Login required to started new threads

Login required to post replies

Prev Next
Re: Front Page Shimano's Dura Ace R9100P Power Meter [Slowman] [ In reply to ]
Quote | Reply
Slowman wrote:
the truest thing you can say about me is that i don't have anything of value to write about this crank's accuracy.
You know, there are people around here who would be reasonably qualified to help you with this, were you so inclined. But I'd understand if you weren't.
Quote Reply
Re: Front Page Shimano's Dura Ace R9100P Power Meter [RChung] [ In reply to ]
Quote | Reply
RChung wrote:
Slowman wrote:
the truest thing you can say about me is that i don't have anything of value to write about this crank's accuracy.

You know, there are people around here who would be reasonably qualified to help you with this, were you so inclined. But I'd understand if you weren't.

yes, i'm blessed with some very smart people in this community on whom i could rely. i will certainly avail myself of the wisdom of these folks - perhaps you among them if that makes sense for you - and on a number of topics certain folks here have gotten a recent reachout from me. i've decided we should spread our wings a little, editorially, so what you're talking about is timely. but for more than one reason i was not ready to tackle accuracy in this article. i felt that cost, basic features, and most of all installation, was enough for this article.

Dan Empfield
aka Slowman
Quote Reply
Re: Front Page Shimano's Dura Ace R9100P Power Meter [Slowman] [ In reply to ]
Quote | Reply
Slowman wrote:
yes, we have a crank length issue, even if this crank is made down to 165mm, but those who need cranks shorter than that are edge-case, and probably not part of shimano's mission.

The last 4 cranksets I've sold were 2x160 (one an R7000), a 155 and a 145. For people all roughly average height for their gender, the 145s were for a road bike. Shimano are take off cranks for me, the only advantage is that they have some resale value, unlike certain placeholder brands. I use 162.5mm on my tri bike. Which is why my original comment is that I don't see a reason for triathletes to be interested - 170mm is a long crank when pushing the boundaries of closed hip angle.

One of those was a Rotor Aldhu with inspider power. That is a good system - heaps of modularity, rechargeable powermeter, goes down to 155, tension adjustment on NDS. Unproven as a powermeter at this point but it avoids the fundamental issues of a lot of other systems so that's a good start.

Slowman wrote:
we may be - what - a year away from a world where you can choose from a force axs PM, an ultegra PM, an FSA PM, as well as from giant and specialized, and these would all be candidates for OE.

OE powermeters on cranks are just going to be a headache as it will be increasing the cost of a part that needs to be changed for fit purposes. Properly modular like Rotor would be OK. As big a fan as I am of powermeters (won't coach without one and have 8 myself) I don't see that OE powermeters are a worthwhile goal - it's just a curiosity for the majority of bike riders.

The problem for me coaching athletes with a DA PM is that we don't know how accurate it is. Shane makes a convincing case that there is a fundamental design issue that means the expensive 9100 isn't really any better than a cheap Stages style system.

One of my athletes (have been fitting him for years, now coaching too) has completely worthless training history. Last year he had a 4iiii on DA that doubled the numbers on one side, then swapped to another 4iiii that underreads. His new team have Shimano (and insist on Shimano Cranks) but I'm left still not knowing whether his race bike is accurate. Which is a real problem when putting a CV out to world tour teams. 400 FTP is good, but every increment above that helps in getting a job. And I have to change his FTP in WKO per bike change while he still has the 4iiii on the training bike.

It will never happen but an industry standard dynamic calibration process would go a long way. Then we might actually be able to know the value of any particular PM.

Slowman wrote:
but i'm not a compleat idiot. (there's a 60s reference there.)

I had to look that up as my knowledge of the 60s is derived from the Monkees, Adam Wests Batman and Beatles documentaries.
Quote Reply
Re: Front Page Shimano's Dura Ace R9100P Power Meter [cyclenutnz] [ In reply to ]
Quote | Reply
cyclenutnz wrote:
Slowman wrote:
but i'm not a compleat idiot. (there's a 60s reference there.)


I had to look that up as my knowledge of the 60s is derived from the Monkees, Adam Wests Batman and Beatles documentaries.

if you were broke, investing in bulk in boxes of kraft macaroni when they went on sale for 25 cents per, then you had a used VW and you relied on how to keep your volkswagen alive, for the compleat idiot. and if you can find a good copy of that now, it does, literally, cost more than the 1964 bus with a blown engine that i and my buddies bought. and we used that book to rebuild the engine (pursuant to taking that bus on a 7,000 mile jaunt thru mexico and central america).

but actually the original reference was probably the compleat angler, which was a famous (at the time) but since largely forgotten then-bible of flyfishing from the 16th century.

