No Depth of Field from -ANY- digital camera ( Updated )

[PlatinumWeaver]

Hi,

My sister took photography ( as part of Media I think ) at her school, so while I'm the one with the dSLR going out and taking shots, she has the "book smarts" so to speak. I was talking to her a few days ago about cameras and photos and this and that and she said that as yet there is no digital camera that has 'proper depth of field'.

Now from what I know of depth of field that isn't the case but it didn't sound like she was being confused about terminology or anything. Can anyone help me work out what exactly she was talking about?

Thanks all.

[MHD]

perhaps she was likening it to the eye's optical system which scans and constantly adjusts fl and once the brain does the interpolation it gives the illusion of infinate DOF.

Maybe she was trying to say that no camera system has full (or infinite) DOF (ie everything in focus)

[Matt. K]

All lens/camera combinations have "depth of field". So I don't know what she is talking about. Maybe you could ask her for more information?

[Greg B]

I dunno Platinum, kids these days

I think the tedious "Please explain" could reasonably be brought into play on this one.

Maybe she was thinking about something else and got a bit confused.

[PlatinumWeaver]

She was specifically saying that Digital cameras do not have this feature that film cameras do have.

I'll definately get more information from her later.

Ta

[gstark]

Dean,

Perhaps she's referring to the PHD* style cameras, which, due to their small sensor size, have a very restricted DoF.

But the D70 does provide effective DoF (and control of it too), and even moreso, if you look at the full frame digital cameras, which take exactly the same images as their 35mm counterparts.

Putting this another way, if the 35mm cameras have DoF, so too must any camera using the same lens systems as those 35mm cameras.



* PHD = Push Here Dummy

[birddog114]

PlatinumWeaver

Please check the following link

http://smad.jmu.edu/dof/index.html

Am very confused with what she said and

[Glen]

Tell her to look at the shots under portrait here by Onyx. The people are in focus the background isn't. What is that except dof?

She might be parrotting back at you something from her course a few years back which is not true now. Something from P&S days not DSLR.

A few weeks ago I was bought a book for holiday reading on photography. It was reprinted and ammended 2004, first run 2001. It said the same sort of things, written by an old photographer with no idea about digital. I want to write to the publisher and offer to rewrite 2 chapters!





Or the short answer is CRAP

[PlatinumWeaver]

PHD - Never heard them described as that before, always just Point-and-click or Point-and-shoot. I like it though!

I think that perhaps gstark and glen might have hit it, that when she was learning all this stuff the digital options were severely limited. I can certainly see that DOF is adjustable and managable with today's cameras.

I'll have a word with her, set things right...

[birddog114]

Yeah! I remembered that too, met an old bloke with his antique Konica in his hand and he said this and that then pointed to my D100 and said: " this bloody brick doesn't have DOF" :

The same thing my son at school (y12) he got photography subject and he asked me last year about the same funny question, and said it was from his teacher

Perhaps they're old Digi P&S but not with the modern DSLR

[Greg B]

Dean,

Perhaps she's referring to the PHD* style cameras, which, due to their small sensor size, have a very restricted DoF.

But the D70 does provide effective DoF (and control of it too), and even moreso, if you look at the full frame digital cameras, which take exactly the same images as their 35mm counterparts.

Putting this another way, if the 35mm cameras have DoF, so too must any camera using the same lens systems as those 35mm cameras.



* PHD = Push Here Dummy

Love "PHD" (in fact, I love all TLAs)

Gary, could same be said of, for instance, the old 110 format film (in relation to the smaller sensor issue)? And if the issue is that a small sensor or a small negative have restricted DoF, would that therefore mean that it was not a specifically digital issue even then?

I am trying to get my head around why a small sensor or film size would have DoF issues - I don't doubt that it is the case, just trying to make sense of it.

[MHD]

Its not the size of the sensor its the size of the optics... see what is important is the absolute size of the iris, which is usually stated as the fstop or as a fraction of fl (eg f/2.8, or f/1 would have an apeture equal to the fl).. For example a 50mm f/2.8 lens will have greater DOF that a 100mm f/2.8 lens as the iris will be bigger on the 100mm lens (0.050/2.8 metres as against 0.100/2.8 metres )

now in a compact, P&S or PHD cam, to give usefull equivenlent FL (to compensate for small sensor size) they use very short optics, eg 8-20mm lenses, meaning the actual size of the apeture is much smaller giving big DOF even when wide open..

