Super 8 depth of field verses 16mm

Forum covering all aspects of small gauge cinematography! This is the main discussion forum.

Moderator: Andreas Wideroe

mattias
Posts: 8356
Joined: Wed May 15, 2002 1:31 pm
Location: Gubbängen, Stockholm, Sweden
Contact:

Post by mattias »

Lucas Lightfeat wrote: Mattias, I'm amazed that you're asking a question about this, as you are normaly such an authority on cinematographic related details, but I can assure you that 35mm has a much shallower dof than 8mm.
*lol* you got no idea what i'm talking about, do you? sorry, but not only are you actually wong, you've also missed the point of this thread completely...

(edited 6.17 pm: sorry. i didn't meant for that to sound that hard. i guess i got way too defensive...)

/matt
Last edited by mattias on Wed Feb 05, 2003 6:19 pm, edited 1 time in total.
mattias
Posts: 8356
Joined: Wed May 15, 2002 1:31 pm
Location: Gubbängen, Stockholm, Sweden
Contact:

Post by mattias »

jessh wrote:'Traditionally 1/500" circle of confusion was used for 35mm filming, and 1/1000" was used for 16mm'(Appendix B)

The reason that a smaller circle of confusion is used with 16mm than 35mm is that ther 16mm frame has to be magnified more in order to cover the same space.
thanks, that's what i thought. however, these numbers might actually prove that format doesn't matter, since such a small circle is smaller than what the emulsion can register. given a resolving power of 120 lines/mm the smallest circle that can be registered is 1/600", so who cares if details even smaller are sharp or not? ;-) perhaps using 1/1000" is enough to be "safe" and to take account for the uneven graininess which as we know can increase the sharpness slightly.

/matt
User avatar
MovieStuff
Posts: 6135
Joined: Wed May 01, 2002 1:07 am
Real name: Roger Evans
Location: Kerrville, Texas
Contact:

Post by MovieStuff »

mattias wrote:these numbers might actually prove that format doesn't matter, since such a small circle is smaller than what the emulsion can register.
Yes, I would tend to agree. Again, as I mentioned in a previous post, "apparent" depth of field can be affected by a variety of outside influences not taken into account in basic depth of field charts. However, from a practical standpoint, I still feel smaller formats like S8 still have a functional DOF advantage over larger formats for most shooting situations, if one wants deeper depth of field. Having done some foreground miniatures in both formats, I can tell you that doing so in smaller formats is quite easy while maintaining enough depth of field in 35mm is a bit of a challenge. That said, one of the things that does help is when the resolution of the film is so limited that differences in COC size become moot and areas of the film not as sharp as others all tend to look the same due to limitations in film resolution and not so much lens resolution. This only becomes really critical if the image is blown up too much and it becomes apparent that parts of the image are not as sharp as they look when viewed small.

For that reason, S8 tends to appear as if it has hyper-depth of field that often defies the numbers on depth of field charts, since S8 is generally viewed small. Even when viewed larger, the distance of the audience is generally such that the field of view is reduced back to the apparent smaller size.

Roger
User avatar
S8 Booster
Posts: 5857
Joined: Mon May 06, 2002 11:49 pm
Real name: Super Octa Booster
Location: Yeah, it IS the real thing not the Fooleywood Crapitfied Wannabe Copy..
Contact:

Post by S8 Booster »

Some general info on DOF from Kodak WEB:
http://www.kodak.com/US/en/motion/story/fact3.shtml
Depth of Field

Depth of field is the measurement of the area in front of and behind the subject that is held in acceptable focus. You can get a simple understanding of the phenomenon by holding your hand quite close in front of your face – and then focusing your eyes on your hand. Notice the quality of the background. It is very soft and out-of-focus. Now slowly move your hand away from your face, continuing to focus your eyes on that hand. Notice that, as your hand moves away from your face, the background comes into sharper focus.

This is a demonstration of one of the factors, distance to subject, that determines the depth of field of a given shot. There are two other variables that the photographer or cinematographer controls – focal length and f or t stop. And then, an inherent characteristic of any given imaging system is the actual size of the aerial image as it converges on the focal plane. This determines the “resting” depth of field. For example, a larger image (like you get with 65mm film) has very shallow depth of field while a relatively small image (like that of Super 8 film) has much greater depth of field.

But, overall film capture systems have a naturally shallow depth of field that can be easily manipulated. This is extremely important in terms of trying to focus the viewers’ attention to different areas in a scene.

