Video - depth of field

Gav O Reilly · May 4, 2005

I've started to shoot more on video than on film recently and i'm trying to get around some of those typical DV differences.

Can anyone explain the science behind the wider depth of field and level of detail in video?

Thanks,

Kevin Zanit · May 4, 2005

This subject has been dealt with in great detail in the past. Do a search.

Simply, DOF is a function of physics: Focal length, aperture, and imager size (be it 2/3", 1/3", 35mm, 16mm, etc.).

The bigger the aperture opening (lower F-stop) the less DOF (shallower)

The bigger the imager size, the less DOF

The longer the focal length, the less DOF

All three of these items are what account for DOF.

In the case of a video camera, it generally will have a smaller imager, thus it will have more DOF (deeper).

There is no real way around this. There are adapters that project an aerial image onto a 35mm ground glass that is then re-photographed that will give you 35mm DOF. These adapters have drawbacks that have been dealt with in the past (search).

All you can do when shooting video (or even 16mm for that matter) to minimize DOF is to shoot as wide open (aperture) as you can, from as far away as you can with a long lens.

Kevin Zanit

Sam Wells · May 4, 2005

The bigger the imager size, the less DOF
The longer the focal length, the less DOF

<{POST_SNAPBACK}>

You can't separate these two. I think it's better to state it as a function of image size than focal length.

-Sam

Patrick Neary · May 4, 2005

well, to split a few hairs here, the size of the imager actually has no bearing on DOF. The shallowing DOF you see in 35 as opposed to minidv is a by-product of using a different focal length lens to achieve the same angle of view.

DOF is blind to format - it is only a result of focal length and aperture.

Edited May 4, 2005 by PatrickNeary

drew_town · May 4, 2005

well, to split a few hairs here, the size of the imager actually has no bearing on DOF. The shallowing DOF you see in 35 as opposed to minidv is a by-product of using a different focal length lens to achieve the same angle of view.

Well that's saying the same thing isn't it? If the image sensor were larger there wouldn't be a shift in focal length, giving you a more shallow DoF. I've never heard it put like that so I won't vouch for its validity. There's plenty of info on Dof in OTHER THREADS THAT ARE EASILY SEARCHABLE.

John Pytlak RIP · May 4, 2005

There's plenty of info on Dof in OTHER THREADS THAT ARE EASILY SEARCHABLE.

<{POST_SNAPBACK}>

Like these: ;)

http://www.nikonlinks.com/unklbil/dof.htm

http://www.svca.org/articles/DigitalDepthofField.htm

http://www.wrotniak.net/photo/dof/

http://home.online.no/~gjon/davisdof.htm

Kevin Zanit · May 4, 2005

For all intents and purposes, the imager size factors into the equation.

He asked what it was bout video that caused the greater DOF, and how he could fix it. The key factor in that problem is the imager size in the video camera vs. 35mm frame size . . .

I did not intend to write a technical journal on the complete physics of DOF, just to give a simple answer, and urge him to search for more details.

Kevin Zanit

Patrick Neary · May 4, 2005

Searching the forums does make for great reading on the subject, but you end up wading through a lot of material for a simple answer...

what would be nice is a clear, concise DOF primer on the FAQ. This Q does come up quite often, and almost every time, the notion of the imager size having something to do with DOF creeps in. I understand why it's in the discussion, but it's not accurate, and we want to be somewhat precise here, right?

In the same way, you could rightly make the argument that DOF is determined by the film's ASA, or the lighting package, or the number of ND's stacked on the lens, when really, those are just ancillary to the actual mechanics of DOF.

The reason I bring this up is that I know for myself, it became much easier to understand the vagaries of DOF when boiled down to it's core, which is not complicated at all; focal length and iris (and you could add the distance the lens is focused at, if you want to get really fancy.)

there, I'm done.

Sam Wells · May 5, 2005

I did not intend to write a technical journal on the complete physics of DOF, just to give a simple answer, and urge him to search for more details.
Kevin Zanit

<{POST_SNAPBACK}>

Fair enough !