Now, if you ripped out the supplied optics on say a CP5400 and stuck a 50mm/1.8 lens infront of it, besides getting a huge crop factor due to the small size of the sensor you would get EXACTLY the same DOF as on a DX sensor as on a full frame sensor as on a piece of film..

(my slightly more than 2c)

[Glen]

Greg, I think some of the P&S (both digital and film) have tiny apertures, tiny sensors and relatively wide lenses. Not too dissimilar to a 20mm lens on your camera stopped down to f20

[gstark]

Greg,

Love "PHD" (in fact, I love all TLAs)

There are some that are better than others.

Gary, could same be said of, for instance, the old 110 format film (in relation to the smaller sensor issue)? And if the issue is that a small sensor or a small negative have restricted DoF, would that therefore mean that it was not a specifically digital issue even then?

I would not argue with this concept.

Consider too the target audience (user) of th PHD camera - it has to be simple to use, and produce images within which the subjects should be recognisable. That means that they should appear to be in focus.

Not necessarily clear, crisp, razor sharp, etc.

Just clear enough to be recognisable.

Thus, with the probable exception of the Pentax SLR, focus free was pretty much the name of the game. Or at the best, zone focus - mountain, tree, two stick-people standing, head'n'chest.

Likewise with larger Instamatic (did it have a number? 125 springs to mind for some reason) which really has an image size closer to FF 35mm.

So with a smaller image/sensor, the goal of an in-focus image is easier for the camera's designer to acheive, and with an easy to get in-focus image (from the end-users' perspective) the camera must, by definition, be easier for them to use and take great looking photos with.

Then again, what comprises a "great looking photo" is always open to interpretation.

[jdear]

is she talking about the DOF preview button? on my old x700 is has a special dedicated DOF preview button which when pressed down, manually stops down the lens to see the current DOF.

maybe she is talking about the absence of this such physical button?

I know my A2 has a digital DOF which can be accessed by holding down a combination of buttons, of which of course i never use.

*PHD - i first heard of the term PHD by Ken Duncan on his "creative photography made simple" dvd. He went to a convention with all these other National Geographic photographers, and thought he was being professional having a fully manual camera (his linhof technorama 617s camera doesn have metering, uses a rangefinder, so he used to do all his metering via his old nikon F90x - mid-late 90's, i think he uses an light meter now)

they told him he needed to get a PHD camera, thats what all the rest of them were using.

I took some photos with my old manual FA, manual focus, etc, and it was a pain in the ass... not suitable for candids. How i wished I had a F6 or DSLR that day! hmmm PHD's... *drools sparatically*

[MHD]

so does d70... just under the lens

[gstark]

Its not the size of the sensor its the size of the optics...

Good point, although I still think Greg's question is right on the mark.

And it's still a convenient way to refer to the issue, especially when the optics are generally quoted as 35mm equivalancy factors, rather than in absolute (e.g. 35 - 105 vs 8mm - 24mm) terms.

The smaller image size virtually implies the shorter optics as being standard.

[gooseberry]

Its not the size of the sensor its the size of the optics...

No. To be all technical, Depth of Field is determined by the aperture, focal length and the circle of confusion (CoC). I read somewhere that the CoC as defined by Carl Zeiss is 1/1730 the diagonal measurement of the frame. So sensor size does affect the DOF.

Maybe Platinum's sister was referring to the fact that a small point and shoot digital camera with a very small sensor has a much larger DOF compared to a 35mm frame film camera for a given focal length/aperture setting.

[PlatinumWeaver]

It was really bugging me that I didn't know what she was talking about, and no-one here knew either, so I went in and annoyed her at work ( my work is slow at the moment, her's is always slow ).

From what she was saying it's less of a Depth of Field issue and more a Depth Perception problem.

Do digital cameras have more of an issue regarding layering in images? Say you have 3 objects at different distances all in focus, with a digital camera, do they look more like they are the same distance away than with a film camera?

I think this is what she is trying to say, she said she would show me articles and things tonight about the problem.

