Old 07-30-12, 02:22 PM
  #9  
jmX
Senior Member
 
jmX's Avatar
 
Join Date: Jul 2010
Location: Orange, CA
Posts: 2,201

Bikes: Roubaix / Shiv

Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 12 Post(s)
Likes: 0
Liked 2 Times in 1 Post
I used "virtual power" (non-) about 2 years ago using my ANT+ sensor and my own software, and then went on to buy a powertap. The numbers are close, but for hard short intervals virtual power isn't very useful. For anything a minute or longer, you're good to go.

Some shots of it from 2010 using "virtual power" aka just calculating using the standard equation everybody uses:





If somebody wanted to get really fancy they'd take the first derivative of speed and apply some modifier to base calculated power. I don't beleive anybody does that yet, altho I've played with it a bit here. Haven't had time to finish.

For instance, if you sped up from 10mph to 20mph in 10 seconds, the normal equations would simply convert speed to power, so it'd show you going from 120w to 220w (or whatever). However, if you really did that in 10 seconds, you would have been putting out probably 400-500 watts until your speed was a steady 20, then it'd go back to 220w to maintain speed. Without those power spikes, you'll never get an accurate TSS or any guesses at sprint numbers.

Last edited by jmX; 07-30-12 at 02:30 PM.
jmX is offline