The Ultimate Depth of Field Skinny

[Jeff Donald]

I stand corrected, I did mention Coc first, because of the Panavision use of non standard Coc values for some formats. My apologies to you and Jay.

I still stand by my assertion that many members and guests have an interest in DOF and Coc. Many are film photographers and while they can't apply all of this to video, it is very applicable to their traditional film photography and darkroom work. This is quite obvious by the number of views this thread is getting and the number of emails I'm receiving.

I've not commented on the link I asked for because it is a lengthy article and I could not easily find the quote. I can't find where Alsobrook is talking about Arricam. When I have more time I'll read the article. Sorry.

The Kelly-Wheel uses a fixed Coc, like the Panavision site.

All five factors are listed in the DOF Skinny thread and DV Info article.

The Coc value changes with the size of CCD or negative used. Once established for your size CCD the value will remain fixed. You have noted the larger format cameras have less DOF. Why? They have a different value Coc.

I thought your asking of Jason's original question was a sort of rhetorical question or something on your part. My answer to Jason's original question is contained in previous posts. Jason stated he would experiment with the suggestions and post back results and further questions. Now I'm pretty much trying to help you understand it.

I previously mentioned that a great many articles and web sites have errors and half truths. This is evidenced by your continued reference to these sites that list only 3 factors for DOF. I thought the American Cinematographers Manual, 8th edition (published by the America Society of Cinematographers) and their published formula would be good enough for you. Their formula requires five values.
1. focal length of the lens
2. f-stop number (size)
3. Coc
4. distance from camera to object (subject)
5. Hyperfocal distance (derived from first 3 values)

In the article and Ultimate thread I list the fifth value as "The viewer's personal standard of the permissible degree of sharpness (or unsharpness)". This is after all a subjective medium. I don't argue sharpness. If you insist your image is in focus, who am I to argue with you.

[Wayne Orr]

I'm sure that would get me a lot of work. Yeah.

Jeff, you and I are never going to walk down the aisle. I think you are obfuscating a subject that to people using camcorders should not be so complicated, based on my considerable experience. It would be totally out of the realm of reality to expect someone with a GL1 shooting a "movie" to sit down and "do the math" in the formula for depth of field you refer to, everytime they set up a shot.

It also does not answer the original question, "how do I blur the background?"

Wayne Orr
Caveat canem

[John Jay]

<<<-- Originally posted by Wayne Orr : Again, my point is that narrow depth of field is impossible to achieve with small chip cameras under normal conditions, so rather than fighting it, work with it and use other methods to give separation to your background. -->>>


Wayne,

you asked for some frames....

Technical Data

f1.6
shutter 425
subject distance 1.6m
background distance 6.3m
Optical zoom 50% approx 22mm
Axe : Sony TRV950E
Ambient temperature 16 C
Weather Cloudy Dull
Frame Rate 50i

Procedure - record to DV then grab stills to Memory stick 640x480 then upload to yahoo briefcase as below

DOF examples of background blur (dofbg) and foreground blur (doffg) found here

http://photos.yahoo.com/bc/agentpurp...c=ph%26.view=t

http://photos.yahoo.com/bc/agentpurp...c=ph%26.view=t

[Wayne Orr]

John, I hardly know what to say. I guess then, your advice to someone looking for the blurry background would be:
Shoot wide open, at 22mm, at shutter speed of 425 (never mind the jerky motion), make sure your background is at least 20 away, and your subject is the size of a liter of Coke, and you will end up with a somewhat blurry background. Perfect.

May I recommend to you the movie, "The Patriot," starring Mel Gibson. I think you will find some of the blurry background pictures in that film are more in keeping with what Jason had in mind. And no motion effects from the hi speed shutter used in your pix.

Sorry, John, no sale.

Wayne Orr
Veni, vidi, vici

The way I do it is like this... it's very technical, so try to stay with me.

Okay. I point the camera at the subject and fuss around with the lens and exposure settings. If I get it to work, cool. If I can't, I blame it on poor lighting or Canon... sometimes both.

