Go Back  Bike Forums > The Racer's Forum > "The 33"-Road Bike Racing
Reload this Page >

Anyone actually use Aerolab in Golden Cheetah?

"The 33"-Road Bike Racing We set this forum up for our members to discuss their experiences in either pro or amateur racing, whether they are the big races, or even the small backyard races. Don't forget to update all the members with your own race results.

Anyone actually use Aerolab in Golden Cheetah?

Old 02-07-12, 08:38 AM
  #1  
stedalus
Senior Member
Thread Starter
 
Join Date: Sep 2008
Posts: 598
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Anyone actually use Aerolab in Golden Cheetah?

I have been working on adding a feature to Golden Cheetah's Aerolab that will automatically adjust CdA, Crr, etc to best fit the VE profile, instead of having to fiddle with sliders until it "looks right." I've been giving it intervals from my rides, and it works well in the sense that it can usually find reasonable-looking values. However, since these are just from random rides, they have a pretty wide spread (not surprising).

So, I am wondering if anyone has a set of ride data that was controlled enough to see, say, differences in CdA or Crr vs changing position or tire pressure using manually fitting sliders. I'd like to compare the manually found values to what my fitting routine says. I'm also curious if the whole virtual elevation concept works well enough to measure CdA/Crr with any precision.
stedalus is offline  
Old 02-07-12, 09:42 AM
  #2  
ericm979
Senior Member
 
ericm979's Avatar
 
Join Date: May 2007
Location: Santa Cruz Mountains
Posts: 6,169
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Cool! Thanks for working on that. It sounds useful, especially for idiots like me.

Post your work to the golden cheetah list. There's people there who have used it and have clean data to run against your code. Also, R. Chung is sometimes on the list.
ericm979 is offline  
Old 02-07-12, 09:50 AM
  #3  
efficiency
Does Not Exist
 
Join Date: Oct 2006
Location: Voltairia
Posts: 2,094
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
<geek>
What's your algorithm?
</geek>
efficiency is offline  
Old 02-07-12, 09:59 AM
  #4  
Hida Yanra
VeloSIRraptor
 
Join Date: Mar 2008
Location: Deschutes
Posts: 4,581
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 21 Post(s)
Shoot me a PM - I can potentially go to our outdoor velodrome in the next week and get you a pretty narrow spread of data.
Hida Yanra is offline  
Old 02-07-12, 10:08 AM
  #5  
stedalus
Senior Member
Thread Starter
 
Join Date: Sep 2008
Posts: 598
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
ericm: Thanks. The mailing list is a probably better place to start. Once the patch is fit for public consumption I'll post it there.

efficiency: I am using a Nelder-Mead minimizer with sum of squares as the cost function. As long as the initial guess is ballpark it can usually find a pretty good fit. It can fit all the parameters, but usually it's best to keep mass, eta, and maybe rho fixed. The minimizer can select which parameters to fit, but I need to put in UI controls for that.

I originally tried a linear least squares (this means you have to take m and rho as given). While this was a lot more stable, it often found crazy values (negative CdA, etc). I don't know of a way to do linear least squares with constraints, but if such a beast exists, I think it might work better.

At the moment I am using the GNU Scientific Library implementations, which adds another library dependency, but the relevant routines are fairly short and could be extracted and put into the GC source tree.
stedalus is offline  
Old 02-07-12, 10:39 AM
  #6  
stedalus
Senior Member
Thread Starter
 
Join Date: Sep 2008
Posts: 598
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Originally Posted by Hida Yanra View Post
Shoot me a PM - I can potentially go to our outdoor velodrome in the next week and get you a pretty narrow spread of data.
Let me see what I can get from the GC mailing list before I send someone out into the cold to do test laps. Thanks for the offer.
stedalus is offline  
Old 02-07-12, 11:26 AM
  #7  
HMF
SkinnyStrong
 
HMF's Avatar
 
Join Date: May 2010
Location: Austin, Tejas
Posts: 1,169
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
I would use it if it were easier. I could never get it to look right though.
HMF is offline  
Old 02-07-12, 11:30 AM
  #8  
echappist
fuggitivo solitario
 
Join Date: Jun 2009
Location: Northern NJ
Posts: 9,060
Mentioned: 31 Post(s)
Tagged: 0 Thread(s)
Quoted: 216 Post(s)
wind makes it a lot harder to make it look right, and you also need correct values for air density, which this site estimates

http://www.gribble.org/cycling/power_v_speed.html
echappist is offline  
Old 02-07-12, 11:33 AM
  #9  
Hida Yanra
VeloSIRraptor
 
