I have a question related to different focal lengths and want to decide which one would work better for me.
I'm planning to go full-frame this year and want to have better understanding of the difference between 85mm and 135mm focal lengths for portraits. I love fixed lenses and looked through several sample shots made using both, but there are things that you can't predict by just looking at those pictures. I'm not able to try both by myself, that's why I'm asking for help here.
The only thing that comes to mind is different "compression" you get using different focal length (the more is your focal length the less is the distance between objects).
I can't really contribute to helping you decide what focal length lens would best suit your planned work, but I do want to clarify the concept of "compression of distance".
This is a "phenomenon" that occurs as the vantage point of the camera is moved farther from the group of objects (or features) that are included in the frame. It actually has nothing to do with focal length (except through another consideration I will get to in a bit).
Consider for convenience a "landscape" shot. There are two identical telephone poles (same height), one of them 100 feet farther away from us than the other along a straight line from any camera position we choose.
Suppose we first place the camera 100 feet back from the nearer pole. Thus the two poles will be at distances of 100 ft and 200 ft. Thus their heights on the image will have ratios of 1:0.5.
Next we place the camera 200 feet back from the nearer pole.Thus the two poles will be at distances of 200 ft and 300 ft. Thus their heights on the image will have ratios of 1:0.67. Their heights in the image are more nearly the same than in the first case.
Because we judge the relative distances of objects known (or believed) to have the same size based on their relative angular size to our eye, or to their relative sizes on a photographic image, it will seem as if the two poles in the second image are "not as far apart" ans the same two poles in the first image.
If we print both images at a scale such that the nearer pole is the same height in both images, then the second pole will be smaller in the first print than in the second. Thus the second pole will seem less far behind the first pole in the second print than in the first - the phenomenon of "distance compression".
And we will actually see this even if we don't print the two shots such that the nearer pole was the same height in both prints.
Note that I have said nothing yet about focal length. I might have used the same focal length for both shots - or maybe not. But the "distance compression" effect comes only from the distance from which we take the shot.
Now why does focal length seem
to get into the act? Because if my object in each case was a picture of the two poles, I would probably have used a greater focal length in the second case (so as to best fill the frame with my two subjects, in the interest of best exploitation of the resolution of the camera).
Said another way, suppose I first took a shot with a 50 mm lens, and located the camera so the nearer pole filled some fraction of the frame height. Then I wanted to try a 100 mm lens. To get the nearer pole to fill the same fraction of the lens height, I would have to position the camera twice as far from the nearer pole as in the first shot.
And, as we saw earlier, that would make the two poles seem "closer together" in the second shot than in the first shot.
Now, I will return to the portrait context. First assume that I set up with an 85 mm lens and position the camera for a certain type of framing ("waist up", for example). I take the shot.
Next I want to try a 135 mm lens, and I still have the same framing in mind. So I will have to shoot from farther back.
The different point of perspective for the second shot causes the face proportions to be different - we have encountered a case of "distance compression" (perhaps desirably). But this is wholly a creature of the difference in camera location - not of the focal length itself.