Managing depth of field for a low end camera

[Todd Terry]

[Bob Cole] “You’ll get a narrower slice of the background, but it will be virtually as sharp”

Sorry, Bob… not to be argumentative, but that simply is not true. If you, say, shoot a subject 20 feet away with an 80mm lens, and then shoot the subject from 5 feet away with an 18mm lens (or whatever exact distances it takes to keep the subject the same size), not only will the field of view be different, but the degree of focus on the background (or foreground) elements will be radically different, assuming the same f-stop for both lenses. Yup, RADICALLY. It’s just the way it is. The same would apply when using different focal lengths for a zoom lens. This is, of course, assuming you are shooting a shallow DoF format (such as 35mm film, or video with a lens converter)… a little 1/3″ camera has such deeeeeep DoF almost everything is going to be in nearly-sharp focus no matter what you do.

I shoot every day (and have for the last 20 years) with camera formats ranging from Super35mm film down to video cameras with little 1/3″ sensors, and can attest that this is simply the way it is. If I had the inclination I could probably spend an hour on the phone and easily find dozens of cinematographers to back that up.

You can test this easily yourself… probably NOT with a small video camera since the depth of field is so deep, but if you don’t have access to a 35mm motion picture camera you can test it with an SLR with a zoom lens (NOT a digital SLR, it will suffer the same “small chip” deep focus as the video camera… use a real 35mm film SLR). The change in depth of field should be readily evident, just by looking through the viewfinder. You don’t even have to take any shots.

If you can’t do a test, DoF/f-stop/focal length comparison charts are readily available… I think there is one in the American Cinematographers Manual (I don’t have mine handy, it’s at the studio and it’s after hours now).

[Bob Cole] “What I’m wondering about right now (as a Sony Z1 owner), is whether there is enough of a DOF difference to make a difference, between the 1/3″ imager and a 1/2″ (as in the new Sony EX).”

Nah, probably not. If you want to compare it to film, your Z1 with its 1/3″ chips has roughly the same DoF as Super8 film (very deep). A “full size” camcorder (Cinealta, Varicam, etc.) wit 2/3″ sensors will have roughly the same DoF as 16mm film (somewhat shallower than 8mm, but much deeper than 35mm). The EX with its 1/2″ chip will fall somewhere between 8mm and 16mm… closer to the 8mm DoF… in other words, still extremely deep. I doubt there will be any noticable DoF difference in the EX compared to 1/3″ cameras.

Personally I would love to use te EX with a lens converter and cine primes. However, I really only like to use that setup with cameras with removable lenses…. and unfortunately to take off the EX lens it requires a hacksaw. And I’m betting the results wouldn’t be too good!

T2

__________________________________
Todd Terry
Creative Director
Fantastic Plastic Entertainment, Inc.
fantasticplastic.com

[Bob Cole]

re: DOF. Please take a look at the links in the earlier post. Once more: https://www.cambridgeincolour.com/tutorials/depth-of-field.htm Navigate to the section headed “CLARIFICATION: FOCAL LENGTH AND DEPTH OF FIELD”

There’s no point in continuing with this he said-she said — at least for my part, as I’m not an expert in the physics of optics and have nothing to contribute except for those links. I was just trying to keep someone from believing that he could fix a 1/3″ camera’s shallow DOF by moving the camera back and using the tele.

But at least this discussion has led to another question, which I’m going to post in separate thread.

Thanks!

“Cole-out”

MacPro 2 x 3GHz dualcore; 10 GB 667MHz
Kona LHe
Sony HDV Z1
Sony HDV M25U
HD-Connect MI
Betacam UVW1800
DVCPro AJ-D650

[Todd Terry]

Well, I won’t continue the “he said / she said” argument past this post either (while wondering who is the “he” and who is the “she”)…

But I will observe that I did look at that guy’s website. I will say that he does appear (just looking at the gallery) a pretty darn good photographer. But judging from his technical explainations he is also one that has a fairly, um, unusual grasp on some of the technicalities of photography.

The simple fact is that anyone who says that DoF is unaffected by focal length is simply flying in the face of well over a century’s worth of proven photographic and optical theory. No two ways around it.

Just keep in mind that in this day and age anyone can throw up a website… but that doesn’t mean the information in it is correct. I haven’t looked, but I’ll bet you ten bucks that within ten minutes of Googling I could find some sites purporting that Elvis is still alive and that professional wrestling is real.

If someone wants to believe that however, well, they can knock themselves out.

I think our UK photographer is partially basing his observations on the fact that he seems primarily to be a digital photographer, where small imagers result in DoFs that seem less affected by focal length change.

