Why should minimum shutter speed for long lenses be driven by the crop factor or "equ

Why should minimum shutter speed for long lenses be driven by the crop factor or "equivalent focal length?

It is often said that the minimum shutter speed for long lenses should be adjusted for the "crop factor". For example, we frequently see advice such as the following:

The minimum shutter speed for a 200 mm telephoto lens should be adjusted from 1/200" up to 1/320" for a camera with an APC or APS-C sensor (crop factor of 1.6).

I have never seen this issue questioned before, but I have come to believe that the sensor size has nothing to do with the minimum shutter speed required for long lenses. In fact it is difficult to find anything about the crop factor of sensors that is of any concern beyond the following facts:

a) the higher the crop factor, the longer is the "equivalent focal length".

b) the higher the crop factor, the smaller is the captured field as compared to a full frame sensor.

Looking deeper, we find that "Equivalent focal length" has some interesting implications!

1) There is a widespread belief that perspective changes with focal length (false).

2) Conventional wisdom says that the longer the focal length, the more likely it is that the image qualilty will suffer from lens motion.

However, we know that perspective is actually controlled by the position of the viewer in relation to the scene. The focal length of a lens may force the photographer to move in closer or step back further, and it is only in that sense that the focal length affects the perspective. See http://www.earthboundl​ight.com …istance-focal-length.html

So, wherever the term "equivalent focal length" is used, there is the implication that we should have to use a faster shutter speed - because of the crop factor.

In my opinion, the "crop factor" or "equivalent focal length" is a bug-a-boo that has mistakenly become the basis for a higher shutter speed requirement, when in fact the only thing that changes as a result of the crop factor is how much of the scene is being captured (how big the capture field is).

Another thing that should be kept in mind is that each lens has an actual focal length, and that actual focal length is what it is - regardless of what size sensor lies in the focal plane.

Your comments are invited.

I don't adjust for the crop factor either since the extra reach isn't really.

Interesting, this is a very good question, one I never thought of, but it makes perfect sense "not" to

Robert_Lay wrote in post #4767516
Why should minimum shutter speed for long lenses be driven by the crop factor or "equivalent focal length?

Agree fully with your thoughts about the so-called equivalent focal length - it's bogus, but useful for those of us comparing field of view between 35 mm and APS-C, or (in my distant past) comparing 35 mm with 4 X 5.

But minimum shutter speed (a rule of thumb, of course) depends on camera movement and ratio of enlargement of the final print or screen. The more the image is enlarged the more apparent camera shake will be.

Just like depth of field - both depend on the viewer's perception at the final enlargement.

Obviously, for a given final size, a 1.6 crop image will be enlarged 1.6 X (linear) as much as a full frame image. So shutter speed needs to be 1.6 X for the same amount of blur in the final image.

-js

The FOV is narrower with a crop camera just as it would be with a longer FL on a FF. The movement of the camera will cover a larger distance at the target just as a longer FL would. Does that make any sense? The movement of the camera may deflect the view by 1 degree and with a smaller FOV that would be a larger percentage of the target. That still doesn't sound very clear.

I'd given up on this one !

Next time someone comes up with it try getting them to explain why...it's great to watch.

Adding a little to the debate though.....
Given that any movement will move a given image point to a slightly different position on the sensor then should pixel density be more relevent to shutter speed and focal length ?

Bob

Canon Bob wrote in post #4767640
I'd given up on this one !

Next time someone comes up with it try getting them to explain why...it's great to watch.

Don't give up, Bob. I tried to explain why - your comments are welcome...

-js

I disagree Bob. This comes to the degree of enlargement for like print sizes. Starting from a smaller sensor with the same effective focal length, the smaller sensor will be enlarged 1.6X more for the same print size and the effects of motion will be enlarged by the same amount.

This is analogous to the well understood phenomenon that larger formats require less resolution performance from the lens since the detail resolved will be enlarged less.

It seems that one of the problems with digital is that it has gotten us out of the darkroom. You never see the tiny negative you are starting from so the enlargement effect is lost on the casual shooter. 1.6X and FF images start on the monitor the same size.

