Depth of Field and CoC

Ashley Barron · February 19, 2009

Hi there,

What is the difference in depth of field between 35mm and 16mm?

Also, I am so confused as to what circle of confusion is and how it affects the focus and what focus puller needs to know about it to do his/her job.

Can someone please explain?

Thanking you,

Ash.

David Mullen ASC · February 19, 2009

Because the 16mm image will be enlarged by twice compared to the 35mm image, for the final viewing image, you need to double the Circle of Confusion figure because focus is twice as critical. Traditionally the figure for 35mm for theatrical use is .001", so for 16mm, you use .0005".

Practically what it means is that, let's say that 16mm is half the size of 35mm to make the math easier. So you'd switch from a 50mm lens to a 25mm lens -- half the focal length -- to maintain the same field of view at the same distance when switching from 35mm film to 16mm film.

If you look on a depth of field chart, you'd see that the depth of field for a 25mm lens is the equivalent of stopping down four-stops compared to a 50mm lens. However, by halving the circle of confusion figure, you lose two-stops of depth of field, it becomes more critical, so that four-stop difference is now a two-stop difference. So you can use the same DOF chart for 35mm when shooting in 16mm if you just look at the figures on the column as if you stopped down by two more stops.

Circle of Confusion itself is hard to explain. See:

http://en.wikipedia.org/wiki/Circle_of_confusion

The rough idea is that the figure is the amount a point of light can be out of focus -- and thus become a disk or circle of "x" diameter -- and still be "acceptably" sharp enough to be considered a point. More or less.

Anyway, it generally works out that the magnification factor between formats of different sizes works out to be the practical difference in depth of field in terms of stops -- for example, a 2/3" CCD camera has a magnification factor of 2.5X compared to 35mm in terms of horizontal view and thus focal lengths used to match field of view, and the difference in depth of field is 2.5 stops. 35mm is roughly 2X bigger than 16mm and thus the depth of field difference is 2-stops.

You have to ask yourself what you are using these depth of field tables for in the first place, because conceptually I have trouble with them -- I think the lens should be focused on what you want to be in focus. The charts are really only useful if either you are doing a deep focus shot and need to know the stop to light for to achieve the range you want, or if you are doing a shot where you can't pull focus and what to find the best distance to hold the most amount of the subject, or when you miss the focus and want to know if it will still be acceptably sharp. Some people use them for trying to split the focus between two people, which I find rarely works except when you are stopped down.

But I feel more comfortable if we are focused on something specific -- not the air between two objects -- and I worry about AC's relying too much on charts to tell them that their focusing was "acceptable."

This is my general conceptual problem, the whole notion of the perception of sharpness, which seems rather subjective and not necessarily scientific. The CoC figure we use tends to be based on the size of the image during presentation, but even in a movie theater, there are different sized screens and people sit at different distances to them. Someone in the front row is going to see focus mistakes a lot more clearly than someone in the back row.

Satsuki Murashige · February 19, 2009

I think the lens should be focused on what you want to be in focus. The charts are really only useful if ... you miss the focus and want to know if it will still be acceptably sharp....

I feel more comfortable if we are focused on something specific -- not the air between two objects -- and I worry about AC's relying too much on charts to tell them that their focusing was "acceptable."

I understand what you're saying David, but I don't entirely agree.

I think you have to use depth of field for steadicam/handheld type shots where the camera is usually fairly close to the subject, and the camera-to subject-distance is always changing as well as changing from take to take, and thus virtually impossible to keep "perfectly" focused down to the inch 100% of the time. I think the 1st AC's dirty little secret is that because we normally shoot these shots with wide lenses, that we're able get away with estimating rather than knowing exactly the focus distance at every moment for this type of shot. Naturally, the bigger the format, the better at estimating we have to be. But if you really think about it, it's just not humanly possible to be a human Panatape while running backwards behind the steadicam op with a Preston and peeking over his shoulder, though some guys get very close to that.

