Large Format Considerations (1 reply and 1 comment)
Dear Roger and David,
I hope you’ve enjoyed the holidays.
David posted this article by Steve Yedlin on Cinematography.com» that I found quite interesting:https://www.yedlin.net/NerdyFilmTechStuff/LargeFormatMisconceptions.html»
It talks about the fallacies and misconceptions that are propagated about shooting large format. In short, paraphrased, Steve asserts that there is no such thing as a “large-format-look”, as depth of field, angle of view and uniformity of magnification can be equivalently manipulated and matched via t-stop and angle of view/focal length on either sensor size.
This leaves me to wonder, what – if at all – would make you opt for shooting large format like Alexa LF or even 65.
What I’ve been hearing the most is the fact that you can use a longer lens on LF for the same angle of view of its S35 equivalent. Given that you can match the depth of field via t-stop, the only difference in the image could then come from the differing uniformity of magnification between the shorter S35 and the longer LF lens. While Steve mentioned that even different lenses of the same focal length can have dramatically different magnifications across their image circle, I would assume that longer lenses (at least in the “normal” range) tend to have more uniform magnifications/less distortion than their S35 shorter AoV-equivalents (though I wonder if modern high-end lenses like Master Primes would really have more distortion on their shorter focal lengths than let’s say the equivalent Signature Primes on LF).
If that’s true, it would be the only significant difference I could see in larger format sensors (besides resolution) – which given the cost and production considerations of large format makes me wonder about the increasing success of LF sensors.
I hope I didn’t miss it elsewhere, Roger, but why for example did you choose to shoot “1917” on a 40mm Signature Prime on LF rather than a 32mm Master Prime on a Mini?
I understand a t1.4 on a Master Prime wouldn’t quite give you the shallowness of a t1.8 on the Signature Primes in LF, but could at least match the depth of field of shooting at t2⅓ on LF?
Thanks a lot and all the best,
You can match field of view between formats of different sizes.
You can usually match depth of field but at some point you run into the widest aperture possible for the lens on the smaller format. So if you manage to find an f/1.4 lens for a full-frame camera, you might have a hard time finding an f/0.95 lens for a Super-35 camera to match that depth of field.
But there is a resolution difference between an Alexa Mini and an Alexa Mini LF. And in larger formats, the same lenses can seem to resolve more because if they are designed to resolve "x" lines per millimeter, then if the sensor has more millimeters overall, you are getting more resolution out of the lens. And the same sensor, if larger, will seem to have less noise simply because the image is being enlarged less to fill the same sized screen. And with less visible noise, you have some improvement in dynamic range since more of the underexposed areas are "usable". For all of these reasons, often larger format images seem to look more "clear" (you often hear the vague term "IQ" -- image quality -- used in photography forums to talk about larger formats.)
But also, some of that is an optical illusion because they often have less depth of field, so what's left in focus looks sharper because of the softness of the background. Remember the early days of shooting video on the Canon 5D DSLR? Even if the recording didn't even resolve normal HD, the images seemed very sharp compared to 2/3" HD because of the shallow focus.
Also keep in mind that particular lenses have personalities, so even if you match depth of field and field of view, you still have design differences that will manifest themselves in different ways, like in terms of fall-off, or "flatness" versus "roundness", etc.
Thank you, David! And happy new year!