New location

Come on over to my new site:

Going to be posting regularly there.

Thursday, December 2, 2010

video/picture bike fit?

The new version of the Retul software is going to come with video capability so that after capture periods with the infrared portion the fitter can show the client pictures and video of them pedaling to correlate the numbers that the fitter has from the infrared, and give it a second visual aspect.

Why did it take so long to incorporate video?  The main reason is that pictures and video are a poor means of measuring in a bike fit.  Even with expensive video systems, like the Dartfish video software, it comes down to the user placing points of rotation on a picture, and making measurements based on that.  Even with a large LCD display the picture of the cyclist might be 12-15 inches tall, at most, and placing an anchor point at the hip for instance, it would be easy to have it be "off" by a large margin. 

In fact, infrared and video systems have been tested side by side, and the video's margin of error consistently comes out to at least 10%.  Infrared measuring on the Retul usually runs at a margin of error of 0.2 mm (two-tenths of one millimeter).

Is it a big deal to be that accurate?  Even clients who are already pretty happy on their bike, benefit from the infrared because small refinements have shown large gains in pedaling efficiency and power.  10% error is too much to significantly impact aberrant mechanics and fix pain patterns on the bike, but it may help with "showing" clients very gross changes and malalignments.

So I guess it was a pretty good idea that Retul decided to use it this way.