Focal length and depth-of-field

As may become obvious, I'm fairly new to photography, currently
using a Canon A590 P&S, and am looking at getting a DSLR. But I
need help understanding something, because it could affect what I
buy.

It seems to me that the one thing I can't get with my A590 that I
could get with a larger-format camera is shallow depth of field -
those wonderful blurry backgrounds. Even when I open up the A590
all the way, everything is still pretty much in focus. Well, not
exactly in focus, but not exactly blurry either. And my
understanding of this has always been that this was because the lens
used on my camera, being appropriate for a smaller camera and a very
small sensor, had a much shorter focal length, and that, all else
being equal, the shorter the focal length of a lens, the deeper the
depth of field of the image.

But I see discussions of this issue which say it's not true - that
crop sensor cameras have the same depth of field as full frame.
Then it turns out that what they really means is they have the same
depth of field if they're using the SAME lens. Well duh.

And then I see something like this from Scott Bourne's Photofocus
site that makes it look as though the sensor size determines depth
of field:

"The larger the sensor, the shallower the depth of field for a given
subject size. Small P&S cameras have large Depth of Field (DOF.)
With a crop sensor, you have larger DOF. This means it’s harder to
get a good bokeh and harder to control the backgrounds."

Well, having thought about all of this, I've just about concluded I
was right in the first place. The sensor size per se doesn't
determine the depth of field, but the focal length of the lens does
(all else like aperture and distance being held constant).

But I want to do a thought experiment, and get the experts here to
tell me if I'm right.

There are two cameras. The first is a full-frame DSLR with a fixed
50 mm lens. It's focused on a "test pattern" subject 10 feet away.

The second camera is a 1.6 crop sensor camera, focused on the same
subject at the same distance. The field of VIEW of photos from the
two cameras is the same. I'm not sure of the math, but I think this
means the actual focal length of the lens on the second camera is
31.25mm.

If both lenses are opened up to the same maximum aperture, I believe
the depth of field from the full-frame camera will be significantly
shallower than from the crop-sensor camera. Is that right?

I understand that a number of crop sensor cameras use standard
full-frame lenses, but not all of them do, and I just want to be
sure I understand what I'm giving up in cases where both the body
and the lens are reduced in size proportionately.

And if there's a link to something that explains this in detail,
including the optics involved, that would be great.

Thanks

 

I'm no expert, but I can look at wikipedia...
http://en.wikipedia.org/wiki/Depth_of_field

So take the below with a grain of salt, but my impression is that:

1) The DOF varies linearly with the sensor size (keeping the lens
focal length and aperture constant), simply because a larger sensor
is magnified less to create the final image, meaning that a "blur
spot" can be proportionally larger before being considered out of
focus.

2) It also seems that DOF varies roughly with the inverse-_square_ of
the lens focal length (keeping the aperture constant).

To keep the angle-of-view constant, when changing the sensor size, you
linearly vary the lens focal length according to the change in sensor
size. When doing this, the two above factors will oppose each other,
but the second one will "win", since it's a square factor, whereas the
first is linear.

Thus, keeping the angle-of-view constant, while changing the sensor
size, the DOF will vary inverse-linearly with the sensor size.

Yes; using the above rule-of-thumb, the DOF in the 1.6-crop image will
be 1.6 times the DOF in the full-frame image. To achieve the same DOF,
you'd need to use an f-number 1.6 as big -- e.g., instead of f2.8, you'd
need to use f1.7.

I don't think it really matters; the only real difference is that
"non-full-frame capable" lenses don't have as large a usable image
field. The behavior in the usable portion should be the same as a
full-frame lens.

-Miles

 

http://www.dofmaster.com/dofjs.html

 

Increasing focal length may decrease DOF, but if you move back and
reframe in exactly the same manner, then moving back increased your
DOF and in the end you'll get the same DOF for same given aperature.

This is something that DSLR

 

For similar fields of view, indeed focal length scales with format size.
So smaller cameras do have a hyperfocal distance that kind of scales
with format size. The result is that since the photographer doesn't
scale down, he sees a nearer hyperfocal distance and finds it harder to
create a thinner area of good focus.

 

What planet in what universe with new laws of physics are you on,
pretend-photographer troll?

 

Say you frame a person in your viewfinder. You are at 50mm f/4. You
have a given DOF. If you zoom to 100mm, keep f/4, and move back to
get the same framing; then you'll have the same DOF. Try it sometime.

That is why it seems like big lenses have small DOFs compared to
smaller lenses. If you are focusing on something from 5' away, DOF of
1' seems pretty good. But if you have a 300mm lens and are using the
same framing, that DOF of 1' seems pretty darned small.

Meanwhile, this is the last post I will make on the subject because I
do not intend to get into a flaming war with you, which I see from the
tone of your post is what you are trying to start. Sorry, but I'm not
taking your bait. Rather, I'll go grab my camera and take pictures.
You should try it sometime. Taking pictures is much more fun than
flaming.

 

While I tend to agree with a lot of what you postulate, your statement that
you can change DOF by changing focal length without losing aperture just
doesn't jive that well with reality.

The very fact that you've changed focal length entails changing the
aperture -- especially with smaller sensored cams using variable aperture
tele lenses.

If you're small sensor has a lens rated at, let's say, f/2.8 - 5.6, with a
(35mm equiv) focal length of 28mm to 130mm, then simply zooming from 28 to
max lenth automatically drops your aperture size down to f/5.6 so you're
going to have to compensate with a change in ISO or shutter speed.

You'd need a rather expensive constant apeture lens (usually only found on
higher end DSLR's) to accomplish your experiment. And, of course, if you
had such a rig, you more than likely wouldn't be overly worried about
limited DOF problems in the first place...

Take Care,
Dudley

 

One of my P&S cameras is f/2.0 at the 36mm wide end and f/2.4 at the 324mm
zoom end. Another has f/2.7 on the 36mm wide end and f/3.5 on the 432mm
zoom end. Just two examples from those I own. In both that is plenty of
focal-length change to select and implement any DOF that I desire and
require. Where on earth do you get your "let's say" figures of f/2.8-f/5.6
for a mere 28mm-130mm (4.5x) zoom range on any P&S camera? Are you
selecting your chosen values based on the severe limitations in one of your
own DSLR pieces of glass? DSLR glass is horrendously poor when it comes to
holding aperture from wide to full zoom. It's a size, weight, and cost
constraint that they'll never get around.

Must you DSLR-Trolls always try to fabricate wild imaginings to support why
you wasted that much money on the body and the expensive accessory glass to
make it at least as functional in image quality as a good P&S camera?

In both of the above mentioned P&S cameras this is less than a 2/3rds EV
change going from wide to full zoom. Well within the realm of an acceptable
exposure margin of error, even if using locked-in manual exposure settings
for any image. Hardly a cause for hysterically running for the hills to go
buy a DSLR body plus the overpriced, always have to keep changing them,
hunks of glass to try to get equivalent images from it. DSLRs frequently
judge the wrong exposure with even more EV offset than that. 1, 2, and even
3 full EV stops wrong exposure frequently common in DSLRs. Look at your own
photos posted in the past as perfect living proof exposure errors as bad.
We've all seen them so you can't lie about it.