my experiences with it so far, and while it is not perfect, it seems to be very consistent under most conditions I have encountered so far. I think consistency is the key when trying to gauge improvement, so I am not too concerned if the readings are off a little bit as long as they are consistently so. I would actually love to put it next to a PT and see how the two compare (Velonews found them to compare very well, for what it is worth).

How can you tell ? If riding the exact same course multiple times, but each with different combinations of wind conditions, barometric pressures and temperatures, how can you know if it's giving you an accurate "power" reading ?


Personally, I would always be wondering if the numbers were right, did I calibrate it properly this time ....
even if the overall numbers from a ride are similar to powermeter numbers, the short interval feedback from a pm can't be matched.
And since their top model is about the same price as a powertap, the choice is pretty clear to me.

It has a very good accelerometer. F=ma you know.

The issue isn't whether the iBike "measures" power or not... it does as much as any other PM. The issue is how far it deviates from reality and how often.
I completely agree with this.

The way it determines power relies on an accurate measurement of several variables that are quite difficult to get right... especially if it's windy, you are going downhill (ie fast), or the road is rough. Errors tend to even out though, if it is properly calibrated and you use the post processing.
Can you do post processing with an iBike Sport?

Incorrect.

It doesn't measure power. There are no strain gauges. It calculates power based on a series of assumptions.

For the price you'd be better served by a downloadable HR monitor.

Well said. The old 305 does great at measuring/recording what it supposed to without pretending to be a power meter. When/if the time comes where an athlete wants to train with power, powertap is the clear choice.

I hope Andy's Chung-On-A-Stick comes into existence.
Me too.

I ran one side-by-side with a Quarq and got MOSTLY matching results. The iBike lagged behind on short/sharp efforts and in some other situations. The big problem is that when it was off (due to weather or whatever) it was way off and often stopped working. Also the constant calibration to make it work, and the fact that it doesn't really work on the trainer are a pain. I don't think it's that bad, but I wouldn't buy one.

How can you tell ? If riding the exact same course multiple times, but each with different combinations of wind conditions, barometric pressures and temperatures, how can you know if it's giving you an accurate "power" reading ?


Personally, I would always be wondering if the numbers were right, did I calibrate it properly this time ....
even if the overall numbers from a ride are similar to powermeter numbers, the short interval feedback from a pm can't be matched.
And since their top model is about the same price as a powertap, the choice is pretty clear to me.

Good question; my opinion of the consistency of the unit is based on these:

- The numbers shown on the screen do not jump around significantly when I am doing a steady effort, including when I am riding over not so smooth roads. Obviously, an increase in effort, whether it is from riding into a headwind or some other factor, results in an increase in power.
- The average power from riding my favorite 20-mile course in what I would consider similar conditions has been consistent. I also went all-out once which resulted in a higher average power than usual. I should note that there are no stops so my times when I am riding solo are very consistent on this particular course.

Again, these do not prove the accuracy of the unit, by as I said before I am not worried about absolute accuracy as long as the conistency is there.

By the way, the calibration is done once and the unit goes through an auto-calibration, which takes about 10 seconds to complete, every time it is turned on. So there is no hassle there.

I will agree with you that the iBike is not going to be very good in tracking short-duration (I would say, 5 seconds or less) efforts because there is some delay between putting the power down and seeing the numbers jump.

Good question; my opinion of the consistency of the unit is based on these:

- The numbers shown on the screen do not jump around significantly when I am doing a steady effort, including when I am riding over not so smooth roads. Obviously, an increase in effort, whether it is from riding into a headwind or some other factor, results in an increase in power.
- The average power from riding my favorite 20-mile course in what I would consider similar conditions has been consistent. I also went all-out once which resulted in a higher average power than usual. I should note that there are no stops so my times when I am riding solo are very consistent on this particular course.

Again, these do not prove the accuracy of the unit, by as I said before I am not worried about absolute accuracy as long as the conistency is there.

By the way, the calibration is done once and the unit goes through an auto-calibration, which takes about 10 seconds to complete, every time it is turned on. So there is no hassle there.

I will agree with you that the iBike is not going to be very good in tracking short-duration (I would say, 5 seconds or less) efforts because there is some delay between putting the power down and seeing the numbers jump.

Sorry, but that's the one thing that should be jumping by +/- 20W due to the uneven resistance encountered in cycling, even if your speed is even, and the terrain looks "even." That the estimated wattage doesn't change dramatically is probably due to the fact that the unit isn't sampling very fast and that the computer compensates by running a rolling average with a long decay constant.

One of the first things you realize when you get a real powermeter is how much the number jumps, even with a 5" rolling average.

