Rehash: DOF on DSLR vs DOF on P&S Digital

Thanks, Ulysses. Did you ever end up with an 828?
Heheh... that's a much more involved question than your question about DOF.

Suffice it to say that I occasionally do shoot with one when I get the chance.
I sold mine
after 6 weeks simply because I missed the buttery-smooth images
from a DSLR (no, I never got rid of my D60, but I did get a DR for
snapshots).
Cool. :-)
Canon still makes the MP-E...if they discontinued it, I'd buy
several "just in case". They are about $800 new.
Wow! That's much lower than I would have expected for this specialty lens. I really liked the results you demonstrated with it.
half the story. As you read in my review, you "need" (although
it's not a requirement) the MT-24EX or MR-14EX macro flash to get
full benefit from it. The MT-24 is about $600 nowadays.
It sounds like it's really about buying a system, rather than just the lens. You'd really have to have a dedicated need for this type of setup. And right now, I don't know that I'd have the subject matter to warrant it at this time.

But man!! I sure do love specialty gear! It's really nice sometimes to find a niche and to excel at it!

--

Ulysses
http://www.ulyssesphotography.com
 
Frank,

Firstly, my dSLR is P&S!

Lets talk about the influence of sensor size on depth of field.

Its easy to do with by looking at the equations and talking about teh adoption of an appropriate allowable diamater of the circle of confusuion, but I think you are not looking for that.

There have been develooed a large number of "intuitive" explanations of why it works that way, but most of those don't work for me.

Here's another appraoch.

"All other things bering equal", the depth of field perforamnce depends on the aperture - not as an f-number, but in terms of an actual diamater of the aperture (in mm, for example). The larger the aperture, the smaller the depth of field (assuming that we use a maximum allowable diameter of the circle of confusion that is consistent as a fraction of the sensor size).

One of the "other thngs" that we reasonably would require to be equal in comparing the depth of field performance of two cameras is the field of view. To get the same field of view on a smaller sensor camera requires a shorter focal length than on a larger-sensor camera.

Then, on the smaller camera, for a given f-number, the actual aperture diamater is smaller (it is the focal length divded by the f-number.) The smaller actual aperture diameter leads to a greater depth of field.

Best regards,

Doug
 
This explanation is not the usual way DOF is explained, but it's very intuitive if you understand parallax (the difference in what each of your two eyes see is parallax).

A lens produces an image by focusing light that strikes its entire surface onto a point on the sensor. That is, if you're taking a picture of a pinhead in the distance, the light from the pinhead that strikes every portion of the lens is combined to produce the image of the pinhead on the sensor.

Due to parallax, the view of the pinhead the left-most edge of the lens sees is going to be different from the view the right-most edge of the lens sees (think of these points as your eyes, with your face as the lens). The bigger the lens' diameter, the greater the parallax and more different these views. The more different these views, the more blurred the pinhead will be in the final picture (assuming it's not in the plane of focus).

So the smaller the diameter of the lens the more DOF you get. Technically, it's the entrance pupil not the diameter (pretty equivalent for telephotos, but very different for wide angles). Entrance pupil = focal length / f-ratio.

Shorter focal length lenses have a smaller entrance pupil, so more DOF.

P&S digicams use shorter focal lengths to achieve the same FOV as 35mm. Shorter focal length = smaller entrance pupil, so they have more DOF.

Higher f-ratio means smaller entrance pupil, so more DOF.
 
Thanks again, Doug, John, and everyone else...I think I'll be able to explain it pretty simply know...and I think I understand it better myself.

Frank
 
Due to parallax, the view of the pinhead the left-most edge of the
lens sees is going to be different from the view the right-most
edge of the lens sees (think of these points as your eyes, with
your face as the lens). The bigger the lens' diameter, the greater
the parallax and more different these views. The more different
these views, the more blurred the pinhead will be in the final
picture (assuming it's not in the plane of focus).

So the smaller the diameter of the lens the more DOF you get.
Technically, it's the entrance pupil not the diameter (pretty
equivalent for telephotos, but very different for wide angles).
Entrance pupil = focal length / f-ratio.

Shorter focal length lenses have a smaller entrance pupil, so more
DOF.
Yes, short focal length lenses have a small entrance pupil. Short focal length lenses (always retrofocus on SLRs) have large exit pupils too. This small entrace pupil, and specifically the ratio of the pupils causes the DOF on short focal length lenses to be LESS than on longer lenses in the macro area where parallax effects things most.

Jason
 
Ulysses,
"All other things bering equal", the depth of field perforamnce
depends on the aperture - not as an f-number, but in terms of an
actual diamater of the aperture (in mm, for example).
But there's the rub. All other things are not equal either, which
is what makes the "layman's" explanation a bit problematic. :-/
Indeed!