Dan Empfield
aka Slowman
Quote Reply
Re: Front Page Shimano's Dura Ace R9100P Power Meter [Slowman] [ In reply to ]
Quote | Reply
I will never forget helping a friend rebuild his bus engine on a picnic table at Joshua Tree with a copy of that book open in front of us.

Might be a poll question here...

Tech writer/support on this here site. FIST school instructor and certified bike fitter. Formerly at Diamondback Bikes, LeMond Fitness, FSA, TiCycles, etc.
Coaching and bike fit - http://source-e.net/ Cyclocross blog - https://crosssports.net/ BJJ instruction - https://ballardbjj.com/
Quote Reply
Re: Front Page Shimano's Dura Ace R9100P Power Meter [fredly] [ In reply to ]
Quote | Reply
fredly wrote:
I will never forget helping a friend rebuild his bus engine on a picnic table at Joshua Tree with a copy of that book open in front of us. Might be a poll question here...

i think you're right, about the poll question. one infamous passage, and this from a 50 year old memory, so, i'll mung it up, but, "if the green light comes on, hop in the back seat with your girlfriend and right about the time you two're done, start the car back up and go, but if the red light comes on..."

Dan Empfield
aka Slowman
Quote Reply
Re: Front Page Shimano's Dura Ace R9100P Power Meter [Slowman] [ In reply to ]
Quote | Reply
I read the article and have a 4iii dual PM on DA 8000. The articles tells me my PM is not accurate. My question is against what? I mean the number is the number. so long as said number is repeatable then I should be fine. And its not like I am using these numbers to get a contract somewhere! However the difference to my wahoo numbers is significant when I run both at same time. Wahoo is approx. 15W lower at IM power of 225. That is significant. I suppose just need to know what number is what for each PM and train accordingly.

@rhyspencer
Quote Reply
Re: Front Page Shimano's Dura Ace R9100P Power Meter [rhys] [ In reply to ]
Quote | Reply
rhys wrote:
I read the article and have a 4iii dual PM on DA 8000. The articles tells me my PM is not accurate. My question is against what? I mean the number is the number. so long as said number is repeatable then I should be fine. And its not like I am using these numbers to get a contract somewhere! However the difference to my wahoo numbers is significant when I run both at same time. Wahoo is approx. 15W lower at IM power of 225. That is significant. I suppose just need to know what number is what for each PM and train accordingly.

i agree with your entire post. as long as the number is repeatable, what's the difference of it's off a bit? but we also rely on other sources for our power number and it would be convenient of our smart trainers and our power meters were congruent.

Dan Empfield
aka Slowman
Quote Reply
Re: Front Page Shimano's Dura Ace R9100P Power Meter [Slowman] [ In reply to ]
Quote | Reply
Yes. 4iiii on road bike. Wahoo indoors. Suppose could use 4iiii as sole source but Bluetooth connections to ITV and cadence built in Wahoo I just go with trainer. It’s get complicated come summer and I start riding my P2s on my Tri bike but come then I suppose I wont be on my Wahoo🤦‍♂️

Really first world problems. It’s just a lot of toys!

@rhyspencer
Quote Reply
Re: Front Page Shimano's Dura Ace R9100P Power Meter [rhys] [ In reply to ]
Quote | Reply
rhys wrote:
I read the article and have a 4iii dual PM on DA 8000. The articles tells me my PM is not accurate. My question is against what? I mean the number is the number. so long as said number is repeatable then I should be fine. And its not like I am using these numbers to get a contract somewhere! However the difference to my wahoo numbers is significant when I run both at same time. Wahoo is approx. 15W lower at IM power of 225. That is significant. I suppose just need to know what number is what for each PM and train accordingly.

There are lots of uses for power data, some of which don't require much accuracy or precision, some of which do.

Training FTP is one of the least demanding things you can do with a power meter -- that's why riders have been able to train effectively with just a wristwatch and a regular training route, or with a HRM, for decades. Some people will only care about their FTP, and they use their power data pretty much like a wristwatch and just try to improve their time; or they use it just like a HRM, and they just ride in their training zone. There's nothing wrong with that.

On the other hand, measuring drag is a highly-demanding use for a power meter, because drag varies nonlinearly with speed and power; that means in order to solve the nonlinear drag equation you need to know the *actual* amount of power. So, you can train FTP with a wristwatch or HRM, but you can't measure drag with just a wristwatch or a HRM. If you were trying to tease out small differences in drag between, say, two different helmets, or two different tires, you're going to need both high accuracy and high precision. Ironically, people used to try to measure drag with coast downs -- that's a case where the power is accurately and precisely known to be exactly zero.