Does this ring a bell with anyone?

p.s. Sorry to hijack the thread away from purely DOF

[MHD]

sorry, I consider the iris (apeture) as being part of the optics... it is an optical element... (being pedantic I know)
quite often the maximum apeture is not defined by the iris but by the smallest peice of glass in the lens configuration.. so the iris you put into a lens is the same size as this smallest element (which acts as the spatial filter in the fourier domain but that is getting off track )

[darb]

say, what? the question is ambigious mate, ask her to clarify what she means? dSLR's have huge ability to control DOF just as film cameras. (differing degrees perhaps she means?)

[/list]

[MHD]

Given the same lens you will get the same results on film or CCD/CMOS/LBCAST

[gooseberry]

Given the same lens you will get the same results on film or CCD/CMOS/LBCAST

Not quite, if the sensor is not full frame the depth of field will be different. A 50mm lens set f/2.8 on a D70 will have a different DOF to when it is on a film SLR.

You can check here for more info

http://www.nikonians.org/html/resources ... ocal2.html

Different format sizes have different Circle of Confusion diameters which thus effect DOF.

[birddog114]

Not quite, if the sensor is not full frame the depth of field will be different. A 50mm lens set f/2.8 on a D70 will have a different DOF to when it is on a film SLR

It's exactly right! that why each lens and each cam has a difference DOF.

[gstark]

Its not the size of the sensor its the size of the optics...

No. To be all technical, Depth of Field is determined by the aperture, focal length and the circle of confusion (CoC). I read somewhere that the CoC as defined by Carl Zeiss is 1/1730 the diagonal measurement of the frame. So sensor size does affect the DOF.

I'm not so sure.

Is the frame that you're referring to here the frame of the image sensor, or the frame that would be provided as a result of the image circle provided by the lens?

My gut feel is that it's the latter, and I find it hard to accept that a physical constraint, unrelated to the manner in which the image is created (in the lens) has such an effect.

Let's put this another way - grab a simple magnifying glass, which of course is a lens. Make up three frames from paper or cardboard, and make one the same size as a FF 35 frame, the second the same size as a D70 sensor, and the third a similar size to, say, a 5700 sensor.

Place a bare bulb somewhere above these "sensors", and using your magnifying glass, locate it so that, for each sensor in turn, the projected image of the bulb is brought into focus.

Does that distance vary between "sensors" ?

What about when that projected image is out of focus? How do the CoC look? What are the distance relationships between the lens and the "sensors" ?

Now grab your 50mm 1.4 or whatever and repeat this series of tests.

It's my contention that the sensor's size - or more specifically, the sensor's frame size - would have no bearing on this.

[MHD]

Without reading the nikonians art, I agree with Gary... you can form an image on a screen or piece of paper... the image is formed by the optical system not system collecting the information..

The BIG difference different sensors make is signal to noise, resolution and colour response (both ballance and depth) only if the image is hopelessly undersampled would the sensor change that compasition of the image...

sorry if I am being pedantic about this! But I am a physicist after all and this is more than slightly my domian

Edited to change horrific grammatical mistakes.. why is when I am writing my thesis that I can not write anything else!!!!

[gstark]

Given the same lens you will get the same results on film or CCD/CMOS/LBCAST

Not quite, if the sensor is not full frame the depth of field will be different. A 50mm lens set f/2.8 on a D70 will have a different DOF to when it is on a film SLR.

The differences, when calculated, are little more than academic.

Lens choice - 50mm, F4, distance 4ft.

35mm Near DoF 3.82 ft, Far DoF, 4.2Ft
D70 - 3.86/4.16


To my mind, that's no difference, in practical terms.

HYperfocal distances ... that's a whole different ballpark, but that's not the current game, is it?

[gstark]

Without reading the nikonians art,

It's not an article; just a DoF calculator.

[gstark]

Platinum,

Please be sure to post those references and articles. This is certainly a fascinating topic.

[Matt. K]

Haven't seen it mentioned here, but depth of field is also strongly affected by camera to subject distance. The closer you get the less you got!

[PlatinumWeaver]

Definately will!

[gstark]

Haven't seen it mentioned here, but depth of field is also strongly affected by camera to subject distance. The closer you get the less you got!

Very true too.

[MHD]

See my post earlier... The most important thing is absolute size of the iris which, for a constant fstop is directly proportional to focal length..

However DOF caused by the angular spread of rays though the system.. The closer a subject is to the lens the greater the spread of angles the optics will collect from a given point.

