Digital Depth of Field

[Pete Johnson]

I've read the Depth of Field and sensor size articles and many of the posts. It seems clear that DoF is different for DSLR cameras than for 35mm film. I'm intrigued, when taking portraits in studio, I seem to get more DoF for my aperture setting than I expected. The original article could suggest that as I'm using a Nikon DSLR with a smaller than 35mm sensor, F14 acts like I would expect from F22... and I could easily imagine this translating down the scale with say, F2.8 giving me the DoF I would have expected from F4 or F5.6 on 35mm.
I'm interested in your thoughts and experiences with this - I would expect it to have an impact on lighting settings if I intend to shoot with both film and digital in the same session, as the DoF at 125th f5.6 would be different for the two cameras.

[dobson]

The reason you are getting less apparent depth of field is that you are using a shorter lens for the same field of view. I did the math on this a little while ago, and it turns out that while you can get close to the look of 35mm by simply opening the aperture you can't get it to look identical. There's a thread on here (sensor size and DOF in medium format, I think), that explains this concept in depth.

Phillip

[Pete Johnson]

Thanks Phillip

I'll look it up. Fortunately I don't need the digital and 35mm to be identical, but I was interested in the idea that the DoF will come out differently - in effect, to get the same DoF with the DSLR I would have to turn down the output of the studio flash so I can use a wider aperture.
There are of course other practical situations where the DoF will be different, for example in Landscape work where I may be able to use f16 (or f11?) to get a more useful shutter speed. If my understanding is right, then I'll need to learn to map the apertures to something larger (around 1.5 stops) to get the same DoF I expect to get from 35mm.

[feppe]

Thanks Phillip

I'll look it up. Fortunately I don't need the digital and 35mm to be identical, but I was interested in the idea that the DoF will come out differently - in effect, to get the same DoF with the DSLR I would have to turn down the output of the studio flash so I can use a wider aperture.
There are of course other practical situations where the DoF will be different, for example in Landscape work where I may be able to use f16 (or f11?) to get a more useful shutter speed. If my understanding is right, then I'll need to learn to map the apertures to something larger (around 1.5 stops) to get the same DoF I expect to get from 35mm.
[a href=\"index.php?act=findpost&pid=111723\"][{POST_SNAPBACK}][/a]

I was wondering the same as I moved from 35mm and 6x6 to DSLR. It gets quite complicated and math-intensive, but google for "depth of field", digital and "circle of confusion" and you should get more info than you need. Also, there are nice DOF calculators where you can enter your own custom circle of confusion based on your preferences.

[Monito]

"Digital" and 35mm film depth of field are the same if you use a full-frame digital camera such as the Canon 5D.  Not all digital is cropped and additionally there are "medium format" digital systems that are larger than the 35mm film frame.

It is best not to write as if "digital" is synonymous with the cropped 1.5 or 1.6 factor sensors.

[Pete Johnson]

"Digital" and 35mm film depth of field are the same if you use a full-frame digital camera such as the Canon 5D.  Not all digital is cropped and additionally there are "medium format" digital systems that are larger than the 35mm film frame.

It is best not to write as if "digital" is synonymous with the cropped 1.5 or 1.6 factor sensors.
[a href=\"index.php?act=findpost&pid=112269\"][{POST_SNAPBACK}][/a]

Thanks for the reminder!
I haven't used a Canon 5D - do you really get the same DoF as for 35mm film? Or do you just get the same focal length?

The articles on this website suggest that the DoF is different for different sensor sizes, and that the optimum sharpness occurs at different apertures (I know this isn't depth of field, but since we're talking about sensors, then maybe it needs including).

Please note, my original thread description said Practical Digital Depth of Field. As a photographer, although I use aperture settings to affect exposure, I set the aperture I use with the Depth of Field in mind (shallow for portraits, deep for landscapes...). It seems to me that DoF is different for a digital camera, and that means I need to learn to correct/compensate to produce the effects I want.

Personal experience shows me that my photos are softer at small apertures (probably due to diffraction) and therefore a smaller aperture isn't always best for getting everything looking sharp. I also find that I get a greater than expected depth of field (using Nikon D200) for a given aperture setting. In practical terms, if I can get the DoF for f4 at f2.8, I can shoot at a higher shutter speed, and don't need a VR lens. I already map the focal length in my head due to the 1.4 crop factor. How much do I need to map the aperture value by when looking at using DoF in a creative way?

