Resolving Detail


Q I recently set up to shoot some interviews and thought about the concept of what framing I should use. Usually, I would just mix up the framing, going from medium to close-up, etc., so that once the footage got into edit they wouldn’t be hampered by jump cuts. But as I set up my gear, I got into a discussion with the interviewer/director about framing the interviews. He brought up the point that if the only reason I changed framing was to avoid jump cuts, how was I to know which bites would be cut against other bites? He suggested that I assumed I knew how things were going to be cut and that maybe I was actually making things worse. It was really an academic discussion. We both agreed that "mixing things up" would provide some options in edit. If, on the other hand, all the shots were the same framing, every edit—not just a few—would be jump cuts.

Fast-forward to my next project with the same interviewer/director. This time we were shooting people on a greenscreen. The idea of shooting the interviews vertically popped into my head. I figured that since the left and right sides of the frame were usually just wasted green, orienting the camera vertically would fill that space. By filling more of the screen, there would be more options for resizing when the people were composited. I’m sure this isn’t a new idea, but I’m curious if the lens would cooperate. Are lenses optimized for shooting horizontally? Or, more simply, is this a good idea?
Via email

A Shooting vertically isn’t a new idea, but asking about lens orientation is a great question. A simple answer could be based on what you might glean from still photographers. I think if you ask any professional photographer if they think twice about shooting vertically versus horizontally because of lens performance, they would say no.

But let’s look at this question a little more methodically. Discussions about lens performance inevitably focus on MTF, the Modulation Transfer Function. MTF is one of the measures of how well a lens will reproduce a scene.

If you think about resolving detail in a scene, you’re dealing with contrast—the difference between darkness and brightness and how quickly that difference occurs across a given amount of space. This is more than just pixel count; it’s really how those pixels are changed from black to white, depending on how frequently the detail in the scene changes from black to white.

Think of poppy seeds on a white sheet of paper. If the lens presents black seeds against bright white, the MTF of the lens is better than if the seeds look gray. But optical tests don’t use poppy seeds. They use charts filled with black and white lines that change in size.

The charts use lines in different orientations to measure different MTFs. If you dive deep into optic specifications, you might see terms like tangential or meridional and sagittal that relate to the orientation of the lines.