Then again, maybe I just don't have enough experience pulling focus yet to be the human Panatape. But I have to think that even the top 1st's in the business start sweating when they have to do these types of shots wide open on anamorphic. If I can ask, what has your experience been with your 1st's over the years as your budgets have gotten bigger?

David Mullen ASC · February 19, 2009

I understand that there are types of shots where you have to hope the depth of field covers what you need, like during a running shot. As a DP, I have to factor that into account, or at least, acknowledge how hard it is going to be for the AC when I talk to the director.

My issue is more in the run of the mill static close-up of an actor sitting where the AC constantly consults a DOF chart before he tells you that he got it -- that unnerves me, the notion that he didn't really "get it" but the chart tells him that it was acceptable so we can move on.

Now and then, that's fine, but doing it all the time drives me up the wall because DOF charts by their very definition cannot be 100% accurate -- we're talking about the vague concept of "acceptably sharp" for the vague concept of "viewing size" which determines CoC.

So any AC who constantly relies of DOF charts rather than focusing on a precise spot is asking for trouble.

I recently had a raking two-shot on a couch, a medium shot, on the Genesis -- we had a day-player AC pulling focus who had worked on some big shows -- using the Genesis. After the first take, the shot was clearly soft and I told him -- he said he was setting the focus between the two people because at f/2.8, on the DOF chart, it should hold the split.

Well, it clearly didn't on the HD monitor. He didn't consult me as to what to focus on, which I had assumed he was going to focus on the person talking through the whole shot -- he just set the lens for a split that couldn't hold because a chart told him he could get away with it.

Three days later a different guest AC did exactly the same thing on another raking two-shot, ruining the first take, and I had to talk to him about why I'd prefer the focus be on one or the other actor in the scene, not just set for the space between them. We actually stopped and set the focus on each actor and the space between them where the DOF chart said it should hold, and I could see on the HD monitor exactly where the lens was set in each case, and that the eyes clearly were not sharp when the focus was not set for them. He just scratched his head and said "well, the chart says I should be able to hold the split."

THAT'S the reason why I have a problem with DOF charts and the AC's who rely too much on them -- because I think a lens should be focused on a subject, not the air between two subjects, unless you have a good reason not to. This is what I tell all my AC's, to focus on something specific and if there is a problem, tell me about it, don't just start using DOF charts and setting the focus on vague areas that you hope the subject will fall into. If I wanted a raking two-shot to hold focus across both actors, I would do something specific to achieve that, like light to a high f-stop or use a slant-focus lens.

--

Oddly enough, I have just the same if not more focus problems today working on higher budgets and better crews than I did in the past. I attribute this mostly to the rise in two-camera shooting, which is a bit of a sloppier technique than single camera. Also, back when I was an operator/DP, I was a lot more critical about double-checking focus, and since I was the DP, I could insist on taking the time if I felt the AC needed it, and double-check his measurements with my eye-focusing. As a DP back at video village, I have less control over the AC-operator relationship and have to trust them when they tell me everything is fine. I find that if I "hover" over them too much and question them too much on focus, it makes them feel like I don't trust them. But the truth is that I now have more focus problems than I did in the past when I was a micro-manager about focus.

I was particularly worried about focus on my four anamorphic features, and I tended to light to an f/4 whenever I could -- in the end, I have better focus in most of them than my Super-35 movies shot at f/2.8 despite the fact that anamorphic at f/4 still has less depth of field. I think that's because I worried more about focus and thus took more time getting it right on the anamorphic shows.

Now and then I do run across a truly gifted focus puller that allows me to shoot at f/2.8 and not worry so much about the focus problems -- trouble is, those types of AC's are highly in demand and are hard to keep.

But the other problem is the tradition of using a B-camera for a tighter shot; generally the A-camera crew is more experienced so they should be the ones getting the harder shot. But everyone looks at me funny when I insist the A-camera shoot the tighter angle in 2-camera set-ups. The B crew looks offended, the A crew was looking forward to getting the easier shot, etc.

Dan Diaconu M · February 19, 2009

I was able to focus on anything the light touches using two video cameras mounted on the sides of the mate box. (those in the field for some time might remember). Google 6160607 for details on the apparatus.

