Shallow depth of field is desirable, particularly in portrait photography, due to its subject and background separation, and its pleasing aesthetic. But while your gut reaction for depth might be to go for the fastest prime you can, you might be surprised at this side-by-side comparison.
There was a lesson I learned early on in photography, quite by accident. In fact, I still have the image that marks the lesson. I was fairly new to the craft, with only a year or two under my belt, and I wanted a 70-200mm f/2.8. The problem was, the Canon edition I wanted was far too expensive for me to be able to afford back then and although I looked at other lens manufacturers, I still struggled. Then I saw the f/4 variant, which while still an L-range lens, isn't anywhere near as expensive due to its narrower widest aperture. I got such a good price that I knew I could resell it if it was a complete disappointment.
When it arrived, I took it out at sunset on a freezing cold winter's afternoon and found some local horses to act as the subject. To my delight and complete surprise, at 200mm, f/4 created some pleasing bokeh and decent subject separation. I hadn't understood the relationship between focal length and depth in my images, or at least not to the degree I thought I did.
This video by Pye Jirsa of SLR Lounge is a more expensive and more recent example of the same revelation. Jirsa compares the Canon RF 85mm f/1.2 with the Canon RF 70-200mm f/2.8, and the results are surprising. At around the 7:50 mark of this video, Jirsa points out that most people would get the two example images the wrong way around with which lens was used, and I wholeheartedly agree. While the 85mm is brilliant, the 70-200mm wins in a lot of categories. What do you think?
If you want to compare subject separation between different lenses divide the focal length with the aperture. The bigger the number the better the separation.
Hmm, if that's true then a 50mm f1.0 and a 200mm f4 would both have the same focus separation. Can't quite put my finger on it but that does not sound right. Maybe it's to do with subject distance - that rule of thumb might be right if the subject distance is kept the same, but then (obviously) you would have very different framing.
Maybe I missed something in this comparison (I didn't watch the whole thing), but when comparing bokeh and separation between two lenses, the same--or very similar--image needs to be shot with both lenses. To the point, what conclusion am I suppose to draw from a tightly framed portrait image of a couple embracing and a wide(r)-angle, landscape shot of an infant with two similar but different backgrounds? The background trees seem to be closer to the couple in the 85mm portrait shot, whereas the baby is in the middle of the path with the background at "infinity" in the landscape image. To note, the trees at the edge of the 200mm image are in clear focus, and the sharpness fades gradually. Not the case in the 85mm shot.
Now, I'm not disputing the author's point. I've gotten great separation with both lenses, and am very pleased with both lenses. However, it's rather frustrating seeing the lack of control in this comparison. Maybe the commentator can redo the test in a more controlled environment?
bokeh is not how unsharp the background is, its the quality of the transition.
Great stuff, but… you are not always at a football field or at the beach. Some people also shoot in a bit tighter spaces … try to do that at 200mm . Not always possible. We are mixing a bit the right tool with flexibility;)
Unless the person is the same size in the shot, how can we compare the separation and the bokeh? In the video, the photographer shows completely different photos, made with the 2 lenses and says 'this looks like a prime'. To whom? Pretty useless 'comparison'.
welcome to f-stoppers :)
Pretty much the same point I made earlier. You just stated it more succinctly (and less diplomatically).