The 2/3" chip size of current digital video equipment results in a smaller capture area than that of film, giving it an inherently deep depth of field – which is not easy to manipulate. That’s usually how you can tell the difference between reality-based television programming, for example, and dramas, sitcoms, TV movies or high-end commercials.
R
..tnx for reminding me Michael Lehnert.... or Santo or.... cinematography.com super8 - the forum of Rednex, Wannabees and Pretenders...
tim
Posts: 422
Joined: Mon May 06, 2002 8:38 am
Location: Norfolk, UK
Contact:

Post by tim »

I am sure that we have gone through this before on this Forum, but here goes anyway.

A lens does not form a point image of a point source, instead about 84% of the light is concentrated into a disk whose diameter (d) depends on the wavelength of the light (l) and the aperture of the lens (A):

d = 1.22xlxA Eq. (1)

This is known as the Airy disk. Lenses for cine cameras have aperture that vary from 1 (very fast) to 22 (slow).
The smallest Airy disk is obtained with a fast lens and the shortest wavelength of light. For an f/1 lens and blue light, the Airy disk is 0.5 micrometer diameter. For an f/16 lens and red light, the Airy disk is 14 micrometer diameter. Note that very few camera lenses are this good.

As we know from earlier discussions, K40 film will reproduce about 1000 lines across a S8 frame. This means that the smallest disk that can be recorded is about 6 micrometer diameter, even if there is no halation.

The image of a point source not in focus is a disk with a fairly sharp edge. Ideally, the depth of focus of a lens is taken to be that for which the out-of-focus image of a point is no greater than the diameter of the Airy disk. To simplify calculation, the permissible diameter of this out-of-focus image is taken as a fixed amount, often f/1000. For a typical S8 camera zoom lens, depending on setting, f/1000 is between 7 micrometer and 40 micrometer. (Longer or shorter lenses will give slightly larger or smaller values.) Taking a usual setting of about 12 mm focal length, f/1000 is 12 micrometer. This is rather greater than the minimum Airy disk diameter calculated above, but slightly smaller than the maximum diameter. However, it is only twice the minimum image size for K40.

To take Airy disk diameter as the diameter of the circle of confusion would greatly complicate the calculation and presentation of depth of field. However, the result is simply that focusing is slightly more critical with lenses whose aperture is between f/1 and f/8 than the tables suggest. Allowing for the imperfections of the optics and camera/projector system generally, it probably doesn't matter anyway!
Lucas Lightfeat
Posts: 716
Joined: Tue Dec 03, 2002 1:09 am
Location: London, England
Contact:

Post by Lucas Lightfeat »

Hi Mattias

With respect, I do have a clue what you're talking about. I understand the mechanics and the theory of all that you have spoken about.

You asked a simple question, because you were unsure, contradicting what you had previously said. I restated to you a basic truth of cinematography, only because you were getting yourself tangled up in theoretic considerations which do not really challenge the basic rules of depth of field. I understand about the circle of confusion and image size considerations, but they do not change the simple fact that 35mm has a shallower dof than Super8 under all but the most unreasonable conditions, and certainly on a TV screen (where most of us view our finished films)

I did, you should note, state that I could barely believe you had gotten yourself so unsure about such a simple thing, but that's exactly what it sounded like - you asked if Super8 actually had a smaller dof than 35 - well, no, it doesn't, which I'm sure is the conclusion you will come to having theorised this topic to its natural conclusion. I in no way meant to bring into question your omniscience in the field of cine ;) I think Jessh explains it well, and all said and done, Super8 produces a very deep depth of field because it uses wider lenses, and I'll stand by that, because regardless of the scientific definitions, this is the practical consideration we need when making films and analysis of the coc will never change this: As you say, the 16mm table works well, when extended into the Super8 wider focal lengths. This is exactly what I learnt at film school, where I studied cinematography.

No hard feelings here,

Lucas
mattias
Posts: 8356
Joined: Wed May 15, 2002 1:31 pm
Location: Gubbängen, Stockholm, Sweden
Contact:

Post by mattias »

Lucas Lightfeat wrote:I restated to you a basic truth of cinematography
but it's not a basic truth, and it's not even a truth. you're talking about something else than the rest of us.
but they do not change the simple fact that 35mm has a shallower dof than Super8 under all but the most unreasonable conditions
you're talking about a difference at the same field of view, and not at the same focal length like the rest of us. everybody knows that 35 mm has less dof than s8 at the same field of view, but we're lightyears beyond that in this discussion. the question is whether you can use a dof chart for a larger format for super-8.
you asked if Super8 actually had a smaller dof than 35 - well, no, it doesn't
yes it does, which three of the later posts in this thread have clearly proven. i must say i find your disrespect for logic and theory quite stunning.