What would really be useful here is practical experience and thoughts rather than theory and references to theory sites. I'm sure that my optimum sharpness comes at f8, with a DoF similar to what I expected from f11 or f16 when I used 35mm. Do others with a D200 agree? Do those with other cameras (any brand) find anything similar - or how different is it for you?

[howiesmith]

Thanks for the reminder!
I haven't used a Canon 5D - do you really get the same DoF as for 35mm film? Or do you just get the same focal length?

The articles on this website suggest that the DoF is different for different sensor sizes, and that the optimum sharpness occurs at different apertures (I know this isn't depth of field, but since we're talking about sensors, then maybe it needs including).
[a href=\"index.php?act=findpost&pid=112841\"][{POST_SNAPBACK}][/a]

From Understanding Depth of Field on this site:

"There was a query in October, 2001 on my Discussion Forum as to whether Depth of Field was calculated any differently for digital Vs. film. The answer is, no. There is no difference whosesoever. DOF doesn't care about the recording media type or size, though a lower COF is used for medium and large format, since the amount of magnification to make a decent sized print is much less than for 35mm."

[Pete Johnson]

From Understanding Depth of Field on this site:

"There was a query in October, 2001 on my Discussion Forum as to whether Depth of Field was calculated any differently for digital Vs. film. The answer is, no. There is no difference whosesoever. DOF doesn't care about the recording media type or size, though a lower COF is used for medium and large format, since the amount of magnification to make a decent sized print is much less than for 35mm."
[a href=\"index.php?act=findpost&pid=112846\"][{POST_SNAPBACK}][/a]

Thanks Howie - I did read this. However I'm referring to what I would call the 'practical depth of field' - 'depth' in my photos that seems to be sharp. I understand that this is goverened by all sorts of theory, but at the end of the day, I would say that most photographers use DoF to create areas of image that are sharp/not-sharp. Should I refer to this as Depth of Sharpness then?


Am I asking the wrong question?

Is this the wrong forum for my search for knowledge?

Are there any other photographers out there who are experiencing similar issues to me?

To be honest, I reckon I'd get a lot more interest if I just set up a post slating Nikon or Canon! Yet I feel that, in practical terms, Depth of Field/Sharpness (whatever) is a key part of usiing my camera to produce the photos I want, and the DoF (DoS?) IS affected by the camera (sensor size, pixel pitch etc) and the lens used.

Come on guys - either help or tell me where else I can gather the info I seek.

Cheers!

[Ray]

If you shoot the same scene with a 3mp D30 and a 10mp 400D, focussing on the same point, using the same lens at the same aperture, the 400D will likely exhibit a slightly shallower DoF than the D30, assuming the lens is good quality.

This is because the parts that are in focus with the 400D may be significantly sharper than the same parts in the D30 shot. But the parts that are not in focus will be similarly unsharp in both shots.

Further explanation: This difference in DoF, despite the sensors being the same sze, will be dependent upon print size. For postcard size prints you would probably see no difference in DoF. At A3+ size you probably would provided you were not using a severely diffraction limited aperture such as f22 or f32, in which case the 400D would be crippled redarding optimal sharpness.

[howiesmith]

The image of the circle of confusion at the limit of depth of field is fuzzy for two reasons.  The first and predominate effect (except for small circle of confusion) is the point of light that is out of focus and appears as a fuzzy disc simply because it is out of focus.  Second, the circle of confusion appears fuzzy because the lens cannot completely resolve the edges of the disc (edge resolution compared to circle of confucsion diameter).  This second effect is very small (second order) for a reasonable good lens (assumed) and is ignored by photographers when actually calculating depth of field.

[Ray]

The image of the circle of confusion at the limit of depth of field is fuzzy for two reasons.  The first and predominate effect (except for small circle of confusion) is the point of light that is out of focus and appears as a fuzzy disc simply because it is out of focus.  Second, the circle of confusion appears fuzzy because the lens cannot completely resolve the edges of the disc (edge resolution compared to circle of confucsion diameter).  This second effect is very small (second order) for a reasonable good lens (assumed) and is ignored by photographers when actually calculating depth of field.
[a href=\"index.php?act=findpost&pid=113146\"][{POST_SNAPBACK}][/a]

The essential point here is that out-of-focus parts of the images  are less dependent upon lens quality and sensor pixel density than the in-focus parts of the image.

We all know that lens quality and sensor resolution are significant factors which generate a massive amount of discussion on forums like this, but basic DoF calculators tend to ignore such issues. One would be forgiven, when looking at such DoF formulas, if one were to draw the conclusion that all lenses are equal and all formats of the same size have equal resolving capability.

