Digital Crop Factor

As we know, any lens projects a circular image to film or sensor. With full frame sensors, a rectangular portion is captured and with less than full frame sensors, a smaller portion is captured. This gives the appearance of magnification but is really a crop factor.

OK, we know all this. Here's the question. What impact does the crop factor have on the old "one over focal length" rule for shutter speed when handholding a lens.

At first blush, I say it should have no effect because the focal length of the lens does not change. Any movement will be magnified over the same "lever" of the same focal length. However, a colleague has suggested that while this is true, and indeed the amount of movement of a point on the sensor due to movement of the lens will be the same in both cases, with a less than full frame sensor, this movement will be across a smaller field hence it will be or appear to be magnified also.

That seems to make sense and now I don't know what to believe. So if I'm handholding a 200mm lens on a 1.6 crop factor sensor camera, do I need 1/200 sec shutter speed or 1/320 sec (200 x 1.6)?

Appreciate any help on this because I get asked a lot at workshops and thought I knew the answer but now am not sure.
John Reed

Everyone's going to have their own view on this but mine, for what it's worth, is that it needs to be the inverse taking the crop into effect.

I disagree with Robert. Cropping the image doesn't affect sharpness. I think the old rule holds. IS or VR helps, but crop factor has no effect.

1/focal length is just a rule of thumb, and a rather loose one at that.
Some folks are as steady as a rock while handholding lenses, while others aren't.
I agree with Bob solely for the sake of playing it safe.

One thing is certain...it's a whole lot easier calculating 1/focal length than 1/focal length x 1.6

Many thanks to all opinions so far. Rest assured I know that the "rule" is just a guideline and I'm not asking to be educated in that regard. I also know that there will be a plethora of "opinions". I'm simply looking for the actual answer based on the physics of optics. Call it a quirk. Rest assured also I will not bore or disinterest students with theory. Not my style anyway.

So, forget about the rule of thumb or teaching approaches and let's stick to figuring out the answer to the question from simply a purists point of view. My mistake for not simply phrasing that way in the first place.
JR

I think that this is a case of 'if all else is equal'.

It's a bit like depth of field. Lots of what we consider to be in focus, isn't. It's just acceptably out of focus (i.e. either side of the plane of focus), and it does the job. Hand-holding will never give you a perfectly shake-free image, just that the shake is acceptable, and the slower the shutter speed, the more obvious this is.

The more we crop an image, the more obvious the shake that's there. So with a full frame and a DX sensor taking an image of something the same distance away, the shake should be more visible with the latter (the image is cropped so the flaws are more visible). Think of it this way - imagine the crop factor was 10, so that your 500 was effectively a 5000mm lens, with an enormous crop factor - any tiny movement would be very visible (just as with macro lenses where the 'rule' doesn't really work). The amount of shake is the same, but its like the difference between holding a laser pointer steady and a lantern - movement is more visible in the former, even if it's the same amount of shake for both.

Therefore I plump for the 1/effective focal length guideline. Willing to be 'out-logiced'!

Apologies for various rubbish analogies



Mark

One of the first things I do with a new lens is shoot a series of handheld photos of graph paper, and then take a look at them blown up to so how they do. I use an Olympus E-3 as my primary camera, and it has camera-based IS, and the effectiveness varies markedly based on the lens length. Today, with all of the data readily available and effectively no cost for a photo, I don't think it's sensible to use these general rules. An hour with the camera will tell you far more.

An example: by experiment, I discovered that I could safely shoot my 300mm at 1/250, but my 90-250mm at 250mm needed 1/500. Perhaps the zoom lens weighs a little more, and decreases my ability to hold it steady. Oh, and the Olympus has a crop factor of 2.0 if that matters.

John, this is an interesting question. I don't think you will get the precise answer you are looking for. However, I will say that I think the crop factor does affect "the rule". While the center of the lens' circle is captured at the same focal length indicated on the lens, the sensor is smaller. This is important as when a print is made or the image projected more magnification is required, so any small flaws are also magnified.

I vote for 1/cropped focal length as the "rule"

Personally I think that what you are seeing with the 1 over rule is magnification of any shaking of the lens or camera as you go to longer lenses and therefore it is independent of the crop factor--LJP

Technical excellence is only one of the reasons I use a tripod.

I like to mull over compositions. Mulling is much easier when you're not holding up a camera.

Pete D. retreating into a corner to think about that...