shutter speeds for handheld natural light macro shots

[LordV]

Does anyone know the approximate forrmula for working out shutter speeds required for normal handheld macro shots in natural light to stop camera motion blur ?

I think I came across a formula something like this
1/ (focal length x crop factor x (1+M)­­­­­­^2)
where M is the lens magnification
but the nomenclature used for the (1+M) squared term was not clear and am not sure why it should be squared.

Brian V.

[DaveW]

There are rough and ready rules of thumb for avoiding camera shake Brian, but I would thing stopping subject motion blur would depend on how fast the subject was travelling? Even vibration reduction lenses only stop camera shake not subject movement, and become more and more useless the higher the magnification, with firms like Nikon advising VR is actually switched off for macro distances.

"Stopping" action is something you never do, just by appropriate shutter speed or flash reduce the length of the motion blur on film, or sensor, to less in the final print or reproduction than the human eye can resolve.

Obviously "stopping" a leaf moving gently in the wind can be done at a slower shutter speed than "freezing" the motion a racing cheetah. And again longer duration flash may "freeze" slower movement than is required to "stop" a bullet in flight. Any formula devised I would think would need to put in the speed of the moving object relative to the camera into the equation.

For camera shake the old rough and ready rule of thumb for 35mm cameras was "a shutter speed no lower than the focal length of the lens in use". Meaning 1/100th sec for a 100mm lens and 1/200th sec for a 200mm lens. With the smaller APS-C sensor I would think that needs modifying to the lenses equivalent field of view on a 35mm camera. Meaning due to Nikon's 1.5 crop factor, no slower that 1/150th second for a 100mm lens not having vibration reduction.

However that rule of thumb only works for a 35mm camera down to around the standard lens and cannot be used for wide-angle lenses, since the slowest recommended shutter speed for hand holding any 35mm camera was always 1/60th of a second.

DaveW

[Harold Gough]

Why is it not the same as for general photography? The minimum shutter speed is the reciprocal of the focal length (effective focal length when using a teleconverter). On that basis, when using tubes, it seems that the shutter speed should be multiplied by the magnification factor which excludes the cropping factor. If the image movement is frozen then further magnification will not introduce blur due to movement.

I think it is the angle of view of the lens which is the relevant factor.

Perhaps my thinking is erroneous but, so far as I have worked with this (no cropping factor), results have been acceptable.

For less than steady hands, or over rough terrain, a faster speed would be advisable.

Harold

[ChrisR]

Bitty thoughts:
Lateral (ie vertical or horizontal) movements wouldn't matter as much for a subject at infinity as they would for macro, because they would be small in relation to the field width.
Angular movements for the same angle of view would be the same,
but a 50mm lens working at 5:1 would have an angle of view more like a 250mm at infinity (that's close but not exact, I'll leave the maths to Rik!).

But it's also easier to hold a long thing than a short thing, so you win a bit back, depending how still your feet are.
If your subject is actually moving across the field of view that's usually overwhelmingly important, cos it'll be doing it faster than you'll be moving the field of view about.

[Harold Gough]

Lateral (ie vertical or horizontal) movements wouldn't matter as much for a subject at infinity as they would for macro, because they would be small in relation to the field width.

I take quite a lot of shots of mountains !

But it's also easier to hold a long thing than a short thing, so you win a bit back, depending how still your feet are.

It depends on the weight. My arms get tired quite quickly with my 300mm f2.8 or my 200-500mm zoom.

If your subject is actually moving across the field of view that's usually overwhelmingly important, cos it'll be doing it faster than you'll be moving the field of view about.

I'm not sure about "overwhelmingly". What I think you mean is that any blur due to motion will be proportionate to its speed relative to the field of view, whether that is movement by the subject or by the lens.