[PlatinumWeaver]

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

This is what my sister was refering to, please debunk as nessesary.

[gooseberry]

Is the frame that you're referring to here the frame of the image sensor, or the frame that would be provided as a result of the image circle provided by the lens?

Hi gstark,

The frame I am referring to is the frame of the image sensor.

"the CoC using the "Zeiss formula" is d/1730, where d is the diagonal measure of the film, in millimeters."

[gooseberry]

Without reading the nikonians art, I agree with Gary... you can form an image on a screen or piece of paper... the image is formed by the optical system not system collecting the information..

...

sorry if I am being pedantic about this! But I am a physicist after all and this is more than slightly my domian

Hehe.. no worries about being pedantic - I get that way too sometimes. I agree that the image is formed by the optical system and not the system collecting it - but to quote someone else on CoC.

"The film size is important because you don't have to enlarge large negatives as much to get a particular sized print. So, a 6 × 4.5 cm frame, being roughly twice the size of a 35mm frame, will have a CoC that's roughly twice the size of that for a 35mm frame. In other words, if a fuzzy disc 0.025 mm wide looks like a point when printed from 35mm film, you can have an 0.043 mm disc on 6 × 4.5 cm film and still have the same apparent degree of sharpness if you enlarge it to the same size print as you made from the 35mm frame."

[gooseberry]

The differences, when calculated, are little more than academic.

That's 'cos the DX size format is not a lot smaller than a full 35mm frame. But it can vary quite a bit depending on focal length, aperture and subject distance.

eg.
50mm f/8 subject distance 5m

35mm DOF is 8.28m - 3.58m
D70 DOF is 7.32 - 3.8m

[PlatinumWeaver]

Any comments on the webpage I linked to? 90% of it is beyond me..

[gstark]

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

This is what my sister was refering to, please debunk as nessesary.

No need to debunk; it makes quite a bit of sense, especially once it's taken into context.

I'm going to pull just two phrases from it, that you need to refer back to your sister however ....

"Digital camera sensors (at least in the cameras below $1000 or so) are much smaller. For example, the active sensor area in the Olympus C-30x0Z series is about 5.27x7.03 mm."

He's qualifying what he's saying. Mostly, he's talking about Oly P&S cameras, and even links his article to a table of Oly P&Ss.

"For almost all digital cameras (except models based on modified SLR bodies by Canon, Nikon, Sigma, and Contax), N varies between 4 and 5.5, hence difference is quite dramatic."

In this section he's providing a formula for the relationships of DoF between different types of camera. He points out that he sourced a reference to this formula in a photographic manual from the 50s.

So much for the digital age.

Bottom line is that this is nothing new, but your sister probably hasn't quite fully understood the context of this.

Yes, there's very little control of DoF with digital cameras, and in many cases, we'll see (in enlargements) image degradation due to pixelation before CoCs become an issue.

But larger pixel counts partially address this, as do larger sensors.

And Gooseberry's drawing to our attention the scale of enlargement - and thus the sensor's frame size - is a factor too.

But I would hold that as being a secondary factor; the optics build the CoC first, and the degree of enlargement is a product of the final image that you're producing, subject to the constraints of the ultimate number of pixels available within which an enlargment can be made.

All clear now?

[Mj]

mmm... Gary beat me to it... so I'll just say that whilst DOF effects on a digital SLR may well present slightly differently to a film based 35mm SLR, DOF certainly is a creative component available to digital SLRs.

The article doesn't spell that out but prefers to focus on differences between film SLRs and little P&S digitals... and so is not incorrect... just selective in the information it presents. BTW... I have in fact seen examples of P&S cameras producing quite nice DOF effects via addon lenses, tubes etc.

Michael.

[Matt. K]

Circles of confusion? Now that's something I know about!

[Onyx]

I do believe the 'digital' was referring to PHD cams as most laypersons aren't aware of the catergory of DSLR cameras.

But PW - you gotta agree with her, as you'll know the female is always right.

[PlatinumWeaver]

Onyx: I agree, I think I might just drop the subject and hope it fades away. I'm glad I started such a flurry of conversation here though...

[Greg B]

Excellent thread, thanks for starting it PW, and thanks to the posters for all the interesting information.

And, ironically, thanks to your sister

[Onyx]

And, ironically, thanks to your sister

This could be taken quite out of context...