Lighting HDV

[Laurent Andrieux]

Hello ! (sorry guys, don't have time to read the 2 David's lasts posts, (I like to read David's posts very carefully...) but, about Stephen and R. Edge posts :

 

 

Yes Stephen you are right and you understood perfectly well, I am sort of working on a theorical answer to these questions.

 

R. Edge : I think David is right also and he didn't tell you it was wasting time.

 

The thread you linked us to is very interesting also.

 

He pointed out the relative and subjective dimension of DOF.

 

He actually took the same example I like to give (and have given in another thread a few months ago) about zeiss and cooke.

 

We here are touching the heart of the subject : try to study DOF on a theorical point of view (that is absolutly necessary) and point how these theorical statements are not disagreeing the practical use.

 

I think many people ask questions about that and this why we're trying to put something worth in the FAQ sections.

 

Because people need it and because when it comes to focus and DOF, ACs need charts or calculators, that invoques CoC, and one has to consider the final release of aproduction as to figure out clearly what "will be acceptably sharp on the screen".

 

It's true these questions are not so easy to answer.

 

I've given an example sometimes ago. In the definition of the hyperfocal distance, the term that is commonly used (f²) is wrong it should be used f X "normal (ie human eye field of view) f...

 

A friend of mine who is a teacher at Louis Lumiere school and PHD wrote his thesis on the DOF, he found a lot of weird things about it (as the one I was just pointing out)...

 

(I'm sorry guys but we have a BIG problem in the school in teach so that I'm very busy these days...)

[Robert Edge]

David,

 

I'm going to do my first super-16 shoot in a couple of weeks and I'll be very pleased if I find that I can easily judge the apparent depth of field that the finished product will have just by looking through the viewfinder. I'm skeptical, given that I'll be probably be using neutral density filters at some point and I'll also be shooting close-ups at small apertures, but hope springs eternal. In the meantime, I figure that looking at some depth of field charts as part of my planning is like chicken soup - couldn't hurt.

 

I asked my original question because I believe that apparent depth of field ultimately depends on how much an image is enlarged and how far the image is from the viewer. Without assumptions about these questions, it is impossible to posit a circle of confusion and, consequently, impossible to construct a depth of field chart. It's also impossible to make a depth of field scale for a lens. I know what the underlying assumptions are when I look at a still camera DOF chart or at a Nikon DOF scale, but I haven't got a clue what assumptions underlie the ASC charts or a Zeiss cine scale. I'd like to know, if only because I am curious.

 

I do realize that there is a significant subjective element to one's perception of depth of field and that charts are very much a rough guide.

 

I asked specifically about DOF for television broadcast because it occurred to me that the resolution of a television screen might have an impact on apparent depth of field and because there may be differences, as between television broadcast and theatrical projection, in enlargement magnification and viewer distance that are absolute rather than relative. I do understand that television screens come in different sizes and resolutions and that viewers can sit as near or as far from a screen as they please. Of course, the same is true of theaters. I'm just interested in whether people have thought about these questions and whether they have come to any conclusions about what the implications are for super 16 depth of field, if there are any.

 

I am interested in depth of field for super 16, because it is the format I am using, rather than in how super 16 and 35mm compare. I have read the ASC Manual. According to the introduction to the DOF charts, at page, 632-33, the COC for 35mm is .001". While this is expressed as applicable to all formats, the text also notes that the traditional COC for 16mm is .0005". In the charts on extreme close-ups, at pages 744-45, the COC for 35mm is .001" and the COC for 16mm and super 16 is .006" (I assume that .006" should be .0006", although maybe I am misreading something).

 

As you know, there are some caveats to these numbers. In the article on lenses, at page 176, there are interesting comments about the impact of current lenses and stocks on circle of confusion assumptions. These comments are presented in somewhat more summary form in the introduction to the charts. Depending on circumstances, these comments could lead one to adopt a smaller COC for 35mm and perhaps a proportionately smaller COC for super 16.

