Herb Chong wrote this in response to my speculation on how image
stabilization could be done:

"that's how it's done in Nikon and Canon and how IS binoculars from various
vendors work too. software implementations require sensors and image
processors far faster than can be reasonably powered right now."

I think you misunderstood the context of what I wrote. Canon's IS technology
uses an optical pendulum type system and is quite ingenious. I was
speculating on the practicality of stabilizing an image by use of
accelerometers to measure camera movement so servo motors could compensate
for the movement. I believe that method would be difficult to implement and
probably wouldn't work very well. This all started with the "why not
stabilize the sensor in the body?"

Tom Reese

Reply via email to