..for that matter, anyone who has a 16mm camera and a zoom lens can walk around with it and see how it works ! Or a still camera with DOF preview and a zoom. etc

even a DV camera altho w/o ND's if you don't mind some possible nuclear images...

-Sam

Jon Amerikaner · May 15, 2005

DOF on video is notoriously deep. To get a smaller DOF, more film like, you need to shoot wide open. This will probably mean using the high-speed shutter and or ND filters for daytime photography. Furthermore, this won't visibly affect the DOF on wide shots. You must zoom in (or use longer lenses) to compress the background and foreground to really see the DOF. And since you're zoomed in, be sure to have a tripod to steady your shots.

Stephen Williams · May 16, 2005

This subject has been dealt with in great detail in the past. Do a search.

Simply, DOF is a function of physics: Focal length, aperture, and imager size (be it 2/3", 1/3", 35mm, 16mm, etc.).

The bigger the aperture opening (lower F-stop) the less DOF (shallower)

The bigger the imager size, the less DOF

The longer the focal length, the less DOF

All three of these items are what account for DOF.

Kevin Zanit

<{POST_SNAPBACK}>

Keven,

In David Samuelson book "A Hands-On Manual for Cinematographers", he states. "Depth of field remains the same, regardless of lens focal length, so long as the image size (and f-stop) is the same. There is no point in changing to a shorter focal length lens and moving closer, because if the image size remains the same so will the depth of field." (Focal Press, London, second edition, 1998, p.218)

Stephen Williams DP

Zurich

www.stephenw.com

Kevin Zanit · May 16, 2005

The physics of what everyone is saying is dead on correct.

But . . .

As I said before (in correcting my previous statement), to keep it simple, the imager's size will indirectly effect your DOF. Having a certain imager size will "force" you into a certain focal length.

You are all correct in stating that imager size does not directly effect your DOF (and I accept that), but in practical use, it will (indirectly) ultimately have an impact on your focal length choice, thus it factors greatly into DOF.

Kevin Zanit

Stephen Williams · May 16, 2005

The physics of what everyone is saying is dead on correct.

But . . .

As I said before (in correcting my previous statement), to keep it simple, the imager's size will indirectly effect your DOF. Having a certain imager size will "force" you into a certain focal length.

You are all correct in stating that imager size does not directly effect your DOF (and I accept that), but in practical use, it will (indirectly) ultimately have an impact on your focal length choice, thus it factors greatly into DOF.

Kevin Zanit

<{POST_SNAPBACK}>

Kevin,

I don't understand what you are saying?

I think you accept that a 25mm at 5 feet, a 50mm at 10 feet, a 100mm at 20 feet & a 200mm at 40 feet all have the same DOF at the same F stop & the same size image.

Stephen

May 17, 2005

This is a little off the topic of what it going on hear but This is somthing that I'm getting ready to order I wanted to see what you guys thought. And it is somthing to look at about DOF

Jay Gladwell · May 17, 2005

Ben, yes that will certainly provide the shallow DOF, but it, like the mini35, would be impossible to use for handheld work. Now the Guerilla 35 is another story!

Jay

Laurent Andrieux · May 20, 2005

Gentlemen,

Sorry I'm reading this thread a little late, and sorry to keep tracking it to its "off-topic" way, but, as someone mentionned here, it's been said in the FAQs related topics, that it would be a good idea to have this question in the FAQs section.

Everybody participates to this thesaurus as he wants, and the thing is that after Dominic Case and I proposed to write something, I began to work on this precise topic. (it's in David's hands at the moment).

So... reading this thread makes me think that maybe I should be even more clear and simple about a couple of facts...

I have the strange feeling that, though everybody who has contributed to this thread is certainly not wrong in what he says, it could sound be a bit contradictory sometimes to newbies for instance (does the image size affect DOF yes or no for Christ's sake ?!! :blink: )

In this behave, would you gentlemen please tell me if this is clear (so I can take your reactions in account and maybe change my text and, also, considering english is not my language, you'd help me trying to be clear about some vocabular points... ) :

1) It has to be precised what one refers to when saying "image size" :

if you refer to the dimensions of the format of the "recorded" image (what I would do) - 16 mm, 35 mm, 1/3" or 2/3" video, or if you refer to what I would call the "growth" (that is defined as G=i/o ie image size devided by object size).