As I said above, generally we buy power meters cuz we're hoping they'll help us answer questions or solve riddles. When the questions are easy or the riddles simple, you don't need much accuracy; when almost everything lines up, you use the data more for affirmation than information. It's when the questions are hard and the riddles are ball-breakers that you don't want to be asking "I wonder if my power meter is working right? I wonder which of these numbers can I trust?"
Quote Reply
Re: Front Page Shimano's Dura Ace R9100P Power Meter [RChung] [ In reply to ]
Quote | Reply
Way above my head mate! . Honestly i have power to keep me engaged. Thats it. I am far on the downside of performance at 49. 20lbs heavier & 10+yrs past my top racing. I just find power fun to look at climbing, pulling a race group in a bike race, staying low and aero and in control in an Ironman, or looking at after for what i knew was incredibly painful workouts. Its just a number. And a game. Played mostly outside!

That said, what helmet do u recommend? I still race a Giro TT from 2006!

@rhyspencer
Quote Reply
Re: Front Page Shimano's Dura Ace R9100P Power Meter [rhys] [ In reply to ]
Quote | Reply
rhys wrote:
Way above my head mate! . Honestly i have power to keep me engaged. Thats it. I am far on the downside of performance at 49. 20lbs heavier & 10+yrs past my top racing. I just find power fun to look at climbing, pulling a race group in a bike race, staying low and aero and in control in an Ironman, or looking at after for what i knew was incredibly painful workouts. Its just a number. And a game. Played mostly outside!

That said, what helmet do u recommend? I still race a Giro TT from 2006!

Hmmm. Sounds like you have exactly the right helmet for your needs.
Quote Reply
Re: Front Page Shimano's Dura Ace R9100P Power Meter [Slowman] [ In reply to ]
Quote | Reply
Interesting. I'll answer this in dot points as I've wasted way too much time on this specific issue, only to be shitlisted by most of these companies using these cranksets to make power meters. There's a reason I'm no longer reviewing other products these companies make and it's not my end. My door is always open....

- I test a LOT of power meters. I have a well established protocol that has proven itself over and over again. It's not perfect, but it works really well.
- ~18 months ago I was getting 'incorrect' data from ALL Shimano based L/R power meters. All of them.
- I could not (and still can not) use any Shimano based dual L/R power meter as a baseline to compare other meters to (smart trainers & power pedals) because of this.
- Diving deeper and deeper into this rabbit hole, I found something interesting with the right side readings.
- The explanation from the manufacturers was varied. From "Yeah, about that....ummm" to "We have no idea". They all know, or if they don't, they simply don't test to the level required for human riding. (constant torque testing is commonly used, humans aren't constant torque machines).
- I reached out to Keith Wakeham who's name is on the patent for this very design of power meter. Thankfully Keith was in a position to explain in great detail what the issue was. His video is all everyone needs to know on this issue. Kick back with a coffee and soak this in. It's brilliant and explains why there's no simple fix for this issue. https://www.youtube.com/watch?v=Zwt4Bx_FGHU
- So that was the issue explained. Finally.
- I set about putting all my data together and an overview of the issue. I sent this to all the named manufacturers two weeks prior to publishing. Two weeks later, I hit publish.
- I've since had a more Shimano based L/R cranks sent my way.. I've tested them, given the company the results (same same), and moved on. I can't be bothered at this time making more "Well, we knew this was a problem. Here's the data showing the problem" videos.
- As above, I don't talk much to the companies who make these meters anymore. I'm not spinning yarns they want to hear, but that's entirely on them. Make a better product. Simple. The 4iiii on the FSA crank was brilliant. It'd be interesting to test a Stages L/R on an XTR crank. But...

Back on topic: Shimano FC-R9100-P.

- As above. It has issues due to the base crank design.
- It wasn't designed as a power meter. It was an afterthought. That's proving to be a problem.
- I have reviewed this specific meter. Shimano said I was using the wrong firmware. (I wasn't. It was 4.0.7, "Update to make output value more accurate"). I updated the firmware anyway to 4.1.7 and retested, same issue. Shimano went quiet on me.
- There's multiple reviewers saying the same thing in regard to the data. DCR. Ben Delaney (over on Bike Radar). Myself. There's no motive here other than to talk on what we see from the data. And the data indicates issues.
- Are these units good enough to train with? Sure. They'll give readings good enough for most people. They just need to be aware of the shortcomings... and that what they've been sold might not be what they get.
- Are these units good enough for me to compare other meters to? No.

