How to get a shallow depth-of-field (DOF)?

dsjtecserv said:

Barrie Davis said:

dsjtecserv said:

No, that one won't work, or rather is only indirectly beneficial. For any given subject framing and f number, depth of field is the same regardless of focal length. The out-of-focus objects will look different because of the different angular view and different amount of background included, but the actual depth of field will not change. In other words, changing the focal length won't buy any less of the subject being in the range of acceptable focus. But the things that are out of focus may look softer and more pleasing.

Click to expand...

Since it is "maximised blur" in the out of focus zones that people mean when they talk about getting "shallowest Depth of Field"... it is not wrong to advise their
separating the background with distance and using a long lens from further away.

Doing so magnifies the background relative to a constant subject size, and the apparent blur of any zones outside DoF gets magnified along with it.

In other words, it DOES work to the best degree possible, even if nothing works to blur backgrounds very much when small sensor cameras are in use.
--

Click to expand...

Well, I'm not sure that "maximized blur" should be equated with "shallowest depth of field" whether people think that's what they want or not!

Click to expand...

But then, it IS equated in the minds of people who ask that particular question, is it not?

dsjtecserv said:

dsjtecserv said:

I agree that there is value in using that factor in order to optimize the aesthetic quality of that which is truly out of focus, so it was certainly worth pointing out. I would just suggest putting it in the context of the apparent quality of the background blur, rather than actually accomplishing reduced DoF.

Click to expand...

Even where beginners don't fully understand the question they are asking -- or its implications -- I think its a good idea to not perpetuate a misapprehension, even for the sake of simplicity. You and the other regulars in this forum do a great job of explaining things, but this one stood out as one that could possibly be misleading without further explanation.

Click to expand...

Well, we can't really blame the people who don't know the answers, for asking the "wrong" questions, can we?

In any case, between us we have answered the question from both directions....

• In "front-to-back-distance" terms, DoF is actually (as you state) the same with different focal lengths used at the same f/number and distance.

• But the appearance of the BLUR in those parts of the image NOT within DoF, is magnified more with longer lenses.

There we are. The statements above are equivocal, with no "rights" or "wrongs" in them.

Peace.
--
Regards,
Baz

Well, I'll see your Cher, and your Streisand... and I'll raise you an Alice Babs!

I have yet to see simulated shallow DoF that looks anything like what is produced by appropriate optics. I am sure someone with enough skill could pull it off with enough patience, but mostly this sort of fakery just looks amateurish and gimmicky. The problem is that the degree of blur is dependant on the distance from the plane of focus and no piece of software is capable of making that judgement.

Cedarhill said:

I have yet to see simulated shallow DoF that looks anything like what is produced by appropriate optics. I am sure someone with enough skill could pull it off with enough patience, but mostly this sort of fakery just looks amateurish and gimmicky. The problem is that the degree of blur is dependant on the distance from the plane of focus and no piece of software is capable of making that judgement.

Click to expand...

Yes, that's "sorts" true...I think it's possible for software to examine OOF areas in images and make a judgment about where it was...not very accurately as it won't have details about the objects. But, this might be OK for some users?

But in the camera, it's a different issue. Modern dSLRs have multiple phase sensors that can locate the position of background (and foreground) objects with some precision. I'm not sure if the camera my cuz Barrie was talking about is doing this:

" ...presumably it does this by identifying which zones are sharp and which are not. "

Regardless of the current state of the art, this IS a possible future development.

--
Charlie Davis
Nikon 5700, Sony R1, Nikon D50, Nikon D300
HomePage: http://www.1derful.info

"If ever a time should come, when vain and aspiring men shall possess the highest seats in Government, our country will stand in need of its experienced patriots to prevent its ruin."
-Samuel Adams, 1776

chuxter said:

But in the camera, it's a different issue. Modern dSLRs have multiple phase sensors that can locate the position of background (and foreground) objects with some precision. I'm not sure if the camera my cuz Barrie was talking about is doing this:

" ...presumably it does this by identifying which zones are sharp and which are not. "

Regardless of the current state of the art, this IS a possible future development.

Click to expand...

I did some search and have found the elusive cameras mentioned above. These are two new Sony compacts, Sony WX5 and TX9 that have a Background Defocus mode which mimics shallow DOF in-camera.
http://www.dpreview.com/news/1007/10070801sonywx5tx9t99.asp

"..As an extra refinement, new Background Defocus (mode) makes it easy to achieve pro-style images, similar to the results achieved with a DSLR camera and wide-aperture lens. Cyber-shot™ rapidly shoots two frames when the shutter button is pressed. Position information between the foreground and background is analysed automatically, creating a single picture with your subject in sharp focus and a beautifully blurred background."

I posted a question in the Sony forum regarding any experience with the Background Defocus mode, but unfortunately those cameras not yet available.