Therefore, I would say, as I was tought at school and as I teach my students, that :

- Yes DOF depends on the image size. It does but the way it does depends a lot on the dimensions of the viewed result, ie on the screening size.

When I read that the larger the image is, the less DOF, I cannot agree if I follow my definitions.

16mm is a smaller image size than 35 mm and it has less DOF because if you print it and screen it on a wide screen, it will be more enlarged.

if you refer to what I call the growth of the image, well the comparision of the DOF in diffrent situations could be analysed as follows :

- It is admitted, though some people don't believe it, that at a given size image, growth doesn't affect DOF and therefore, yes, a 25mm at 5 feet, a 50mm at 10 feet, a 100mm at 20 feet & a 200mm at 40 feet all have the same DOF at the same F stop & the same growth, but, as the field of view changes, the DOF on the set, behind

the focus plan changes, because its growth changes...

- As the image size (format) is smaller in 16 mm than in 35, you need a shorter lens to get the same object's size (let's say, the director wants a CU) for instance and, yes, of course, the focal length affects DOF... but you're not having the same growth !

Of course the focal length affects the DOF every other parameters remaining the same, what David Samuelson refers to is a situation where the focal length is changed and the growth remains the same.

The way DOF is affected by the image size (format) is took in account in what is defined as the circle of confusion.

it depends on the screening size. If you shoot 16 mm and compare it on a wide screen with 35mm, it will have less DOF than 35 mm. If you compare the 16 mm on a TV set at 4 or 5 times the diagonal, to the wide screened 35 mm at 1.5 times the diagonal, 16 mm as more DOF !

Laurent Andrieux · May 20, 2005

I figure out, with the help of systran that what you people call image size is what I would call object size (CU, medium shot, etc.) is it so ?

Laurent Andrieux · May 20, 2005

another point, as this is the video section.

I've read here and there comparision beetween 35 mm DOF and video DOF. S35 adaptator, etc. and, mainly saying that video has more DOF than 35 mm.

Would you clear something for me : do people compare the rendering of DOF on a TV set to the typical 35 mm rendering on a wide screen, or both on a same "output", (TV set or wide screen) ?

I mean, with the adaptator, are pple looking for video to have the same DOF as 35 in the same conditions ? If so, I don't understand... a video (1/3'' or 2/3" video image will have less DOF than 35 mm if you considerthe same "output", tv set or wide screen !)

David Mullen ASC · May 20, 2005

another point, as this is the video section.

I've read here and there comparision beetween 35 mm DOF and video DOF. S35 adaptator, etc. and, mainly saying that video has more DOF than 35 mm.

Would you clear something for me : do people compare the rendering of DOF on a TV set to the typical 35 mm rendering on a wide screen, or both on a same "output", (TV set or wide screen) ?

I mean, with the adaptator, are pple looking for video to have the same DOF as 35 in the same conditions ? If so, I don't understand... a video (1/3'' or 2/3" video image will have less DOF than 35 mm if you considerthe same "output", tv set or wide screen !)

<{POST_SNAPBACK}>

You're assuming the same focal length lens is compared in different formats but the circle of confusion is different.

The reason 16mm or 2/3" video has more depth of field than 35mm is because you have to use shorter focal lengths to achieve the same field of view as 35mm, and shorter focal lengths have more depth of field than longer ones. So the imager size is irrelevant EXCEPT in how it determines what focal length you choose. This is a bigger factor (focal length / aperture / distance-focused) in determining depth of field than what circle of confusion you choose.

Personally I don't even bring up circles of confusion because it's, well, confusing. The ASC Manual just lists depth of field info for all formats by focal length (i.e. it assumes one circle of confusion.)