Harold

[rjlittlefield]

Brian, I think the reference you want is http://forums.dpreview.com/forums/read. ... e=29244641. The author there offers the explanation that:

... At high magnifications the distance from the sensor plane to the image is the sum of the image and object distances or: [(1+M) + (1+1/M)] f = [(1+M)**2 / M] f where M is the magnification and f the focal length. Since the image is magnified by a factor of M this means if we shift our view laterally by a distance of a/M (where a is again the sensor pixel width) we will shift the image by one pixel on the sensor. So the angle required for this 1 pixel shift is:

(a/M) / {[(1+M)**2 / M] f} = a/[f (1+M)**2]. This is exactly the same as the non-macro case above except for the factor of (1+M)**2 - and in fact the earler results is just the limiting case for small M.

So the upshot is that the 1/f rule is still valid for macro, but must now be modified by multiplying f by the factor of (1+M)**2. So for example at 1:1 (1+M)**2 =4 and the shutter speed for sharp handheld imaging needs to be 1/4th as long as would ordinarily be the case for the focal length.

There was a long thread about this, roughly a year ago. At that point, I wrote as a snippet of summary that:

No doubt the exact form of the correction would depend on the exact geometry of the shake, like whether it's purely rotational and where its center of rotation is. But worrying about that level of detail would be a waste of time. The key thing is that macro needs shorter exposures, and the required correction is something like 1/((m+1)^2). In other words, shooting at m=5 requires an effective exposure time that is roughly 1/36 as long as shooting a landscape with the same focal length lens.

Now, if you stick a 65 mm lens on a full-frame 35mm camera, the standard 1/f rule says that you need at most 1/65 second. Extend that lens to get m=5, and the modified 1/f rule says at most 1/2340 second. Put that lens on a 1.6 crop factor sensor, and the time gets even shorter.

Does that cover what you needed?

--Rik

[LordV]

Thanks for the discussion everybody
Rik - yes that was what I think I originally saw but was not certain purely from the way it was printed whether the (1+M) term was squared or not- indeed I'm still not certain why it is squared - my natural instinct (probably wrong) is that it should be a linear function with magnification.

You are correct though that I'm just thinking about stopping handshake/camera shake here not other subject movement issues and the linkage to flash duration times needed when doing flash shots at highish mag handheld.

Brian V.

[rjlittlefield]

I'd give real good odds that by **2, the original author intended squaring.

I'd have to review the author's model to be sure, but just offhand I suspect that the question of linear versus squared has to do with what you assume about the shake. For distant scenes, the only relevance is rotation, and it really doesn't matter where the center of rotation is. For macro, if you assume that the shake is purely rotational, then it matters a great deal where the center is. Offhand, I think that your model of linear is probably correct if the center of rotation is the entrance pupil, where the author's model of squared is probably correct if the center is the sensor and the lens is thin. But I would have mess a lot more with the symbols to be sure.

This problem illustrates one of the pitfalls of mathematical models -- or any other kind of model for that matter. The result that you get is only more or less accurate depending on how well the model matches reality. Sometimes the model is pretty sensitive to assumptions.

Of course what that usually means is that the real system is sensitive to the details. Consider, for example, shooting with a microscope objective on bellows, and imagine that the angular shake is fixed at some particular value while you get to pick the center. Given the choice, would you center at the camera or the lens?

--Rik

[Harold Gough]

For distant scenes, the only relevance is rotation, and it really doesn't matter where the center of rotation is.

Perhaps I fail to grasp your use of "rotation" here but surely any movement in the plane for the image, at the sensor, is important.

Harold

[rjlittlefield]

By "rotation", I'm talking here about what aviators would call "pitch" and "yaw", not "roll". See HERE.

The effect of "roll" does not depend on anything else. The effects of "pitch" and "yaw", in macro photography, depend on magnification, angle of view, and where the center of rotation is.

--Rik

[Harold Gough]

By "rotation", I'm talking here about what aviators would call "pitch" and "yaw", not "roll". See HERE.

The effect of "roll" does not depend on anything else. The effects of "pitch" and "yaw", in macro photography, depend on magnification, angle of view, and where the center of rotation is.

Thanks, Rik. In that case I agree with you completely.

Harold