Apparent Depth of field with Canon 20D

Folks,

I recently did a shot where I bracketed f-stops to tailor depth of
field. The shot was of eye glasses sitting on a page of print. I was
surprised to see that the viewfinder image appears to have way more
depth of field than the actual photographed image (20D with 50mm f
1.4). With the lens wide open (f/1.4), I could almost read the print on
the page (focus was above the plane of the page on the eyeglasses) when
looking through the viewfinder. On the photographed image, the page was
a total blur. It seems the viewfinder appeared to have significantly
greater DOF than the actual image. Any ideas why this would be?

W wrote:
Folks,

I recently did a shot where I bracketed f-stops to tailor depth of
field. The shot was of eye glasses sitting on a page of print. I was
surprised to see that the viewfinder image appears to have way more
depth of field than the actual photographed image (20D with 50mm f
1.4). With the lens wide open (f/1.4), I could almost read the print on
the page (focus was above the plane of the page on the eyeglasses) when
looking through the viewfinder. On the photographed image, the page was
a total blur. It seems the viewfinder appeared to have significantly
greater DOF than the actual image. Any ideas why this would be?

Can you post the photo and link to it in this thread? IAE, tho, I
suspect it's the proper working of the human eye in play here.....

--
John McWilliams

Jim Townsend wrote:
W wrote:

Folks,

I recently did a shot where I bracketed f-stops to tailor depth of
field. The shot was of eye glasses sitting on a page of print. I was
surprised to see that the viewfinder image appears to have way more
depth of field than the actual photographed image (20D with 50mm f
1.4). With the lens wide open (f/1.4), I could almost read the print on
the page (focus was above the plane of the page on the eyeglasses) when
looking through the viewfinder. On the photographed image, the page was
a total blur. It seems the viewfinder appeared to have significantly
greater DOF than the actual image. Any ideas why this would be?

I'm not quite sure I follow you...

Note that when you look through the lens, it's always wide open and
that's the DOF you'll see. The lens iris doesn't shrink to your
selected aperture until you press the shutter release.

You need to use the DOF preview button on the side of the camera
to see what the DOF will look like at smaller apertures.


What you say is true, but that simply makes his observation harder to
explain.
Dave Cohen

W wrote:
Folks,

I recently did a shot where I bracketed f-stops to tailor depth of
field. The shot was of eye glasses sitting on a page of print. I was
surprised to see that the viewfinder image appears to have way more
depth of field than the actual photographed image (20D with 50mm f
1.4). With the lens wide open (f/1.4), I could almost read the print on
the page (focus was above the plane of the page on the eyeglasses) when
looking through the viewfinder. On the photographed image, the page was
a total blur. It seems the viewfinder appeared to have significantly
greater DOF than the actual image. Any ideas why this would be?


Do Canon lenses work like Nikon?
AFAIK with Nikon the maximum aperture "at rest" is f2.8 (or is it f2?) -
if you look at a f1.4 lens, when set at f1.4 the aperture blades are
closed slightly - when you press the DOF preview button - or make an
exposure the aperture opens to the fully wide f1.4 position.

Folks,

I'm afraid to say this may have been a false alarm. It appears that
some of the shots were not properly focused (OOPS....sorry). I would
have to carefully repeat this experiment. If/when I do, I will report
back here.



frederick wrote:
W wrote:
Folks,

I recently did a shot where I bracketed f-stops to tailor depth of
field. The shot was of eye glasses sitting on a page of print. I was
surprised to see that the viewfinder image appears to have way more
depth of field than the actual photographed image (20D with 50mm f
1.4). With the lens wide open (f/1.4), I could almost read the print on
the page (focus was above the plane of the page on the eyeglasses) when
looking through the viewfinder. On the photographed image, the page was
a total blur. It seems the viewfinder appeared to have significantly
greater DOF than the actual image. Any ideas why this would be?


Do Canon lenses work like Nikon?
AFAIK with Nikon the maximum aperture "at rest" is f2.8 (or is it f2?) -
if you look at a f1.4 lens, when set at f1.4 the aperture blades are
closed slightly - when you press the DOF preview button - or make an
exposure the aperture opens to the fully wide f1.4 position.

W wrote:
Folks,

I recently did a shot where I bracketed f-stops to tailor depth of
field. The shot was of eye glasses sitting on a page of print. I was
surprised to see that the viewfinder image appears to have way more
depth of field than the actual photographed image (20D with 50mm f
1.4). With the lens wide open (f/1.4), I could almost read the print on
the page (focus was above the plane of the page on the eyeglasses) when
looking through the viewfinder. On the photographed image, the page was
a total blur. It seems the viewfinder appeared to have significantly
greater DOF than the actual image. Any ideas why this would be?

I observed this with the 50mm 1.8 on 20d. I think the reason might be
that the focusing screen lets some of the image forming rays pass thru
uninterrupted.

To approximate, the image of the 50mm in the viewfinder of a DSLR is
magnifyed 1:1 so the viewfinder also has 50mm focal length. This means
a very limited actual aperture of say 5mm (taking the iris in account),
if there was no focusing screen. This gives an actual f10, not f1.8!
I tryed that on some trees in the distance while focusing on 0.45m.
There were no trees in the photo at 1.8 at all, while at f10, the shape
resembled what I saw in the viewfinder only not in such detail.

This means it's a combination of the two "images" we are observing in
the viewfinder, so it should appear sharper than the real image formed,
and it realy does.