[John Jay]

<<<-- Originally posted by Wayne Orr : Again, my point is that narrow depth of field is impossible to achieve with small chip cameras under normal conditions, so rather than fighting it, work with it and use other methods to give separation to your background. -->>>

Wayne,

I objected to your statement above and have provided two real world examples.

[Chris Hurd]

Members are reminded to review our FAQ regarding disagreements with other members.

Folks -- please note: In a few of these preceding posts, I've had to edit out a few comments that were somewhat nasty... please proceed with this debate, but do so on an elevated level like the mature adults we all are. I would like to think that we can disagree with each other but still buy each other the round of beer. This place is for friendly discussions and debates, or at least it will have that *appearance* but I would rather not have to baby-sit. Let's please relax, keep the thread open, and proceed with cool heads and warm, fuzzy feelings for each other... or else! Much respect,

[Jeff Donald]

Wayne, I'm not sure what your trying to say

Shoot wide open, at 22mm,

An inch is a little more than 22mm. John says the distance is 1.6mm or a little over 5 feet. A little close, but reasonable for an interview. The subject size is fine for product shots and similar type work that I do. For an interview the subject size is too large and would require either less zoom or increase the camera to subject distance (either will reduce subject size and increase DOF).

Other than for Ken's last post, I get the feeling you guys are leading Jason to believe that there is actually something he can do to create a narrow depth of field with a quarter inch chip camera, and that simply is not going to happen in a real world situation. (I am not talking about shooting extreme close-ups of the head of a coin)

John has used a Sony TRV950e and it uses 1/4.7 inch (.21 inch) chips. The DOF is about 19% greater with John's camera than the 1/4 inch chips in Jason's GL2. By using the Panavision Calculator for the 1/4 inch chips we see the DOF for Jason's camera is about 6 inches front to rear. Using the same information and changing the CCD size to 1/3 inch (XL1 and others) the DOF drops to under 3 inches. John used a real world situation that was not the head of a coin. He met your challenge. Live with it.

No one is implying or stating that 1/4 inch, 1/3 inch or even 2/3 inch CCD's will achieve the DOF of a 35mm or even a 16mm film camera. But just because you don't have the same tools, it doesn't mean that you give up. I am suggesting that people learn the whole truth about DOF and use it as best they can to their advantage.

The shutter speed argument is irrelevant. Shutter speed can be controlled through the use of ND filters. Neither John, nor his example should be criticized for his lack of owning or using ND filters to control shutter speed. Someone with considerable experience would realize that.

It would be totally out of the realm of reality to expect someone with a GL1 shooting a "movie" to sit down and "do the math" in the formula for depth of field you refer to, every time they set up a shot.

As I previously stated this type of shooting does not lend itself to run and gun or ENG style shooting. The formula is too complicated for you to use your slide rule. That's why they make Kelly-Wheels, DOF charts and Palm programs. They use the formula and the user inputs the non fixed, variable data. It takes only a few seconds to input the data. I know many camera operators that use Palms (PDA's) on the set for just such a purpose.

I think you are obfuscating a subject that to people using camcorders should not be so complicated

I'm sorry you think I am clouding the subject. Quite to the contrary, from reader mail, I would say I'm clearing up and making DOF and Coc clearer for many people. The vast majority of my students understand the principles, concepts and practical uses in less than an hour. The actual implementation of the principles is a life long pursuit and study.

Caveat canem, can you explain why I should beware the dog?

[John Jay]

Jeff

Thx for wrangling this one!

Just a few points

The 950 was used to prove the point that even small chips can get slim DOF - useful for cutaway shots to spice up your edits

Also, I have many ND filters, but since it was a quick n dirty test it was easier to ramp the speed to fix the exposure issue.


Also heres a nice duck shot I found from J Beale site, shot with his DVX100

http://www.bealecorner.com/dvx100/frames/duck01.jpg

[Jeff Donald]

Also, I have many ND filters, but since it was a quick n dirty test it was easier to ramp the speed to fix the exposure issue.

I assumed as much, thanks for the clarification.