Even though in general I really like intuitive explanations, in this matter nothing does it for me but looking at the equations, and then adopting some set of conditions for "other things being equal" that make sense to me in light of the specific comparsion I am interested in at the time.

There's actually a rather elegant "intuitive" rationale based on information theory energy considerations and the fact that (assuming a comparable ISO sensitivity) a smaller total amount of photometric energy is required to capture an image on a sensor of smaller area than one of larger area. It involves the concept that lesser depth of field is a price we pay for capturing a greater amount of luminous flux from any given scene (that is, a greater amount of photometric energy in any given exposure time).

But I don't float that outlook very often!

Best regards,

Doug
 
achieve the nice shallow DOF and nice bokeh effect that you could
get from a 50mm lens at f/1.8?
The smaller the sensor the smaller the subject you can get a nice bokeh effect with. So, a G5 can only do it say with a person's head, while a DSLR can do it with a person's body. A large format camera might be able to do it with a group of people. All of this ignores the perspective based effect of background blur you get with telephoto lenses of course, which can help the situation if your goal is a blurry background. Don't get this blurry background confused with DOF, because they are different.
Based on the comments in this thread a G5s 7.2mm-28.8mm lens has
the same properties (DOF) as a 7.2mm-28.8mm lens (if it existed!)
for a 35mm EF body.
This is incorrect. A 7.2-28.8mm lens on a DSLR would have FAR greater DOF. You're forgetting to calculate the linear term of enlargement of the sensor. Specifically, the DOF of the lens on a G5 would be like a 35-140mm lens with an aperture of f/10 at the wide end and f/14 at the telephoto end. The equivalent lens on a 1.6x DSLR would be a 22-88mm with f/6 on the wide end and f/9 on the telephoto end.
This should mean that a shallow DOF is nearly impossible to obtain
on most P&S cameras.
It is for a large subject. Small subjects it is still possible for. Basically the subject has to be about 3x smaller than an equivalent subject you can shoot with a DSLR.

Jason
 
Even though in general I really like intuitive explanations, in
this matter nothing does it for me but looking at the equations,
and then adopting some set of conditions for "other things being
equal" that make sense to me in light of the specific comparsion I
am interested in at the time.
I did fairly well in school, but I sure wish I were more of a mathematician in the real-world application, Doug. It's a problem I lament nearly every day. :-[

Heheh... can you be a theoretician without having the math under your feet? :-)
There's actually a rather elegant "intuitive" rationale based on
information theory energy considerations and the fact that
(assuming a comparable ISO sensitivity) a smaller total amount of
photometric energy is required to capture an image on a sensor of
smaller area than one of larger area. It involves the concept that
lesser depth of field is a price we pay for capturing a greater
amount of luminous flux from any given scene (that is, a greater
amount of photometric energy in any given exposure time).

But I don't float that outlook very often!
Hmm... I've never heard it put that way before. Interesting! And no numbers for me to chew on. I like it! :-))

--

Ulysses
http://www.ulyssesphotography.com
 
This is incorrect. A 7.2-28.8mm lens on a DSLR would have FAR
greater DOF. You're forgetting to calculate the linear term of
enlargement of the sensor. Specifically, the DOF of the lens on a
G5 would be like a 35-140mm lens with an aperture of f/10 at the
wide end and f/14 at the telephoto end. The equivalent lens on a
1.6x DSLR would be a 22-88mm with f/6 on the wide end and f/9 on
the telephoto end.
Killer explanation. I'm dead.
 
Thanks for responding. After reading so many technical
explanations below, I think that "MrFoopy" hit it well: "Yet
another way to put this is that all lenses on digicams are by
nature wide angle lenses, and they have the correspondingly deeper
depth of field of wide angle lenses."
I disagree, but it works. I don't believe wide lenses have inherently wider DOF ( I've tried my 50 from 5 feet against my 100 at 10, for example ), but this is a good "laymen's terms" explanation, and helps people understand what's going on.
The reason I asked this question, and to answer your question
below, is because I'm currently working on my book before
submitting it publishers. I still can't decide if I want to do an
"artistic" book or a "how to" book, so I'm going right up the
middle with an "artistic how to" book. That being the case, I want
Are you sure you want to do this? Could work out better if you take the world by storm with your art work ... then publish a follow-up how-to book once you've got us hooked? ( "Us" meaning those of us who aren't already fans. )
to be able to explain a lot of the technical mumbo-jumbo in terms
that are so simple that even a guy like me can understand it.
A guy like you can make photos the rest of us have never dreamed before, and have written reviews on the most advanced macro lens ever created, then had these reviews published / carried on the major photography web sites.