/matt
jessh
Posts: 512
Joined: Fri May 10, 2002 5:10 am
Location: Austin, Tx, USA
Contact:

Post by jessh »

mattias wrote:given a resolving power of 120 lines/mm the smallest circle that can be registered is 1/600", so who cares if details even smaller are sharp or not? ;-)
where are you getting this number? My calculations are saying that for 120lines/mm the smallest point would be 1/3048"

To get this number I am simply conveting 1/120 of a mm to inches by dividing it by 25.4 (1inch=25.4mm) so I am doing 1/((1/120)/25.4). My number sounds about right to me, but please feel free to prove me wrong :-)

Assuming my caluclations to be correct, you would want to use 1/2000" if you planned on blowing up to 35mm, but you can probably get away with 1/1000" if you just plan on viewing on a TV or small screen. Although you may want to use 1/2000" to give you a larger margin of error.

It seems that with newer film stocks and higher res lenses the standards for circle of confusion have gotten smaller. "With newer, sharper lenses and film stocks, sometimes a smaller circle of confusion is used. David Samuelson suggests that 1/1000" or 1/1400" should be suitable for most modern lenses and stocks without any diffusion" (The Filmmakers Handbook, in reference to 35mm) and the palm program I use (pCAM) uses 1/1000"(25.4microns) for 35mm and 1/1667"(15.24microns) for 16mm. I am currently trying to decide what circle of confusion to use with it, and I think I am going to go with 1/2000"

~Jess
Lucas Lightfeat
Posts: 716
Joined: Tue Dec 03, 2002 1:09 am
Location: London, England
Contact:

Post by Lucas Lightfeat »

I believe that this is just semantics. The past few posts have actually shown that it all depends on the size of the circle of confusion used, which is dependant on the filmstock chosen as well as the format/ screening magnification of that circle you choose to impose on the equation. These are arbitrary factors, and will be different for each user.

I will continue to use the 16mm chart with it's 1/1000" coc. I believe this provides a margin of error, and is workable for Super8 because of the smaller projection. Who projects onto a 40' screen anyway? Not I.

For reference, the Schneider chart posted previously has a coc of 1/1270" (0.02mm)

I used 1/50 x 1/25.4 = 1/1270 to get this.

Lucas
User avatar
S8 Booster
Posts: 5857
Joined: Mon May 06, 2002 11:49 pm
Real name: Super Octa Booster
Location: Yeah, it IS the real thing not the Fooleywood Crapitfied Wannabe Copy..
Contact:

Post by S8 Booster »

Seen this?: http://tangentsoft.net/fcalc/

Do not cover S8 by default but seem like S8 preferences may be set.
f/Calc is a calculator for photographic formulas. These formulas are at the heart of many interesting questions about the technical side of photography. You could compute these functions by hand, but they range from tedious to difficult, even with a good scientific calculator. When you let f/Calc do the hard work, you gain an intuitive feel for what the results mean, because you can quickly try new input values to see how they affect the result.
Apart:
Should be fairly simple to load the data into a spreadsheet to calculate it all for all formats to verify the differences or similarities between teh formats.

R
..tnx for reminding me Michael Lehnert.... or Santo or.... cinematography.com super8 - the forum of Rednex, Wannabees and Pretenders...
tim
Posts: 422
Joined: Mon May 06, 2002 8:38 am
Location: Norfolk, UK
Contact:

Post by tim »

Don't confuse them with facts!
User avatar
S8 Booster
Posts: 5857
Joined: Mon May 06, 2002 11:49 pm
Real name: Super Octa Booster
Location: Yeah, it IS the real thing not the Fooleywood Crapitfied Wannabe Copy..
Contact:

Post by S8 Booster »

:oops: Sorry Tim. I´ll never do it again :!:

R
..tnx for reminding me Michael Lehnert.... or Santo or.... cinematography.com super8 - the forum of Rednex, Wannabees and Pretenders...
Post Reply