Sorry, but that's the one thing that should be jumping by +/- 20W due to the uneven resistance encountered in cycling, even if your speed is even, and the terrain looks "even." That the estimated wattage doesn't change dramatically is probably due to the fact that the unit isn't sampling very fast and that the computer compensates by running a rolling average with a long decay constant.

One of the first things you realize when you get a real powermeter is how much the number jumps, even with a 5" rolling average.

I said it doesn't jump around significantly; of course the numbers jump around a bit. I have no idea what the sampling rate is, but I would be surprised if it is much slower than a PT.

Incorrect. It doesn't measure power. There are no strain gauges. It calculates power based on a series of assumptions.

And other PMs calculate power based on a series of other assumptions. They both rely on physics.

Can you do post processing with an iBike Sport?

No... you have to be able to download.

Again, these do not prove the accuracy of the unit, by as I said before I am not worried about absolute accuracy as long as the conistency is there.

IME the reality of using the unit is the other way around. With post-processing I can get an AP for a ride that is within a couple % (probably). But in real time, it can vary a lot. Crosswinds, rough roads, and fast descents seem to be the worst case. Also, if there is significant drafting, or the wind shifts then the post-processing doesn't work so well either.

just read the blog. More disturbing than the thought of using the iBike as a power meter is the idea of doing a 20 minute test on a MUP.

I said it doesn't jump around significantly; of course the numbers jump around a bit. I have no idea what the sampling rate is, but I would be surprised if it is much slower than a PT.
You missed the point. Point being it SHOULD jump significantly if you look at the file. That's the first thing you realize when you get a powermeter: how much jump there is in the data. On a flat road on which i'm averaging 240W, i can expect to see variation between 200W to 300W. See below. Note how the speed really doesn't deviate much.

https://lh3.googleusercontent.com/_ZS79zroFGuE/TZyyxNKKloI/AAAAAAAAA-g/IdK7eoKpGi8/s800/untitled1.JPG

I ran one side-by-side with a Quarq and got MOSTLY matching results. The iBike lagged behind on short/sharp efforts and in some other situations. The big problem is that when it was off (due to weather or whatever) it was way off and often stopped working. Also the constant calibration to make it work, and the fact that it doesn't really work on the trainer are a pain. I don't think it's that bad, but I wouldn't buy one.

And it just so happens that training by power is most useful for efforts at and above VO2 max, for which HR and other indicators are not that reliable. I guess it would work great for triathletes though as their race is at tempo pace and the most intense stuff they do is at FTP range.

No... you have to be able to download.
Yeah, not being able to download is a deal killer. My first experience with PMs was a bit more than a decade ago when I had one of the original Power Tap Prologues on a couple month long loan. Like the iBike Sport, that model wasn't downloadable. People who know me can understand the deep irony of me having a PM that I couldn't get the data out of. However, even for less-OCD users than I, the ability to review, summarize, store, and share the data collected on rides is pretty important.

just read the blog. More disturbing than the thought of using the iBike as a power meter is the idea of doing a 20 minute test on a MUP.

Why is that?

You missed the point. Point being it SHOULD jump significantly if you look at the file. That's the first thing you realize when you get a powermeter: how much jump there is in the data. On a flat road on which i'm averaging 240W, i can expect to see variation between 200W to 300W. See below. Note how the speed really doesn't deviate much.


And it just so happens that training by power is most useful for efforts at and above VO2 max, for which HR and other indicators are not that reliable. I guess it would work great for triathletes though as their race is at tempo pace and the most intense stuff they do is at FTP range.

I will post an iBike data file on here for you all to look at when I am able to download the data; however, I imagine it would look similar to what you posted.

And I don't see why you couldn't train with the iBike for efforts at and above V02 max, but that is OK; I am glad you are happy with what you are using.

Why is that?

Because MUPs are not meant for high speed riding due to the mixed population of users: unpredictable kids, dog walkers, skaters ...

Because MUPs are not meant for high speed riding due to the mixed population of users: unpredictable kids, dog walkers, skaters ...

I guess I should have clarified; this one has a 2-mile long access path that leads to a 7-mile loop with a wide asphalt paved road. I do my riding once I get in the loop where there are essentially no kids or dog walkers. Everyone else does a good job staying to the right. Plus there is hardly anyone in the loop at 7:00 AM when I start riding.

But thanks for the concerns.

Because MUPs are not meant for high speed riding due to the mixed population of users: unpredictable kids, dog walkers, skaters ...

