Focal length question

[batmura]

I have a question about focal lengths of two different lenses. At 18mm on my Sigma 10-20, I get a "wider" result than at 18mm on my Nikon 18-105. The 18mm on my Nikon produces "bigger" background subjects. Maybe it's just me, but shouldn't they be the same? Or does the Sigma produce different results because it's an ultra-wide angle lens?

[xpatUSA]

You can look up lens specifications here:

http://slrgear.com/reviews/showcat.php/cat/2

Angles of view (degrees) of your lenses are:

Sigma 102.4 to 63.8
Nikon 76 to 15.3

The image circle diameter produced by different lenses can be different. "Better" lenses can have a larger image circle which reduces vignetting, I have read. Also, when it says "18" on the lens barrel, there is a small range of actual focal lengths, called a "tolerance" in Engineering. Let's say +/- 2mm. Then, with adverse tolerances, one lens could be actually 16, the other 20mm. Unlikely but still possible.

How much "bigger" is the one compared to the other?

[royphot]

In theory, yes they should. In practice when the Sigma is at 18 on the scale the focal length might well only be 16. The Nikon should be at 18 when it is at its widest, but who knows and how could it be checked.

Pre digital I could get access to an electronic shutter speed and exposure tester. Very few of the cameras I tested had speeds which matched what was set. 1/500th could quite easily be 1/350th.

The aperture setting were considerably closer to what was set but very few were exactly correct.

The important factor was that the speeds should maintain the correct relationship up and down the range. 1/500th should have the shutter open almost twice as long as 1/1000th, even if they actually were 1/387th and 1/802th.

These results were much the same for almost every make and model, except for Leica where there was no access to the focal plane. So the declared focal lengths of lenses might just be as inaccurate as the declared speeds, but does it actually matter?

Roy

[DanK]

I think there are two things going on.

First, the FOV is a function of nothing but focal length and sensor size, specifically (where d is a dimension of the sensor and F is focal length),

FOV=2arctan(d/2F).

So, if the two lenses are really at the same focal length, the field of view should be identical. But:

However, apart from manufacturing tolerances, there may be a couple of reasons why you are not actually getting identical focal lengths from the two lenses. First, were you focused at infinity? I vaguely recall reading that the focal length labels refer to FL when focused at infinity and that the effective FL changes as you focus closer. If that is right, the change is probably not similar between the two lenses. Second, I wonder about the algorithm the lenses use to report focal length to exif. For example, if my memory is right about the first issue, perhaps the different manufacturers use a different algorithm for adjusting (or not adjusting) reported focal length as focusing distance changes.

Perhaps someone who knows more about the effect of focusing distance on effective focal length can chime in.

[xpatUSA]

Thanks for the reminder, Dan.

Yes, the nominal focal lens of a lens is stated for the infinite focus condition. I do some macro work, and the effective focal length increases when focusing at less than infinity. At 1:1 magnification, the effective focal length can be doubled, the effective aperture doubled and equally the diffraction. Some cameras ( e.g. Nikon) do report the effective aperture, but not focal length AFAIK. This is seen when, for example, a fixed f/2.8 macro lens is reported by the camera as a bigger number when at full magnification. Some lenses have extra gimmickry e.g. micro-Nikkor 60mm f/2.8 so it doesn't actually go as far as f/5.6 at 1:1, if I recall correctly.

Or is it indeed the lenses that do that reporting? My Nikon D50 + micro-Nikkor did but I can't remember what the D50 + Sigma 50mm macro (old) did. Since the OP mentions a Nikon, we need more input from true Nikonistas.