I know you're being modest, but c'mon! If you're a simpleton, what does that make the rest of us???
I have enjoyed and valued all of your input over the years,
Forrest...thanks for helping me out.
Likewise!! I hope you realize this goes both ways, and that your photos are an inspiration, and raise the bar.

How is the book coming, though? Have you found a publisher?
 
...but a friend of mine is Melissa Springer, who did "Warrior Women" a photo piece about breast cancer survivors (my wife's sister is in that one), and the photography in the book "Salvation on Sand Mountain" about the infamous snake handlers who worship by handling snakes, plus another delightful book called "Important Things".

Melissa is going to put me in touch with some of her contacts in NY and hopefully that will lead somewhere.
Thanks for your encouragement...
How is the book coming, though? Have you found a publisher?
--
My Extreme Macro Bug Gallery: http://www.beautifulbugs.com

Ever wondered about the MP-E lens? Read my review here:
http://www.vividlight.com/articles/2914.htm

 
So people can idly speculate about 5mm lenses, 2mm lenses, 12-nanometer sensors, etc.....

Tsk. With so many (mildly erroneous, "hunch-based") explanations, why not just DO THE MATH? DoF functions are 8th-grade stuff, and can be reduced to a couple of simple diagrams for even small children to understand.

--
http://www.photorant.com/
 
Hi Kevin,

I hear you. I think diffraction is ignored because most photographers have never seen a side-by-side comparison of two images where one was taken at an f-stop that provided just enough depth of field to make a three-dimensional subject space appear uniformly sharp, with the other one taken using an aperture small enough to induce visible diffraction. At a given enlargement factor and viewing distance, it's very easy to detect a defocused Near or Far point in the subject space (insufficient depth of field) because subjects that lie closer to the plane of sharp focus are right there, in that same print, for comparison.

Everyone has seen prints with defocused elements surrounded by subject matter that is sufficiently focused. I submit that everyone has also seen prints that were degraded by diffraction - they just don't know it when they see it, because the effect softens the entire image more or less uniformly (disregarding for the purpose of this discussion that Airy disk diameters vary with the color of the light) -and- they never have a diffraction-free version of that image lying alongside for comparison.

Stopping down in the quest for depth of field, most photographers have, on many occasions, ignorantly forced diffraction's Airy disks to diameters that actually exceed the diameter of the largest circles of confusion produced by defocus. It is pointless to shrink your circles of confusion by stopping down if doing so will make your Airy disks even larger.

I submit that most depth-of-field aware photographers don't really know what aperture is necessary to obtain only "just enough" depth of field at the time of exposure. It can't be done without anticipating both the enlargement factor and the viewing distance of the final print prior to making the exposure. (See my last post.) If you don't know in advance precisely how large a print you will be making nor how closely it may be scrutinized, then you had better shoot for the worst case - the largest print your camera's format can bear, for viewing at a distance of 10 inches.

Failing to pre-visualize your final product before exposure can still hurt you even if you shoot for the worst case (big print viewed closely). You'll find yourself always working with very little depth of field (to force small CoC's) and no freedom to use the smallest apertures available on your lenses (to force small Airy disks). If you just continue to shoot as you always have, with no concern for print size and viewing distance at the time of exposure, the majority of your exposures will fall into two categories: Insufficient DoF for the print you end up making and viewing -or- too much DoF in the final print.

What's wrong with too much DoF (assuming your intent was not to use selective focus as a compositional tool) ?

1) As already mentioned, you can induce a softening of the entire image, by stopping down further than necessary.

2) Stopping down further than necessary will also force you to use a shutter speed that's slower than you could have used - increasing your vulnerability to camera and subject motion.

I understand that working in the field with a DoF calculator (like the customizable DoFMaster disks) is just too cumbersome for some shooting styles or assignments. But I can testify it only takes about TEN SECONDS before each exposure to use a spinning disk calculator like the ones you can produce with Don Fleming's freeware:

http://dfleming.ameranet.com/

I shoot fairly static subjects, always working from a tripod, but it just does not take long to ask myself, "How large a print do I want to make for this subject? When I've answered the question, I know which calculator to reach for. Ten seconds later, I'm ready to make the exposure. I've got three calculators for each focal length I carry. (For three different print sizes.) With a permanent marker, I've also written the diffraction-limiting f-stop on each calculator - the f-stop below which I should never stop down for that size print. (See my last post.)

If you can do a little "homework" up front, you'll be equipped to make exposures that produce really sharp prints.

Mike Davis
http://www.accessz.com
 

Keyboard shortcuts

Back
Top