Other discussion points:

- "Why bang on about the inaccuracy of these and not left only, surely left only is less accurate". Because left only cranks ARE accurate for what they measure. Of course they won't represent your true effort if you're left/right wonky.. BUT... the left side is reading accurately.

- "Why do the pro peloton use the FC-R9100-P?". They're paid to. They also don't test power meters against other sources. They're professional cyclists. Maybe they know what's coming from Shimano so only have to spend 1/2 the season on these existing meters?

Take outs:

- I have no beef with Shimano. I love my Shimano Di2. It's brilliant because it's proven to work.
- I love the Favero Assioma, Quarq DZero, hell, even the random InPeak POWER Crank (Shimano Left only) that all do what they say they do. They work.
- I highly suspect Shimano will (or should) move to a spider based power meter for the pending DuraAce refresh. If they do, it's case closed on this issue. Write it off, move on.
- Pioneer exited this game a few weeks back. They were 100% reliant on Shimano cranks. What did they know that we don't?... Hmmm....

Anyhow. That's my braindump on this one. I just want to see power meters that work in the application I use them for. This doesn't.

Shane Miller - GPLama
YouTube | Web | Twitter | Instagram | Facebook | Strava
Quote Reply
Re: Front Page Shimano's Dura Ace R9100P Power Meter [gplama] [ In reply to ]
Quote | Reply
Shane: Thanks.

Do you remember when the Stages was first released? Ray asked me and Tom to help him set up test protocols and analyze the data. Part of Ray's remit was that we wouldn't look at the single-sided issue -- in retrospect, I think he was right: when I review academic papers we're told to review the paper that was written, not the paper that should have been written. Ray wanted us to help him review the product that Stages had made, not the product that Stages should've made. Anyway, what we found was that there were problems with the data from that first model. Stages denied there were any problems, said they had expensive calibration tools and that we didn't know how to evaluate power data. Then they tested and found out that we were right, delayed their release by six more weeks, and tried to release firmware updates to address our issues. The bottom line is that, in the end, consumers got a better product than if we hadn't done a careful analysis and identified their problems. *That was a good outcome.* I just wish they hadn't trashed us but you can't have everything.

That's what you've done. When Shimano gets around to improving their power meter, it'll be because you (and Ben Delaney, and Keith) raised a ruckus. That's pretty neat. I just wish they hadn't trashed you but you can't have everything.
Quote Reply
Re: Front Page Shimano's Dura Ace R9100P Power Meter [Slowman] [ In reply to ]
Quote | Reply
Dan....an important aspect of all reviews these days is some form of declaration by the reviewer about the source of the unit and payment. DCR does it all the time and has a clear policy.
So questions would be:
Is this a unit supplied by shimano for you to test?
Is this a paid review in anyway?
I just think in this era any review should make it clear from the start whether it is truely independent etc.....
This is not an accusation, but just an important component of reviews in today's world.
Quote Reply
Re: Front Page Shimano's Dura Ace R9100P Power Meter [Amnesia] [ In reply to ]
Quote | Reply
Amnesia wrote:
Dan....an important aspect of all reviews these days is some form of declaration by the reviewer about the source of the unit and payment. DCR does it all the time and has a clear policy.
So questions would be:
Is this a unit supplied by shimano for you to test?
Is this a paid review in anyway?
I just think in this era any review should make it clear from the start whether it is truely independent etc.....
This is not an accusation, but just an important component of reviews in today's world.


very reasonable post. no problem. we state quite clearly what our policy is, and often in duplicate. our partners are always listed at the bottom of every page. we don't engage in "paid media." we often, also, state it in the article just to be clear. we list shimano as a partner, tho this brand is not in for the long haul with us, month in, month out. shimano absolutely does occasionally run a campaign with us thru the year, they wrote us checks last year, and they almost certainly will this year.

therefore, i think it's safe to assume that we will cash a shimano check this year, and you should filter what we write through your cynic's goggles. that's likely to be the case with a lot of our reviews this year, as it appears we are (for whatever reason) in favor with the manufacturers in 2020. as gplama explained above, they can be fickle.

if this PM proves to still generate errant numbers one side, there won't be any hiding that here.

yes, the unit was supplied by shimano. i don't intend to keep it when i'm done. and i never sell product that is provided for review. my policy is that i prefer to pay for product that i review, but [rant mode] i don't feel the need for the manufacturer to profit from this. brands usually don't in my experience have an industry price just for those of us who're buying a lot of products only for testing purposes. [/rant]

Dan Empfield
aka Slowman
Last edited by: Slowman: Feb 22, 20 8:11
Quote Reply
Re: Front Page Shimano's Dura Ace R9100P Power Meter [gplama] [ In reply to ]
Quote | Reply
thanks for the information and the transparency. that's a very helpful overview, and it comports with what i've heard you say in your videos on crank-based PMs deployed on this crank platform.