 

Although it isn't an issue for me, if one does want to compare DOF as between super 16 and 35mm the example you present in your post is interesting:

 

A 50mm lens at f/4 focused at 5' has a depth of field of 4'9" to 5'4".  You switch to a 25mm lens on a Super-16 camera at f/4 focused to 5' and the depth of field is from 4'0" to 6'7" -- a HUGE difference.  If one is shooting either format for the SAME end use (theatrical presentation, TV, whatever) I have a hard time believing that differences in circles of confusion used to calculate depth of field would ever reverse the fact that a 25mm has more depth of field than a 50mm at the same f-stop and distance focused.

<{POST_SNAPBACK}>

 

It appears that the greater depth of field provided by the 25mm lens is indeed significantly reduced if one changes the COC from .001" to .0005". The DOF for the 25mm lens becomes 4'5" to 5'8". This is not necessarily good news if one WANTS DOF. On the other hand, if narrow DOF is desired, it appears that the smaller COC, if it is an accurate predictor of behaviour of the image, and for the example posed, will indeed result in similar depth of field for the two formats.

 

The question is, does the smaller COC reflect the reality of what the image looks like when it is broadcast or projected. This brings us back to the question of magnification of the image and viewer distance.

 

It would be interesting to see what happens with other examples, because I am certainly not questioning the fact that a smaller format/shorter focal length tends to yield more depth of field than a larger format/longer focal length.

 

Thanks very much for your comments.

[David Mullen ASC]

The question is why depth of field is a such particular concern of yours, as opposed to focusing, which is a different issue.

 

Obviously if you want something in focus, you focus on it. Depth of field, on the other hand, is really about: (1)WHAT ELSE is also noticably in focus besides what you focused on; and (2) if you missed the focus, was the subject sufficiently in focus to be acceptable.

 

Trouble is, deciding what is "acceptably sharp" is never going to be an exact science because it is about perception -- that and because the same project can be seen on a wide variety of screen sizes, even if just for TV. (I remember this old argument with someone who felt that 480P video was sufficient resolution for indie filmmakers as long as you could make sure that the audience never saw it on a large screen or sat too close to the screen. Trouble is, no one has that kind of control over how their movie is seen.)

 

So if your interest in depth of field is because you want a "Citizen Kane" type of deep focus effect, then obviously you should light at a sufficient level to stop down the lens AS MUCH AS IS POSSIBLE AND IS PRACTICAL. And shoot tests if you're really concerned that you can't see the effect in the lens. But my experience is that you stop down the lens as much as you can anyway and hope it is enough -- because if it isn't, you already stopped down the lens as much as you could anyway, so you're probably going to have to live with it.

 

On the other hand, if your interest in depth of field is to know how much leeway you have to get something to look acceptably sharp even when you misfocus, then use the more critical figures that a chart gives you. However, on most shoots, if you really know you missed the focus, generally you do another take rather than always just hoping depth of field will hide all your focusing mistakes.

 

All of this is to say that after thirty features, I rarely have dealt much with depth of field charts other than in academic discussions like this. Sure, my 1st AC occasionally scrambles to check something if he's worried, but if it a question of splitting the focus between two subjects, I'm more of the belief that it's always better to actually focus on something than in the dead space between two objects. But that's just my philosophy. If I were doing more "Citizen Kane" type shots, I may be trying that trick of splitting focus, etc. more often after I had stopped down sufficiently.

 

Now if I were shooting miniatures, that's a whole other issue, then depth of field charts are critical for calculating how far you really have to plan on stopping down, in order to get enough lights for the shoot. But even then, ultimately an efx artist shooting miniatures will also use their eyes and shoot tests to determine what LOOKS right to their eyes, and on screen, regardless of calculations. Its just like the "rules" regarding frame rates for scaling motion for miniatures.

 

But my overall point is that depth of field, by its very nature, always involves approximations. For example, the .0010" versus .0005" issue. It's not like those are magic numbers with nothing possible between them, that it's either one or the other. If one figure is for TV and one is for theatrical, then what about big screen TV versus small screen theatrical? What about viewing distances? It seems you could just as easily say that you should split the difference and make the c of c .00075"...

 

So my question is, what specifically are you trying to achieve in your shots where a depth of field calculation is so critical?

[Guest Trevor Swaim]

PS : I generally find out your abreviations, you american and english people (funny game to guess actually) but this time, i need help : IMHO ? :blink:

<{POST_SNAPBACK}>

 

imho = in my humble opinion