Join Date: Mar 2008
Location: Deschutes
Posts: 4,581
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 21 Post(s)
Originally Posted by stedalus View Post
Let me see what I can get from the GC mailing list before I send someone out into the cold to do test laps. Thanks for the offer.
I live in Seattle... it isn't cold

but sure, just PM me if necessary and we can work out a protocol.
Hida Yanra is offline  
Old 02-07-12, 12:15 PM
  #10  
plantrob
a runner no more
 
plantrob's Avatar
 
Join Date: Oct 2010
Location: Allentown, Pennsylvania
Posts: 261

Bikes: Raleigh Competition C6 Fusion

Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Originally Posted by echappist View Post
wind makes it a lot harder to make it look right
Yeah. And it's always windy here. Would it be possible to assume a constant wind speed and direction (as additional user inputs), infer direction of travel from the GPS data, and apply some kind of additional correction?
plantrob is offline  
Old 02-07-12, 03:07 PM
  #11  
echappist
fuggitivo solitario
 
Join Date: Jun 2009
Location: Northern NJ
Posts: 9,060
Mentioned: 31 Post(s)
Tagged: 0 Thread(s)
Quoted: 216 Post(s)
don't you have trexeltown? conditions are quite a bit calmer in the summer
echappist is offline  
Old 02-07-12, 04:20 PM
  #12  
kleinboogie
Senior Member
 
Join Date: Mar 2009
Location: San Diego, CA
Posts: 2,606
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
I have a couple time trials where I tried hard to get it to look right. I used a road bike, aero helmet, skin suit, shoe covers on the first one. Added a wheel cover and worked on my position on the 2nd one and did see a significant change which coincided with a 1mph gain over 20km. I may not have all the weather data but I think I took a picture of the Aerolab screen.

Originally Posted by HMF View Post
I would use it if it were easier. I could never get it to look right though.
This. My fat hands and trackball make it almost impossible to slide in the numbers. It would be nice if it remembered what I entered for each ride so I don't have to take screenshot and re-slide all the values. Great tool. If you can make the UI easier please do. Cheers

Last edited by kleinboogie; 02-07-12 at 04:26 PM.
kleinboogie is offline  
Old 02-07-12, 06:40 PM
  #13  
RChung
Senior Member
 
Join Date: Sep 2009
Posts: 1,447
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 264 Post(s)
Originally Posted by stedalus View Post
I have been working on adding a feature to Golden Cheetah's Aerolab that will automatically adjust CdA, Crr, etc to best fit the VE profile, instead of having to fiddle with sliders until it "looks right."
There are certainly ways to solve for the (CdA,Crr) pairs analytically, and sometimes to find a unique solution. That depends more on the protocol than on the estimation algorithm.

I'm also curious if the whole virtual elevation concept works well enough to measure CdA/Crr with any precision.
Yes, but you need to be pretty careful.
RChung is offline  
Old 02-07-12, 07:22 PM
  #14  
climber7
Senior Member
 
Join Date: Feb 2011
Location: New England
Posts: 797

Bikes: 2010 Jamis Xenith Comp

Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
i can sometimes get it to fit for a small portion of a ride but never for the whole thing. i'm guessing that's largely a result of wind, but do any of you guys actually get good matches for entire rides just by adjusting the Crr/CdA sliders?
climber7 is offline  
Old 02-07-12, 07:37 PM
  #15  
RChung
Senior Member
 
Join Date: Sep 2009
Posts: 1,447
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 264 Post(s)
Originally Posted by climber7 View Post
i can sometimes get it to fit for a small portion of a ride but never for the whole thing. i'm guessing that's largely a result of wind, but do any of you guys actually get good matches for entire rides just by adjusting the Crr/CdA sliders?
Depends what you mean by "good." I can quickly scroll through a bunch of rides and often tell from the Aerolab profile which route I was riding without looking at the notes page.

However, remember that VE doesn't take into account brake usage, either. It's best for segments of rides where you know you held your position, you didn't use your brakes, and there wasn't any wind. That's one of the reasons why the v3.0 implementation is better -- you can zoom into a part of the ride and ignore the parts where you know those conditions don't apply.
RChung is offline  
Old 02-07-12, 10:32 PM
  #16  
Fat Boy
Wheelsuck
 
Fat Boy's Avatar
 
Join Date: Jun 2007
Posts: 6,158
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Originally Posted by climber7 View Post
i can sometimes get it to fit for a small portion of a ride but never for the whole thing. i'm guessing that's largely a result of wind, but do any of you guys actually get good matches for entire rides just by adjusting the Crr/CdA sliders?
I generally attribute this to my Garmin's altitude going wonky. I always have a completely different altitude at home when I'm going that when I'm returning. It's annoying. They should have an algorithm that returns you to the same altitude when you're in the same GPS spot. I can ride on a MUP and back and it will be different each way.

So I end up trying to match up the virtual elevation in sections, and it works reasonably well.
Fat Boy is offline  
Old 02-07-12, 11:14 PM
  #17  
davids0507
Senior Member
 
Join Date: Jun 2008
Posts: 215
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
I use it mostly to figure out the climbing elevation.

But I also like using it to explain to people what riding in the peloton is like. "Riding with the group is like descending at a hundred feet per minute!"
davids0507 is offline  

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off


Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service

Copyright 2018 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.