Originally Posted by Bean
Will there be ABSOLUTELY no handshake blur when photos are taken handheld? I feel that the above statement about no handshake blur is flawed and thus render your theroy and explanation of the guideline is 1/actual focal length flawed.
Using your theroy that there is no handshake blur to begin with, why do we still need the 1/focal length guideline? 2mm or 20000mm, there will be no handshake blur to be magnified, unless we are talking about placing the camera on a tripod.
And just like having higher magnification (of handshake) through higher focal length, smaller sensor also magnifiy the image (and handshake). Thus, I stand on the guideline of 1/equiv. focal length.
Peace. Just a healthy discussion.
|
1) The first point is that if you take any of your own pictures which have no handshake blur (i.e. means no handshake blur is recorded) and magnify it, do handshake blur suddenly appears out of nowhere even though there is no handshake blur in the original picture but handshake blur appears just because you magnify it?
The answer is obviously, no.
The first point is that if there is no handshake blur recorded on a photo to begin with, there won't be handshake blur present in the photo even if it's magnified 100 times.
2) So the natural question is when is handshake blur recorded?
Obviously again, handshake blur is recorded by the sensor if the image move too much on the image sensor because of too much handshake when the shutter is open.
Common sense tells us that how much the image moves on the image sensor depends on the physical handshake as well as the focal length (which magnifies the angle shift) and of course, the shutter speed.
Does the size of the back screen (i.e. image sensor) affects the image shake? Again, common sense tells that the answer is No because it only affects what is included in the frame.
Put it in another way : With steady hands at 72mm focal length at 1/72 shutter speed, a picture is taken. If there is no handshake blur recorded on a 36mm x 24mm image screen behind, do you expect handshake blur to be recorded just because you change your image screen size to a smaller 22mm x 15mm? The answer is obviously "No" because it is exactly the same image as the one on the bigger sensor but only the Field of view is changed (i.e. it is now a cropped picture of the bigger sensor).
===========
Taking (1) and (2) above together :
Now a picture is taken at 1/72 with a focal length of 72mm on a full frame (i.e. 35mm) image sensor camera and found to have no handshake blur recorded. (1/focal length guideline is followed).
The original image is 36mm x 24mm (full frame sensor size) = 864mm^2
Now crop this very same image to 22mm x 15mm (APS sensor size) = 330mm^2
This cropped picture will be exactly the same image recorded by a APS-sized sensor with the same 72mm focal length taken at 1/72 and with the same amount of handshake. Obviously, if there is no handshake blur in the original image, do you expect handshake blur to appear suddenly just because you crop it to a smaller field of view now? Obviously no. But then, you would notice that 72mm on a 22mm x 15mm sensor would be about 110mm on 35mm format equivalent.
You want a 4R (6" x 4") size picture and so need to blow the 2 images up.
For the first picture, its blown up by 4.2 times by length or 18 times by area.
For the second picture, it's magnified by 6.9 times by length or 48 times by area.
If both pictures are the same image except for field of view (due to the crop) and there is no handshake blur recorded in both pictures, then do you expect the 2nd picture to suddenly show handshake blur just because it's blown up a lot more? The answer is obviously "No" again.
Take all the above together, handshake blur risk is the same regardless of sensor size and, as explained in details above, there is no reason for the guideline to change with the image sensor size. In anticipation of some who may still be confused, I'm not saying that you will surely avoid handshake blur if you follow the 1/focal length guideline. I'm just saying that if the guideline is 1/actual focal length (or whatever it is) for the 35mm format to avoid handshake blur to be recorded, then it should be the same for other sensor/film sizes (i.e. format.) because it's the same image formed at the same focal length with same amount of handshake at same shutter speed but only different in FOV (field of view).
Obviously you didn't read the part which talks about prosumer cameras in particular.
Theory not much use. Let's look at real life practicals.
Practically, this can be best vindicated by prosumer cameras without image stabiliser or with image stabiliser turned off. Many have been shooting handheld at 1/125 for the long end which is 280mm-over 400mm on the equivalent 35mm format. All I can say is that many users can attest to this.
Read the 2 links in my original post to see actual practical results to believe.