I train M-F on the 7 mile loop fa63 is referring to...while the early mornings and after work hours get a little crowded, throughout the middle of the day its quite deserted. Weekends are another story altogether. Its a wide paved path through the water wellfields. There's a 2 mile long access path to get to the 7 mile loop that IS usually more filled with children on bikes, walkers, etc. I typically use that as part of warm-up...however, once you get on the 7 mile loop its mainly cyclists, rollerbladers, and runners (the families turn around to go back to the parking lot and the runners often stick to the ground/grass rather than the asphalt). I'll do a 20 minute test in the middle of the day without seeing anything except snakes, deer, gopher tortuous, buzzards, and an occasional bobcat.

We are EXTREMELY lucky to have this area near us...riding on the roads around here is taking your life into your own hands.

And other PMs calculate power based on a series of other assumptions. They both rely on physics.

Pul-lease.

Getting you gum from a gumball machine relies on physics too but that doesn't make a gumball machine a power meter.

Strain gauges generate actual, repeatable hard data based on their quantifiable physical properties. The iBike doesn't. Hardly the same thing.

The OP did make a good point about being consistent. If the iBike generates consistent numbers for you under most conditions then it may be of some use. Those number will not be comparable to the numbers generated by other riders using more accurate systems, however.

If this suits your needs then it may be a viable option for you.

For me, I'll stick with the PM I have at the moment.

To me the problem with the Ibike stuff is the value proposition, i.e. price. The Ibike Sport is $249, but not dowenlaodable and jumps to $399 for wireless.

For me, and I would guess most people training with a Power meter, you really want to be able to download your data. So you have to go to the Ibike Pro, and now we're talking $459 to $599 which is way to close to a Powertap.

I've contemplated getting the Ibike Aero, (for the virtual wind tunnel with a strain based power meter, and to use on the tandem) but every time I choke on the price.

I train M-F on the 7 mile loop fa63 is referring to...

You should ride next to him so that you can glance over at his iBike and see what kind of power you're putting out.

You should ride next to him so that you can glance over at his iBike and see what kind of power you're putting out.

Correction - "estimated power"

Getting you gum from a gumball machine relies on physics too but that doesn't make a gumball machine a power meter.

Cute. All PMs rely on instruments and physics to translate measured quantities into power. The only issue with the iBike (or any of the others) is how accurate it is at determining those quantities, and how much the inaccuracies effect the power number. The iBike needs to determine slope and windspeed with great accuracy, and this is very difficult. Small errors in slope and windspeed can become large power errors... especially if you are going fast. If slope and windspeed could be determined with great accuracy, then the iBike would be a fine PM.

Cute. All PMs rely on instruments and physics to translate measured quantities into power. The only issue with the iBike (or any of the others) is how accurate it is at determining those quantities, and how much the inaccuracies effect the power number. The iBike needs to determine slope and windspeed with great accuracy, and this is very difficult. Small errors in slope and windspeed can become large power errors... especially if you are going fast. If slope and windspeed could be determined with great accuracy, then the iBike would be a fine PM.

If wishes were fishes....

Details matter. A strain gauge is subject to far less measurement uncertainty than the measurements the iBike makes.

To me, the ibike vs pm seems like the equvalent of telling the temperature by:

A) Using a thermometer (power meter) or
B) Putting a glass of water outside, very carefully measuring the rate of evaporation over a fixed time, based on wind speed, relative humidity, amount of sunlight, starting temp of the water ....
too much can go wrong too easily ... should we stop now ?

By rruff's review it seems the only time the ibike doesn't seem to work very well is in windy, cross winds, drafting, fast, downhill or rough roads. Sheesh it sounds like a deal cause I never have any of those conditions on a daily ride, even better as I never experience any drafting or going fast in a race. :rolleyes:

A strain gauge is subject to far less measurement uncertainty than the measurements the iBike makes.

Isn't that what I said?

A strain gauge is subject to far less measurement uncertainty than the measurements the iBike makes.

Isn't that what I said?

Me too.

I figured you would. :)

@rruff
I have no idea, but I guess it was worth repeating... You seem to be hedging around the subject of accuracy by introducing a lot of wishful thinking.

Not at all. I'm just pointing out that the concept is not inherently flawed, and that it is possible (at least theoretically) to build an accurate PM that works like the iBike. You just need to do a very good job of measuring slope and airspeed accurately.

Pul-lease.

Getting you gum from a gumball machine relies on physics too but that doesn't make a gumball machine a power meter.

Strain gauges generate actual, repeatable hard data based on their quantifiable physical properties. The iBike doesn't. Hardly the same thing.

The OP did make a good point about being consistent. If the iBike generates consistent numbers for you under most conditions then it may be of some use. Those number will not be comparable to the numbers generated by other riders using more accurate systems, however.

If this suits your needs then it may be a viable option for you.

For me, I'll stick with the PM I have at the moment.