So the future is (a sort of..) already here!..

Cedarhill said:

I have yet to see simulated shallow DoF that looks anything like what is produced by appropriate optics. I am sure someone with enough skill could pull it off with enough patience, but mostly this sort of fakery just looks amateurish and gimmicky. The problem is that the degree of blur is dependant on the distance from the plane of focus and no piece of software is capable of making that judgement.

Click to expand...

But, strictly speaking, ALL out-of-focus backgrounds are fake , aren't they? I mean REAL optical out-of-focus-ness is restricted to the art of photography,† because it certainly doesn't happen in human vision.... even if people think it does.

To see what I'm getting at, put a single finger up in front of your face about 8" from your nose and focus and converge your eyes on it. Now, WITHOUT re-focusing or de-converging, "mentally" observe the features in the room beyond your focused finger. Is it out of focus back there? Like Hell, it is! It's very much double-imaged , yeah.. but BOTH those background images are hardly fuzzy at all, and this stays true all the way out to the horizon.

... Healthy human vision has deep DoF much like compact cameras, in fact!

And as a further fact, when Petzval came up with his super wide-aperture portrait lens (1840) the blurred backgrounds it generated were considered an uncomfortable and ugly FAULT by the first users, and something they wished to overcome if they could.

Just thought I'd mention it while people were talking about what's "fake" and what's not... ;-)

† well, de-focused image blur is also seen in other focusing optical systems like binoculars, telescopes microscopes, etc.
--
Regards,
Baz

Well, I'll see your Cher, and your Streisand... and I'll raise you an Alice Babs!

Barrie Davis said:

dsjtecserv said:

Barrie Davis said:

dsjtecserv said:

No, that one won't work, or rather is only indirectly beneficial. For any given subject framing and f number, depth of field is the same regardless of focal length. The out-of-focus objects will look different because of the different angular view and different amount of background included, but the actual depth of field will not change. In other words, changing the focal length won't buy any less of the subject being in the range of acceptable focus. But the things that are out of focus may look softer and more pleasing.

Click to expand...

Since it is "maximised blur" in the out of focus zones that people mean when they talk about getting "shallowest Depth of Field"... it is not wrong to advise their
separating the background with distance and using a long lens from further away.

Doing so magnifies the background relative to a constant subject size, and the apparent blur of any zones outside DoF gets magnified along with it.

In other words, it DOES work to the best degree possible, even if nothing works to blur backgrounds very much when small sensor cameras are in use.
--

Click to expand...

Well, I'm not sure that "maximized blur" should be equated with "shallowest depth of field" whether people think that's what they want or not!

Click to expand...

But then, it IS equated in the minds of people who ask that particular question, is it not?

Click to expand...

Indeed. But as a mentor you have the opportunity to gently move their understanding forward to ward something that will be more accurate -- and useful -- for them in the future.

Barrie Davis said:

Barrie Davis said:

Barrie Davis said:

I agree that there is value in using that factor in order to optimize the aesthetic quality of that which is truly out of focus, so it was certainly worth pointing out. I would just suggest putting it in the context of the apparent quality of the background blur, rather than actually accomplishing reduced DoF.

Click to expand...

Even where beginners don't fully understand the question they are asking -- or its implications -- I think its a good idea to not perpetuate a misapprehension, even for the sake of simplicity. You and the other regulars in this forum do a great job of explaining things, but this one stood out as one that could possibly be misleading without further explanation.

Click to expand...

Well, we can't really blame the people who don't know the answers, for asking the "wrong" questions, can we?

Click to expand...

No, we can't blame them, and I wouldn't suggest we do so. But we can help clarify things in a way that will help them understand and apply thing as they move forward. In this case, understanding the distinction between actual depth of field and degree of blur is important. Many (probably most) beginners accept the oversimplified "rule" that focal length affects depth of field, and assume that it applies regardless of subject distance or framing. I know I labored under that assumption for years and was flummoxed when it didn't seem to work that way. Now I know better, and understand how I can still use focal length to achieve pleasing results, but I wish I had had that understanding previously. My suggestion here is to just make sure we plant the seed of understanding early in the conversation, even if the questioner didn't know to ask that question.

Barrie Davis said:

In any case, between us we have answered the question from both directions....

• In "front-to-back-distance" terms, DoF is actually (as you state) the same with different focal lengths used at the same f/number and distance.

• But the appearance of the BLUR in those parts of the image NOT within DoF, is magnified more with longer lenses.

There we are. The statements above are equivocal, with no "rights" or "wrongs" in them.

Click to expand...

Well stated.

Barrie Davis said:

Peace.

Click to expand...

Absolutely.

Barrie Davis said:

--
Regards,
Baz

Well, I'll see your Cher, and your Streisand... and I'll raise you an Alice Babs!

Click to expand...