Dan Empfield
aka Slowman
Quote Reply
Re: Front Page Shimano's Dura Ace R9100P Power Meter [Slowman] [ In reply to ]
Quote | Reply
Slowman wrote:
the questionable-futured pioneer,

It kills me that Pioneer is getting out of the business, though I admit I'm surprised they got into it. Love, love, love those cranks. If I didn't already own more than one, I'd pick up another one now. They're selling what remains all at a good discount.
Quote Reply
Re: Front Page Shimano's Dura Ace R9100P Power Meter [Slowman] [ In reply to ]
Quote | Reply
Slowman wrote:
but i parse between the PM proper, and the crankset. in any case

You can't for this PM. They're one and the same. Stages, et al, had an excuse (if a weak one). Shimano has none.
Quote Reply
Re: Front Page Shimano's Dura Ace R9100P Power Meter [trail] [ In reply to ]
Quote | Reply
trail wrote:
Slowman wrote:
but i parse between the PM proper, and the crankset. in any case


You can't for this PM. They're one and the same. Stages, et al, had an excuse (if a weak one). Shimano has none.

right. what i meant is that all these power meter makers - pioneer, stages, shimano - seem to come up with a similar problem: a bad measurement on the right side only. what they all have in common is this problem on this crank. if the labors of these PM makers were pointed toward another crank design, then they'd have less difficulty one assumes. hence my comment (if badly written) that the engineers at shimano may have been doing everything right, in the design of the PM. that shimano couldn't even get it to work while building into the crank is further evidence that there's something maddeningly odd about this crank.

it's a damned curious problem. i'll be in my workshop in a little while comparing power numbers across a number of power measuring devices including this one. i think it's a little harsh to say that shimano's engineers have no "excuse". i suspect they're beating themselves up over this. there have been firmware upgrades since ben delaney first wrote about this problem 2 years go, and I've asked for detail on each. gplama was generous enough to trade some PMs with me yesterday on my simple question of: why didn't they simply log the variance from actual and assign an algorithm to normalize?

Dan Empfield
aka Slowman
Quote Reply
Re: Front Page Shimano's Dura Ace R9100P Power Meter [Slowman] [ In reply to ]
Quote | Reply
Slowman wrote:
...
it's a damned curious problem. i'll be in my workshop in a little while comparing power numbers across a number of power measuring devices including this one. i think it's a little harsh to say that shimano's engineers have no "excuse". i suspect they're beating themselves up over this. there have been firmware upgrades since ben delaney first wrote about this problem 2 years go, and I've asked for detail on each. gplama was generous enough to trade some PMs with me yesterday on my simple question of: why didn't they simply log the variance from actual and assign an algorithm to normalize?

GPLama linked to a Youtube post by Keith Wakeham on his site. Wakeham is an engineer who's designed power meters before. I'm not an engineer, and I didn't fully follow his arguments, but I do recall him saying that the error can't be compensated for in software. It's just too complex, and it stems from the asymmetric design of the crank spider.
Quote Reply
Re: Front Page Shimano's Dura Ace R9100P Power Meter [weiwentg] [ In reply to ]
Quote | Reply
weiwentg wrote:
Slowman wrote:
...
it's a damned curious problem. i'll be in my workshop in a little while comparing power numbers across a number of power measuring devices including this one. i think it's a little harsh to say that shimano's engineers have no "excuse". i suspect they're beating themselves up over this. there have been firmware upgrades since ben delaney first wrote about this problem 2 years go, and I've asked for detail on each. gplama was generous enough to trade some PMs with me yesterday on my simple question of: why didn't they simply log the variance from actual and assign an algorithm to normalize?


GPLama linked to a Youtube post by Keith Wakeham on his site. Wakeham is an engineer who's designed power meters before. I'm not an engineer, and I didn't fully follow his arguments, but I do recall him saying that the error can't be compensated for in software. It's just too complex, and it stems from the asymmetric design of the crank spider.

yeah. GPLama explained this to me in some PMs we exchanged in the last few days. and, i watched keith's video. very interesting.

Dan Empfield
aka Slowman
Quote Reply

Prev Next