[Bob Cole] “I was just trying to keep someone from believing that he could fix a 1/3″ camera’s shallow DOF by moving the camera back and using the tele”

But that would be trying to keep someone from believing something that is provably true. You can indeed reduce DoF by using longer lenses. You may not get the field of view you want (probably won’t) and perspectives may be flattened in an unwanted way… but the DoF will definitely be reduced.

T2

__________________________________
Todd Terry
Creative Director
Fantastic Plastic Entertainment, Inc.
fantasticplastic.com

[Boyd Mccollum]

There are really only 4 factors in depth of field for the same camera (film/sensor size) based on specific constants (there are also specific mathematical formulae to describe these various relationships):

1. At any given focal length, DOF increases as the aperture gets smaller.
2. At the same aperture and focal length, DOF increases as the distance to the subject increases.
3. At the same aperture and distance to subject, DOF is greater for a shorter focal length than for a longer one.
4. At the same aperture and image size (the subject framing remaining constant), DOF remains approximately constant for all focal lengths.

In this thread, it appears Bob is talking about situation number 4, whereas Todd is talking about number 3. Both are making valid points.

The situation many people encounter is too much DOF for the framing/composition they want – for example, in an interview. The common advice given is to move the camera back and zoom in. However, when you do this, to achieve the same frame composition, you have to move the camera back to a point where the DOF of the new focal length matches what you previously had. This is demonstrative phenomenon.

However, using a longer focal length can help with the illusion of a shallower DOF by narrowing the field of view and magnifying the background blur of the image (making the background larger in relationship to the foreground).

An excellent article with specific photographic examples and reference material can be found at:

https://www.vanwalree.com/optics/dof.html

So yes, focal length does affect DOF. But there are areas where you achieve the same DOF with different focal lengths – same subject framing being one of them.

[Boyd Mccollum]

There are really only 4 factors in depth of field for the same camera (film/sensor size) based on specific constants (there are also specific mathematical formulae to describe these various relationships):

1. At any given focal length, DOF increases as the aperture gets smaller.
2. At the same aperture and focal length, DOF increases as the distance to the subject increases.
3. At the same aperture and distance to subject, DOF is greater for a shorter focal length than for a longer one.
4. At the same aperture and image size (the subject framing remaining constant), DOF remains approximately constant for all focal lengths.

In this thread, it appears Bob is talking about situation number 4, whereas Todd is talking about number 3. Both are making valid points.

The situation many people encounter is too much DOF for the framing/composition they want – for example, in an interview. The common advice given is to move the camera back and zoom in. However, when you do this, to achieve the same frame composition, you have to move the camera back to a point where the DOF of the new focal length matches what you previously had. This is demonstrative phenomenon.

However, using a longer focal length can help with the illusion of a shallower DOF by narrowing the field of view and magnifying the background blur of the image (making the background larger in relationship to the foreground).

An excellent article with specific photographic examples and reference material can be found at:

https://www.vanwalree.com/optics/dof.html

So yes, focal length does affect DOF. But there are areas where you achieve the same DOF with different focal lengths – same subject framing being one of them.

[Nate Graham]

I helped a crew shoot a short this summer. They used a redrock converter and I must say that there is something very appealing about narrow DoF. Eventually they took the redrock off because they were having “focus” problems. But I must admit that while having out of focus backgrounds looks very nice, it doesn’t make a video look any more “like film” but it definitely doesn’t look so much like video either. It might taste like chicken but it still just isn’t quite chicken somehow.

Todd,
Your talk about only using quality lenses and converters raises a question (begging your pardon for my lack of knowledge on the subject). What makes a good lens, well, good?

My father is a still photographer and he’s always talking about this lens and that lens. I finally asked him what makes these high quality lenses more desirable than others. He admitted that he didn’t really know. He just goes off of what his piers say is good, but they never really say why, either. I have yet to see someone look at a photograph, film, video, etc. and say “WOW, this must have been shot with (insert lens model here)” I can understand distortion is factor on wide lenses but what other things does one look for in quality lenses?

The best lenses are always primes, too. Why? Is there no such thing as a really good zoom lens?

Thanks to all of you who so patiently share your knowledge with those of us who are not as learned.

Nate

[Todd Terry]

[Nate Graham] “What makes a good lens, well, good?”

The only thing that makes a lens “good” is whether it shoots good images to your satisfaction.

That being said, cheaper lenses are generally mass produced, without the same attention to detail and quality control that that is available in high-end cinema lenses. High-end cine lenses are typically sharper and somewhat more contrasty than inexpensive lenses. High end cine lenses are designed to “resolve” an image at a higher quality (since they were designed for actual film), whereas video cameras typically have lenses that will resolve only for video-level resolution, if you’re lucky (be it SD or HD). Motion picture lenses are generally made/assembled by hand by guys you would more consider craftsmen…. the optical elements, coatings,
machining and all of that will be surperior. Of course the lack of mass production comes with a price… they can be very expensive. Often times on a real film set the lenses can be the most costly thing around… a single prime can be worth more than the camera itself. A matched set of primes most certainly will be.