Edit...Let me emphasize. I strongly disagree. Think long and hard about the effects of enlargement on shake blur and then tell me why a normal P&S digicam cannot be handheld for a good shot a 1/6 second even though the focal length of the lens is only 6.3mm

If you had a FOV of 100 degrees and you deflect the camera by 10 degrees it would be 10% of the image that was changed. If the FOV is 200 degrees the change would be 5%. Still not very clear. It would be so easy with a chalkboard.

I agree that the "rule-of-thumb" that has worked for 35mm film cameras needs to be modified by the "crop factor" value because of the extra degree of enlargement to achieve the displayed or printed image as compared to 35mm film or images from so-called "full-frame" DSLR's.

Any blur due to camera/lens motion is more obvious with a greater degree of enlargement (difference in size between the image on the film/sensor and the final image, whether printed or displayed on a computer screen).

number six wrote in post #4767655
Don't give up, Bob. I tried to explain why - your comments are welcome...

-js

I guess that depends whether you frame the shot the same with the two sensor sizes or stand in the same place and end up with a different frame.
Are you walking or cropping?

Bob

number six wrote in post #4767584
Agree fully with your thoughts about the so-called equivalent focal length - it's bogus, but useful for those of us comparing field of view between 35 mm and APS-C, or (in my distant past) comparing 35 mm with 4 X 5.

But minimum shutter speed (a rule of thumb, of course) depends on camera movement and ratio of enlargement of the final print or screen. The more the image is enlarged the more apparent camera shake will be.

Just like depth of field - both depend on the viewer's perception at the final enlargement.

Obviously, for a given final size, a 1.6 crop image will be enlarged 1.6 X (linear) as much as a full frame image. So shutter speed needs to be 1.6 X for the same amount of blur in the final image.

-js

have to agree here, though since most people are pixel peepers instead of printers, i'll add a bit more to it. the derivation of 1/focal length is from the resolution of film, no matter what size sheet of film, it'll always have the same possible density (think it's something like 100 microns, and any way you slice it it's at least 10x larger than even a 6mp 1.6 crop). at a certain focal length, the projected real image will "travel" across the film at a rate proportional to the focal length (the higher the number, the faster the travel). in order to have you hand movements not distort the recorded image (i.e. not "travel"), the rate at which the projected image travels must be smaller than the shutter speed times a constant (which is previously determined by the film and focal length) when the "travel" is less that that threshold, the loss of quality by the blur is minimized to an acceptable level. on a camera like the 1d (original, 4mp 1.3 crop), the sensor (comparable to the film), has a lower density than a camera like the 40d (which is 10mp and 1.6 crop, meaning it's pixel density per unit area is several times larger, many several times), and hence needs only a slower shutter speed to make sure that the projected image stays within the boundaries of that pixel. by this same principle, ccd shift sensors can manage to stabilize the image to get longer shutter speeds without blur (do note however, that as the focal length increases, the effectiveness of the shift decreases faster). yes, this is a vast over simplification of the physics behind the rule, and i'de be happy to derive it for your specific camera as a function of focal length, but i'm not in the mood for math and using the rayleigh criterion and so on and so forth.

Canon Bob wrote in post #4767719
I guess that depends whether you frame the shot the same with the two sensor sizes or stand in the same place and end up with a different frame.
Are you walking or cropping?

Bob

If you are using a FF camera and cropping to 1.6X on every shot:
1. You should be using the 1.6X rule of thumb at the least....
2. And realistically you should get some longer lenses or a smaller format.

Who uses a full frame camera full time and always discards 70% of the image?

It's like the question: "If you have a minivan and you only use it for drag racing, should you remove the seats?"

Kind of generates a few other questions, eh?

Y'all are over-thunkin' it. Just get IS on your longer lenses and keep the SS as high as you need to stop the subject movement as desired.

Robert_Lay wrote in post #4767516
Why should minimum shutter speed for long lenses be driven by the crop factor or "equivalent focal length?

Everyone else has already elaborated on it but another thing to think about.

the rule of thumb, 1/focal length is a correlation of focal length and motion blur. Correlation does not equal causation.

In reality it's the does not have a direct affect on blur.