I find it interesting that this point keeps being brought up (and I am assuming you and others that make this statement haven't used an iBike before). What do you guys think about the evidence out there, some even from credible sources like VeloNews, that suggests the numbers between the iBike and the other systems are very close? Maybe the unit really does a decent job of measuring wind speed and gradient in most conditions, and the assumptions in rolling and aerodynamic resistance are pretty good as well?

What do you guys think about the evidence out there, some even from credible sources like VeloNews

Just because a magazine says something, doesn't make it true. Trust me, I used to edit one and have written for many.

The iBike is better than nothing. It's not in the same conversation as strain gauge based units or even the (may god have mercy on it's soul) Ergomo. It's wholly inaccurate under a fairly wide set of circumstances. It's accurate under others. IF the speedometer on your car operated like this you'd get a lot of tickets.

To me, it wouldn't be so much about accuracy but consistency. If it's your only source of information and you derived your FTP with it and trained as such with it, as long as it's consistent, it makes no difference if it's accurate compared to whatever set of standards you want to compare it against. However the fact that it's ability to give consistent results is questionable, compromises it's ability to do the job for which it's designed. Therefore rendering it virtually useless. Second of all, it's ability to give immediate feedback also compromises it's usefulness. When doing tempo stuff and longer rides, you can effectively use HR, it's the intervals and such that the PM is really useful where HR is delayed and apparently so is the iBike.

Just because a magazine says something, doesn't make it true. Trust me, I used to edit one and have written for many.


Of course, but does that mean that there isn't any truth to what they reported? Here is the link, I will everyone make up their own mind:

Link (http://www.ibikesports.com/documents/VN_2010Feb.pdf)

Here is the link to an article Florida Racing Magazine published:

Link (http://www.ibikeforum.com/viewtopic.php?f=39&t=1919&start=0&st=0&sk=t&sd=a)

Once again, I just wanted to share my experiences; hopefully it will help someone one way or the other. I obviously like the unit, but not having used a strain-gage based PM, maybe my expectations/standards are low for what a PM should do...

Of course, but does that mean that there isn't any truth to what they reported?

Well, this is true and really an unintentional slam of the unit:

"But since roll-down tests are done in your standard riding position these fixed numbers produce accurate results most of the time"

Lay folks would be pretty shocked at what seemingly subtle changes in position can produce in CdA; pencil that out to watts and you'd realize what a big +/- reality you end up with.

I find it interesting that this point keeps being brought up (and I am assuming you and others that make this statement haven't used an iBike before). What do you guys think about the evidence out there, some even from credible sources like VeloNews, that suggests the numbers between the iBike and the other systems are very close?

Looking at averages and saying "they're very close most of the time" is less important than knowing the specific conditions under which they're not. Ron is way more credible about that than a review you read in Velonews -- and, as an aside, I'm not sure I'd be pointing out a review by the salesman who sells ad space for a cycling magazine. Just sayin'.

[Edit:] BTW, I don't have a dog in this fight. If you like your iBike Sport, that's fine by me. For many people something like the iBike Sport is a gateway drug and it sounds like you're interested in moving up to the hard stuff. Maybe Velocomp should make the first one free.

Lay folks would be pretty shocked at what seemingly subtle changes in position can produce in CdA; pencil that out to watts and you'd realize what a big +/- reality you end up with.

IMO that is more of a feature than a bug... because physiological stress also varies with position. Calibrate it for your "fast" riding position and it will automatically penalize your watts a little when you are less aero.

If that was the only source of error, I wouldn't be bothered at all.

I've had one for almost 3 years now... paid ~$325 for wired unit on sale, then upgraded to the newest generation when it came out. I use it to track my AP and get a TSS and TSB (NP is usually too high, so I have to guess on that). Many of the roads around here are rough enough to effect the slope determination, plus the wind and hilly terrain make it a very challenging environment. +-200W errors on a fast descent are common. I don't worry about the numbers til I get home, when I'm usually able to correct the data pretty well.

It's a very nice bike computer... but limited as a PM.

.... However the fact that it's ability to give consistent results is questionable, compromises it's ability to do the job for which it's designed. Therefore rendering it virtually useless. Second of all, it's ability to give immediate feedback also compromises it's usefulness.....

When doing tempo stuff and longer rides, you can effectively use HR, it's the intervals and such that the PM is really useful where HR is delayed and apparently so is the iBike.

Bad assumption. Not true at all. I've been using the iBike Sport since 2009 and it works fine to this day.

The delay you mention is just fractions of a second.


... Heart Rate... lol


I collect the data using the Garmin 500 that I need anyway and download and analyze the power data just like any other power meter.

Another great benefit to this thing is that I don't have to use a heavy wheel or ever change batteries in the hub unit (like the PT), and I can swap it around to my TT, Track, or various road /training bikes without any hassle.