That was moving camera/targets, wide open, constant or random moves. Constant being dolly-in or out and random being swinging target (think of a pendulum) combined with dolly-in or out. In short, it is possible to keep anything in focus each and every time (first take included) without rehearsals. On coc, personal taste or confidence using charts, no comment.

Cheers,

Dan.

PS. David, if you are curious to try the device, it is closer to you than you think. PM me if interested (beer)

Satsuki Murashige · February 19, 2009

THAT'S the reason why I have a problem with DOF charts and the AC's who rely too much on them -- because I think a lens should be focused on a subject, not the air between two subjects, unless you have a good reason not to. This is what I tell all my AC's, to focus on something specific and if there is a problem, tell me about it, don't just start using DOF charts and setting the focus on vague areas that you hope the subject will fall into. If I wanted a raking two-shot to hold focus across both actors, I would do something specific to achieve that, like light to a high f-stop or use a slant-focus lens.

Ah, now I understand your frustration! It's really terrible when you feel like your AC doesn't have your back and give you what you need. If the image looks soft, then it's SOFT and arguing with the DP or anyone else that it "shouldn't be" because a chart said so is a waste of time. I had that lesson beaten out of me early on working on my first Steadicam show with a HPX500/Digiprime combo that had serious back focus issues. I never argue with a DP about focus, if he's watching monitor and he tells me it's soft then either I missed it or something's wrong with the backfocus, etc. Then I work out what the problem is and I fix it.

I think you're exactly right about why you had better focus on those anamorphic shows. If the 1st isn't pushing him or herself to get the best focus possible and being hypercritical of their own work, then unfortunately the DP or operator has to push them to that place. We get paid to have ulcers, just like the DP. Like Chris Keth said awhile back (think it was Chris), nothing makes you a better focus puller quickly like hearing "it's soft" on set a lot.

Patrick Neary · February 20, 2009

So any AC who constantly relies of DOF charts rather than focusing on a precise spot is asking for trouble.

Hi-

I've run into a similar situation where the AC consistently sets the focus to hyperfocal, regardless of what we're shooting- the best example being a wide establishing shot where nothing is closer than about 50' away, and yet the Ac sets focus at 8' or whatever based on the chart.

It makes no sense at all; you're just insuring that now everything will fall on the very outer edges of what's "acceptably sharp" rather than focusing on what's important...

Chris Keth · February 22, 2009

Like Chris Keth said awhile back (think it was Chris), nothing makes you a better focus puller quickly like hearing "it's soft" on set a lot.

Hehe, yeah. That was me. I did a show with a director who was very loud about it (rather than having a nice operator who will quietly tell you and just ask for another take "for camera") so, naturally, I wanted him to shut up. :lol:

Gregory Irwin · February 22, 2009

This is an interesting thread. Reading all of these posts confirms my belief that the wide shots are at times way trickier and requires much more consideration than the tight ones. Since wider shots have a broader angle of view, the depth of field may not cover the entire frame thus requiring some decision making on where and how to play focus. This especially applies to the anamorphic format. The tighter shots are obvious. Keep the actor's eyes sharp or whatever the subject matter is. There's not much decision making there, you just need to get it right. If you are not so sure on how to play focus, absolutely confer with the DP.

Best,

Greg

Edited February 22, 2009 by Gregory Irwin

Chris Keth · February 26, 2009

This is an interesting thread. Reading all of these posts confirms my belief that the wide shots are at times way trickier and requires much more consideration than the tight ones. Since wider shots have a broader angle of view, the depth of field may not cover the entire frame thus requiring some decision making on where and how to play focus. This especially applies to the anamorphic format. The tighter shots are obvious. Keep the actor's eyes sharp or whatever the subject matter is. There's not much decision making there, you just need to get it right. If you are not so sure on how to play focus, absolutely confer with the DP.

Best,

Greg

I can definitely see that. It's much more common for a wider shot to have more than one possible subject, too. Tight shots are often tight for the sole purpose of excluding all but one subject.