[howiesmith]

I've read the Depth of Field and sensor size articles and many of the posts. It seems clear that DoF is different for DSLR cameras than for 35mm film. I'm intrigued, when taking portraits in studio, I seem to get more DoF for my aperture setting than I expected. The original article could suggest that as I'm using a Nikon DSLR with a smaller than 35mm sensor, F14 acts like I would expect from F22... and I could easily imagine this translating down the scale with say, F2.8 giving me the DoF I would have expected from F4 or F5.6 on 35mm.
I'm interested in your thoughts and experiences with this - I would expect it to have an impact on lighting settings if I intend to shoot with both film and digital in the same session, as the DoF at 125th f5.6 would be different for the two cameras.
[a href=\"index.php?act=findpost&pid=111711\"][{POST_SNAPBACK}][/a]

Rather than get into another extended discussion that generates a lot of heat but no fire, I would make the following suggestion to you, Pete:

You have changed "cameras" (35mm film to DSLR?) and the DoF changed.  Look very carefully at everything else you changed when you just changed cameras and then determine what changed factors actually affect the DoF you expected.

[howiesmith]

I've read the Depth of Field and sensor size articles and many of the posts. It seems clear that DoF is different for DSLR cameras than for 35mm film. I'm intrigued, when taking portraits in studio, I seem to get more DoF for my aperture setting than I expected. The original article could suggest that as I'm using a Nikon DSLR with a smaller than 35mm sensor, F14 acts like I would expect from F22... and I could easily imagine this translating down the scale with say, F2.8 giving me the DoF I would have expected from F4 or F5.6 on 35mm.
I'm interested in your thoughts and experiences with this - I would expect it to have an impact on lighting settings if I intend to shoot with both film and digital in the same session, as the DoF at 125th f5.6 would be different for the two cameras.
[a href=\"index.php?act=findpost&pid=111711\"][{POST_SNAPBACK}][/a]

You don't give any information about focus distance, focal length of lens, f/stop, CoC, size of print and cropping, etc., so it is very hard to help.

Could this be a focus problem?  If you are focusing on a face for a protrait, there may be little or esentially no "subject" in the foreground to judge the near limit of DoF.  If the camera is focused farther away than expected, the background may be getting more in focus, giving the impression of greater DoF.

[Pete Johnson]

Thanks for the support.

I use a Nikon F80 for film and a D200 for digital. I use Nikon D type lenses for both, or a Sigma f2.8 zoom. The lenses work happily on either camera body.

What I find is that the 'practical' depth of field I get with digital is greater than with film. This may well be a function of the different recording media - (smaller)sensor vs film.

This means that I seem to be able to open the lenses up an stop on digital and achieve the same region of sharpness that I get on film. This has two main effects in my photography - I get a 1 stop advantage in light (and hence can use a faster shutter speed - similar to using a VR lens). The second effect is that for the same intended studio image, I dial my studio lights down a stop in output because I'm using digital and an f-stop wider aperture. When I use film, I dial it back up again.

What interested me was this practical difference in my camera equipment. I've not seen any discussions about this, and wondered whether other photographers had noticed the same thing... you know sometimes you spot something, and ask yourself 'Is it just me?'...

I'm in the process of collecting a range of images from both cameras to compare and prove the effect.

In practice I have learned what works for my own equipment. Where things do get more interesting is when you try to help someone else, or when you compare pictures, since a variation in 'practical' depth of field between cameras would render aperture information less useful.

[howiesmith]

This means that I seem to be able to open the lenses up an stop on digital and achieve the same region of sharpness that I get on film. This has two main effects in my photography - I get a 1 stop advantage in light (and hence can use a faster shutter speed - similar to using a VR lens). The second effect is that for the same intended studio image, I dial my studio lights down a stop in output because I'm using digital and an f-stop wider aperture. When I use film, I dial it back up again.

[a href=\"index.php?act=findpost&pid=114028\"][{POST_SNAPBACK}][/a]

Something for nothing.  Science fails once again.

[AJSJones]

Two common issues that surface when DoF is raised are degree of enlargement (particularly of the circle of confusion diameter) and viewing distance of  the print (and therefore its resolution relative to the human eye's perception of details).  However, in the "digital age" other things come into play.

I often print at a size determined by the "information content" to use a term loosely, but I really mean original pixels; in the case of some images, upressing seems acceptable in the final print and in other cases it is readily apparent and therefore undesirable. In other words I print as large as the pixels allow.  Sometimes original pixels can be made to look very good at 240 per inch, while by 300 it is rare that they are unacceptable.  I was therefore curious to see what the geometric enlargement factors were, given the variable pixel density on the sensor and these two final ppi resolutions.  (To head of a potential wildfire, I should say I am not inferring that all pixels have the same quality)  The table below represents the degree of enlargement (linear not areal) that results from printing from the indicated cameras at 240 or 300 original ppi.  Of course, the actual print size will vary based on MP of the sensor.  

