oooh... my brain.... I'm going to take a wild guess, mostly because I want somebody to jump in and correct me.
The sensor has to move more, the longer the lens. At some point it's going to be too much. I don't think the shutter speed has anything to do with it, since even on a slow exposure, the sensor has to be moving when the exposure starts. But, if the sensor has to move more, it will have to move faster too. And that could be a problem also. I wouldn't doubt.. and I guessing again... That head movement would be harder to correct than camera shake. The logic behind this reasoning is you have to put in a focal length. So, the SR must be sensing the movement of the camera and compensating. So I would say, plaster the camera against your face, and keep your head still for the best results from SR. That way, any axial movement can be sensed by the camera.
Originally posted by kevbirder You may not think that you need shake reduction at high shutter speeds, but I hand hold a 500mm mirror lens on a K10D. On the basis that it's equivalent to a 750mm lens, without SR I would be thinking of 1/750th minimum, preferably 1/1000th shutter speed.
From much of what I have read, using SR seems to be little understood, but I will ask anyway if anyone knows if SR is effective at high shutter speeds. I can easily get camera shake at 1/750th if I am not really careful, but I am not convinced that the SR has much effect on this. I had one shot that was completely blurred (motion not focus) with SR on taken at 1/1000th. I am wondering if this shot was taken before the little hand appeared though and might be blur induced by the SR system.
Most people seem to test SR with slow shutter speeds and standard lenses and maybe SR works great at 1/15th, but does it react fast enough to be useful at fast shutter speeds? I have tested but my tests have been inconclusive.