--
http://www.pbase.com/dsjtecserv

I would really like to see how these cameras simulate out-of-focus backgrounds. If they really do take distance from the focal plane into account, then it might be possible to simulate shallow DoF convincingly.

Barrie Davis said:

And as a further fact, when Petzval came up with his super wide-aperture portrait lens (1840) the blurred backgrounds it generated were considered an uncomfortable and ugly FAULT by the first users, and something they wished to overcome if they could.

Click to expand...

Yes, that's an important point...let me extend it a bit...

Things like blurry vs sharp backgrounds, color vs B&W, square vs rectangular vs round vs oval, vignetting, borders, using mats, glossy vs matte vs textured, accurate vs vivid colors, proper ways to hold cameras, etc. are just cultural trends. They change all the time, like fashions.

There is no right with fashion trends... :-0
--
Charlie Davis
Nikon 5700, Sony R1, Nikon D50, Nikon D300
HomePage: http://www.1derful.info

"If ever a time should come, when vain and aspiring men shall possess the highest seats in Government, our country will stand in need of its experienced patriots to prevent its ruin."
-Samuel Adams, 1776

There may be no "right" with fashion trends, but there are certainly wrong ones.

Cedarhill said:

There may be no "right" with fashion trends, but there are certainly wrong ones.

Click to expand...

That statement describes the observer, not the styling trend. What is wrong today according to current standards (by some group of people) will be judged differently in another time frame.

Remember that in his day, Frank Lloyd Wright was thought to produce wrong architectural designs by his peers. Today he is considered a master.

--
Charlie Davis
Nikon 5700, Sony R1, Nikon D50, Nikon D300
HomePage: http://www.1derful.info

"If ever a time should come, when vain and aspiring men shall possess the highest seats in Government, our country will stand in need of its experienced patriots to prevent its ruin."
-Samuel Adams, 1776

Lighten up Charlie. You are taking this way too seriously. The only opinion I can offer is my own and I am the word's greatest expert on it. My opinion, based on the images I have seen blurred by DoF post processing is that they look fake and would have looked better in their original state. Several posters in this and a couple of other DPR forums have tried to deceive readers with these efforts and it never works. These folks need to be informed that there is more to it than they think.

--

The greatest of mankind's criminals are those who delude themselves into thinking they have done 'the right thing.'

• Rayna Butler

What is with dpreview. It changed open in the previous heading to jibberish.

--

The greatest of mankind's criminals are those who delude themselves into thinking they have done 'the right thing.'

• Rayna Butler

Bokeh 2, another software that creats amazing virtual DoF results (and thanks to DigiDan for the URL )
http://www.alienskin.com/bokeh/examples.aspx

Let me add a note to the general discussion here. By now it is clear to me that only large sensor and bright lens will do authentic DoF. But we the compact users have already decided to compromise whatever it takes. So we will be happy to get even a "faked" DoF either by using a software or by some direct in-camera manipulation. c'est la vie..

Some of the samples in your link look pretty good. Based on my experiments with other software, I would guess you would still have to be very skilled to get a result that looks as though it were made with a camera lens. There is also the idea you mentioned that Charlie also implied. The image doesn't have to look like authentic lens blur to be appealing. In this instance, you are not talking about recreating camera lens bokeh but rather creating an interesting image by using selective blur. I still have to say that my opinion of most efforts in this regard is that they don't look as good as the original photo would have. That is both a matter of skill (of the creator) and taste (of the viewer).

Thanks for the link. Some of the OOF examples looked rather good. But at $200, it's a rather big "accessory" to add to your P&S!
--
Charlie Davis
Nikon 5700, Sony R1, Nikon D50, Nikon D300
HomePage: http://www.1derful.info

"If ever a time should come, when vain and aspiring men shall possess the highest seats in Government, our country will stand in need of its experienced patriots to prevent its ruin."
-Samuel Adams, 1776

I had a Fujifilm S200EXR with a Pro Focus mode but after reading some discussion here in the Fujifilm forum, I never tried it. The comments were that it looked smeary.

The S200 is a big bridge camera with a big (for a non dSLR) sensor.

The F300EXR seems to do it differently, with focus bracketing and face detection. It will take another pic of the background deliberately out of focus and combine the two images. I've not seen it in samples other than Fujifilms website.

http://www.fujifilm.com/products/digital_cameras/f/finepix_f300exr/features/page_03.html

With that 2.0 you can get some good DOF and bokeh
--
http://www.spots2share.com

Cedarhill said:

The image doesn't have to look like authentic lens blur to be appealing. In this instance, you are not talking about recreating camera lens bokeh but rather creating an interesting image by using selective blur.

Click to expand...

I fully agree. The terminology must be kept strictly. "Shallow depth of field" should be saved only for authentic optical lens output. The rest - software and in-camera filter output might be reffered as "background defocus", "selective blur", etc.