[Wayne Orr]

Jeff

Since you are having trouble understanding what I have to say, I will try to explain in more detail.
John said he used the "optical zoom 50%, approx 22mm. That is, his focal length was 22mm, approximately, since these cameras do not have exact focal length markings, which also makes it difficult to calculate exact depth of field. With me so far?

Extrapolating his information to my PD150, 50% focal length would be approx. 33mm. We will use the f/1.4, although in the real world, the PD150 only opens to f/1.6. Entering the same subject distance of 5.2ft (1.6m), we use the Panavision calculator (nice to see you using it, although we know it is flawed) and arrive at a depth of field of 2.82 inches. Hey, we agree. We also agree that the "field of view" this represents is way too restricted for "normal" shooting conditions. That field of view is, 9.162 inches across by 6.8 inches high, or, about the size of a "liter of Coke," as illustrated by John's picture. Doesn't leave much room for head and shoulders, does it?

Hooray. We now have narrow depth of field with a small chip camera. How did we do that? In this case, by opening up to f/1.4. Easy? No. In John's pictures he had to increase the shutter speed to 425 to eat up enough light to get to f/1.4, which is fine for still pix, but results in undesireable motion characteristics when people and cameras move. "The shutter speed argument is irrelevant. Shutter speed can be controlled through the use of ND filters." Excuse me, its not the shutter speed we are trying to control, its the aperature. John chose to control the aperture by increasing the shutter speed; what you mean to say is, control the aperture by adding ND's in front of the lens. How much additional ND? I don't know off hand. And it will vary depending on the lighting conditions. And actually I don't care because I feel that John's pictures support my contention. What?

You quoted me: "I get the feeling you guys are leading Jason to believe that there is actually something he can do to create a narrow depth of field with a quarter inch chip camera, and that simply is not going to happen in a real world situation." Yes, I said that (seems like weeks ago) and I still stand by it. In John's photos, despite using an extremely narrow angle of view (6 degrees approx), which as has been pointed out would not contain a head and shoulders shot, despite the lens being set at maximum aperture, and despite the background being at least 20 feet away, the best John was able to produce, was a modestly blurred background. Because the depth of field is so great with these cameras, the focus will soften slowly over distance. And you are telling me to "live with it"? Don't be so arrogant, Jeff, that dog might bite you in the a**.

But in the end, it's up to you people who are being entertained (and maybe enlightened just a bit) to make the final decision. You can shoot your films outdoors with long lenses and distant backgrounds, trying to get those elusive blurry backgrounds, or, you can spend your creative energies working on composition, lighting, art direction, and other techniques that will improve your images, working within the limits of the medium.

I wish you all the very best.

Wayne Orr
Dominus Vobiscum

[Chris Hurd]

Bob Harotunian:

"I have a couple of DOF frames but need some advice on how to post them."

Sorry for the delay in getting back to you. Just email them to me, and I'll host them for you from the web site and post the link to them.

[John Jay]

Jeff,

If one is looking for more real estate in a Slim DOF shot, may I suggest the use of a full height anamorphic adapter. I have upped a shot to illustrate this at

http://photos.yahoo.com/bc/agentpurp...c=ph%26.view=t

The anamorphic adapter works like an optical compander and has no net effect on DOF but does have the advantage of a 33% increase in image width. Thus head and shoulder shots are possible with differential focus from the background.

If one is unhappy with the degree of blur in the background a matte can be created using an unsharp mask to allow anything to be done to the background in post.

The real issue here is that there has to be some differential focus sufficient to give a definite threshold for the unsharp mask to determine a sharpness silhouette, after which it is a matter of a few mouse clicks and some render time - this is well within the capability of small chip cams and is a far cheaper solution to renting arricam/ moviecam, a daVinci best light, stock plus processing and maybe a colourist.

Another low cost approach would be green/blue screen or a difference matte for non-moving backgrounds

[Wayne Orr]

OK. Now that I have everyone thinking I am against blurry backgrounds, I am going to show you some of my stills with blurry backgrounds, taken from a little project I did last year. They have attached comments.

http://www.digitalprods.com/DofF.jpg

I'll take my turn in the barrel, so feel free to take your best shot. So to speak.

[Rob Lohman]

Those shots look really nice Wayne!