That makes sense, the majority of light rays form the image on the
screen, but the ones that pass are still enough to increase the DOF.

Okay, I repeated this experiment using careful focusing, mirror lockup,
and cable release. The effect I saw was indeed real. I shot at f/1.4
with a 50mm f1.4 lens. The shot was of a pair of folded glasses sitting
on a page of small text. The focus was on the glasses probably about an
inch to inch and a half above the page. The camera was 2 to 3 feet away
from the subject. I could read the text when looking through the
viewfinder but it was out of focus and unreadable in the captured
image.
I am not sure how to post the image here. Also, that would be of little
use because I cannot post what I actually see through the viewfinder.


W wrote:
Folks,

I'm afraid to say this may have been a false alarm. It appears that
some of the shots were not properly focused (OOPS....sorry). I would
have to carefully repeat this experiment. If/when I do, I will report
back here.



frederick wrote:
W wrote:
Folks,

I recently did a shot where I bracketed f-stops to tailor depth of
field. The shot was of eye glasses sitting on a page of print. I was
surprised to see that the viewfinder image appears to have way more
depth of field than the actual photographed image (20D with 50mm f
1.4). With the lens wide open (f/1.4), I could almost read the print on
the page (focus was above the plane of the page on the eyeglasses) when
looking through the viewfinder. On the photographed image, the page was
a total blur. It seems the viewfinder appeared to have significantly
greater DOF than the actual image. Any ideas why this would be?


Do Canon lenses work like Nikon?
AFAIK with Nikon the maximum aperture "at rest" is f2.8 (or is it f2?) -
if you look at a f1.4 lens, when set at f1.4 the aperture blades are
closed slightly - when you press the DOF preview button - or make an
exposure the aperture opens to the fully wide f1.4 position.

Very interesting. I am not sure I understand the explanation. How do
uninterrupted rays get focused in the eye?

Bhogi wrote:
W wrote:
Folks,

I recently did a shot where I bracketed f-stops to tailor depth of
field. The shot was of eye glasses sitting on a page of print. I was
surprised to see that the viewfinder image appears to have way more
depth of field than the actual photographed image (20D with 50mm f
1.4). With the lens wide open (f/1.4), I could almost read the print on
the page (focus was above the plane of the page on the eyeglasses) when
looking through the viewfinder. On the photographed image, the page was
a total blur. It seems the viewfinder appeared to have significantly
greater DOF than the actual image. Any ideas why this would be?

I observed this with the 50mm 1.8 on 20d. I think the reason might be
that the focusing screen lets some of the image forming rays pass thru
uninterrupted.

To approximate, the image of the 50mm in the viewfinder of a DSLR is
magnifyed 1:1 so the viewfinder also has 50mm focal length. This means
a very limited actual aperture of say 5mm (taking the iris in account),
if there was no focusing screen. This gives an actual f10, not f1.8!
I tryed that on some trees in the distance while focusing on 0.45m.
There were no trees in the photo at 1.8 at all, while at f10, the shape
resembled what I saw in the viewfinder only not in such detail.

This means it's a combination of the two "images" we are observing in
the viewfinder, so it should appear sharper than the real image formed,
and it realy does.

That makes sense, the majority of light rays form the image on the
screen, but the ones that pass are still enough to increase the DOF.

"W" writes:
Okay, I repeated this experiment using careful focusing, mirror lockup,
and cable release. The effect I saw was indeed real. I shot at f/1.4
with a 50mm f1.4 lens. The shot was of a pair of folded glasses sitting
on a page of small text. The focus was on the glasses probably about an
inch to inch and a half above the page. The camera was 2 to 3 feet away
from the subject. I could read the text when looking through the
viewfinder but it was out of focus and unreadable in the captured
image.

What kind of focusing screen does the camera have? If it's not a ground
glass, you may not be seeing the effect of all of the light that comes
through the lens. If light from the outer edges of the lens doesn't
actually make it into your eye, the effect is as if the lens had a
smaller aperture (and more DOF). So what you see doesn't match what
the film sees.

I am not sure how to post the image here. Also, that would be of little
use because I cannot post what I actually see through the viewfinder.

You *could* take a picture of what's seen in the viewfinder by using a
second camera to shoot into the viewfinder.

Dave

The focusing screen is the standard 20D screen (not removable). Not
sure exactly the type but it does not appear to be 'pure' ground glass
screen.
Taking a picture through the viewfinder that's too hard . But I can
assure you I can read the text through the viewfinder and not on the
captured image.

Dave Martindale wrote:
"W" writes:
Okay, I repeated this experiment using careful focusing, mirror lockup,
and cable release. The effect I saw was indeed real. I shot at f/1.4
with a 50mm f1.4 lens. The shot was of a pair of folded glasses sitting
on a page of small text. The focus was on the glasses probably about an
inch to inch and a half above the page. The camera was 2 to 3 feet away
from the subject. I could read the text when looking through the
viewfinder but it was out of focus and unreadable in the captured
image.

What kind of focusing screen does the camera have? If it's not a ground
glass, you may not be seeing the effect of all of the light that comes
through the lens. If light from the outer edges of the lens doesn't
actually make it into your eye, the effect is as if the lens had a
smaller aperture (and more DOF). So what you see doesn't match what
the film sees.

I am not sure how to post the image here. Also, that would be of little
use because I cannot post what I actually see through the viewfinder.

You *could* take a picture of what's seen in the viewfinder by using a
second camera to shoot into the viewfinder.

Dave