Some cinematographers really do know their lenses well, and can tell just by looking that such-and-such movie must have been shot with a particular model of Cookes because it’s more contrasty, or that film must have been shot with so-and-so model of Zeiss because it is wamer… etc. Most of us can’t tell.

The “best” lenses are generally primes, largly because the more glass an image has to go through, the more the image degrades. Primes have just a few elements in them, whereas zooms have a whole boatload. Zooms typically are much slower than primes as well. That’s not to say there are not good zooms, as there are some out there. Really good ones are quite costly though.

A great deal of it has to do with fit/finishing/machining, too, not just the optical elements themselves. Many people try to shoot video with DoF converters and use SLR still lenses, and readily learn the difference in inexpensive lenses and higer-end cinema lenses (although some people still do it with great success). Firstly, the “fit” on the SLR lenses (even good ones) do not have nearly the tolerance of cine lenses. Some even “wiggle” a bit… which is inconsequential for still photography, but a killer when shooting motion footage. Many SLR lenses, even expensive ones, breath a bit (the image size changes slightly when focusing), whereas good cine lenses should never breath. They can be difficult to use, too, especially for racking or follow focusing. This is because to go from near to infinity on an SLR lens you only have to turn the barrel a tiny bit, whereas a cine lens will make almost an entire barrel turn to go through its focus range.

A lot of still lenses DO have excellent elements in them, it’s just that they were not mechanically designed for cine use. For example, four of the superspeed lenses in my primes set are Leitz lenses, and started their life as lenses for Leica still cameras. The actual glass in them is second to none, but they would not have been useful for motion work because of the reasons I mentioned. However the optical elements were rehoused by Panavision, and now they have the right mounts, gear rings, focus travel range, collimation, engraving, and all that good stuff… so even though they were once still lenses they are now as good as any cine lenses that started life that way.

I could say a lot more, but those are the basic points. Again, what makes a “good” lens is if it looks good to you. And price isn’t always a factor… some excellent primes will easily cost as much as a new car, but I also have some excellent Russian LOMO cine primes that were dirt cheap (maybe only a few hundred bucks a piece) and they give stunning results as well.

It seems a lot of people are catching onto the fact that using cine primes on an HD camera (often with a DoF converter) can give stunning results. I’m basing that guess on the fact that the market for cine primes has really dried up. A few years ago you could sometimes find a decent small set of superspeed primes in the $10-15K neighborhood. Today unfortunately you can easily pay twice that or more, if you can find them. If and when RED finally releases their primes they will go for about $20K/set, which might be worth looking at. Incidentally, they will also have an 18-50mm zoom for $6500 and a 50-150mm for $8500. They are both T3… which is reasonably fast for a zoom, but pretty slow compared to primes.

There are two test instruments that you should use to make your final and most critical judgment….. your eyeballs.

T2

__________________________________
Todd Terry
Creative Director
Fantastic Plastic Entertainment, Inc.
fantasticplastic.com

[Nate Graham]

Wow, that was an awesome post, thanks Todd!

One other question. I’ve noticed that on The Hunt for Red October every time they do a rack focus there’s an odd sort of distortion. Other films don’t seem to have this same effect when doing a rack focus. Do you know why it looks so different on this film?

Thanks again for the info. You should do a COW video on lenses. I’d be interested in learning more.

Nate

[Todd Terry]

[Nate Graham] “I’ve noticed that on The Hunt for Red October every time they do a rack focus there’s an odd sort of distortion.”

Hmmmmm…. dunno what that was. Haven’t see that film in a while, but next time I catch it I’ll look out for that. Maybe the lenses were breathing, maybe they specifically selected a vintage set of primes (probably anamorphics) for their certain “look,” which happens often. If I recall, the DP on “Red October” was Jan de Bont, and he certainly knows what he is doing so anything that happens on screen was probably done on purpose by design.

T2

__________________________________
Todd Terry
Creative Director
Fantastic Plastic Entertainment, Inc.
fantasticplastic.com

[Steve Wargo]

[Todd at Fantastic Plastic] “professional wrestling is real.”

So that’s why they don’t have the results in the newspaper and on the TV. Darn!

Steve Wargo
Tempe, Arizona
It’s a dry heat!

Sony HDCAM F-900 & HDW-2000/1 deck
5 Final Cut (not quite PRO) systems
Sony HVR-M25 HDV deck
Sony EX-1 on the way.