As a point of reference, for the original DoF scales on 35mm film camera lenses, the dimension of the 24mm original image became 8 inches in the "8x10" held at "standard" distance for viewing and assessment of DoF. This represents about an 8.5x enlargement.  The table just happens to be by ascending pixel density which yields ascending geometrical enlargement.

The reader is left to peruse and draw conclusions about the relevance of these numbers to their own workflow  

[Ray]

I use a Nikon F80 for film and a D200 for digital. I use Nikon D type lenses for both, or a Sigma f2.8 zoom. The lenses work happily on either camera body.

What I find is that the 'practical' depth of field I get with digital is greater than with film. This may well be a function of the different recording media - (smaller)sensor vs film.

What interested me was this practical difference in my camera equipment. I've not seen any discussions about this, and wondered whether other photographers had noticed the same thing... you know sometimes you spot something, and ask yourself 'Is it just me?'...
[a href=\"index.php?act=findpost&pid=114028\"][{POST_SNAPBACK}][/a]

Dear me! This matter has been discussed ad nauseum on this forum over the years   .

Basically, if your shooting position remains unchanged and the Field of View remains unchanged, then the smaller the format the greater the DoF at any chosen F stop. I think even Howard would agree with this   .

The D200 is a smaller format than 35mm. For the same FoV from the same shooting position,  in circumstances where you would use a 50mm lens with the D200, you would use a 75mm lens with 35mm film. In order to get approximately the same DoF with the 75mm lens on the 35mm camera, you would need to multiply whatever f stop you were using with the 50mm lens on the D200 by a factor of 1.5, ie. F2.8 with the 50mm lens becomes equivalent to f4 (approx.) with the 75mm lens, regarding DoF.

Similar differences exist between 35mm and MF.

[Pete Johnson]

Dear me! This matter has been discussed ad nauseum on this forum over the years   .

Basically, if your shooting position remains unchanged and the Field of View remains unchanged, then the smaller the format the greater the DoF at any chosen F stop. I think even Howard would agree with this   .

The D200 is a smaller format than 35mm. For the same FoV from the same shooting position,  in circumstances where you would use a 50mm lens with the D200, you would use a 75mm lens with 35mm film. In order to get approximately the same DoF with the 75mm lens on the 35mm camera, you would need to multiply whatever f stop you were using with the 50mm lens on the D200 by a factor of 1.5, ie. F2.8 with the 50mm lens becomes equivalent to f4 (approx.) with the 75mm lens, regarding DoF.

Similar differences exist between 35mm and MF.
[a href=\"index.php?act=findpost&pid=114315\"][{POST_SNAPBACK}][/a]

Thanks Ray - this is what I thought was happening.

I'm interested that this has been discussed 'ad nauseum over the years', yet I find few people mention the difference when assisting or teaching someone else. In a studio set up I can see that it isn't practical with a group to keep turning the lights up and down, so you might go for a compromise...
Then there's the view that, from a DoF point of view, an f2.8 lens on a smaller sensor DSLR gives a faster shutter speed, with a similar effect to a VR lens...
And finally there's the issue of small aperture diffraction, which ought to be easier to deal with in a smaller sensor DSLR (because the practical DoF is greater for a given aperture, so you won't need to use f22...) although I'm not convinced, as a smaller sensor seems more prone to the diffraction softening effect, and film probably has a finer grain.

Interesting thoughts though, and I'd suggest the scaling effect due to a smaller than 35mm sensor has more impact than most people would think.

And although it looks like I gain something for nothing here, I should add that I don't use wide angle much... and anyone with a full frame camera will be happy to point out that I can't achieve the same field of view as they can!

[howiesmith]

Change from film to digital only, no change in DoF.

Change enlargement factor and focal length, likely change DoF.

Change f/stop to compensate for the change in focal length and enlargement effects on DoF, change the flash exposure.

If you understand DoF and know what (and how much) you all are changing when you change cameras, it all works out fine.

[Jonathan Wienke]

For clarity's sake, I define "in focus" to the part of the image that has pixel-sharp detail. This equates to a CoC equal to the pixel pitch of the sensor. Using this DOF calculator will accurately predict the distance range in which focus will be pixel-sharp for any digital camera.