Ad
Ad
Ad
Pages: [1]   Bottom of Page
Print
Author Topic: Why its hard to focus DSLR's  (Read 7257 times)
free1000
Sr. Member
****
Offline Offline

Posts: 400


WWW
« on: October 13, 2007, 02:50:50 AM »
ReplyReply

I saw this fascinating sidebar in the BJP on page 14 titled "Deceptive Focus", it seems to explain why its so hard focusing DSLRs with those f1.2 etc lenses.  I wonder if this is right.


Autofocus SLR's, with their split light paths, neeed brighter screens and the advanced designs which replaced ground-glass offered a solution. Prismatic or lenticular, less opaque and interceptive, they ere extra bright. No one bothered to point out that with such a screen, you fit an f/1.4 lens and see f/4 depth of field.

[...]

Using the magnificent 85mm f/1.2 USM lens, I was able to confirm that the depth of field is exaggerated in the viewfinder nevertheless. Try it yourself by putting the camera on a tripod. Focus on an extreme differential focus subject (I used the bars of my window and the garden beyond), and view through the finder. Now check the Live View. The view the SLR finder shows resembles the lens at around f/4.5.

The focusing screen, fresnel and eyepiece combination is reducing the exit pupil of the lens. No wonder so many users - professionals included - are baffled by the unsharpness and limited depth of field present in so many digital shots with high speed lenses used wide open. The viewfinder has been fooling you into thinking more of your subject is sharp.
Logged

@foliobook
Foliobook professional photography folio for iPad
www.foliobook.mobi
n1x0n
Newbie
*
Offline Offline

Posts: 18


« Reply #1 on: October 13, 2007, 06:38:06 AM »
ReplyReply

Quote
No one bothered to point out that with such a screen, you fit an f/1.4 lens and see f/4 depth of field.

A simple DOF calculator will tell how misleading is the above statement. In most cases f-stop, f-stop and a half compensates DOF difference between 1.6 crop and FF.
In case of f1.4 lens - that gives you f2.0 or f2.4, but by no means f4.0.

Quote
it seems to explain why its so hard focusing DSLRs with those f1.2 etc lenses.

If BJP states, that deeper DOF is a problem for focusing, than brighter lenses should make focusing easier, not harder...

And above all - lighter/darker viewfinder is not affecting your DOF in ANY way.
Logged
Tim Gray
Sr. Member
****
Offline Offline

Posts: 2002



WWW
« Reply #2 on: October 13, 2007, 07:07:24 AM »
ReplyReply

Quote
A simple DOF calculator will tell how misleading is the above statement. In most cases f-stop, f-stop and a half compensates DOF difference between 1.6 crop and FF.
In case of f1.4 lens - that gives you f2.0 or f2.4, but by no means f4.0.
If BJP states, that deeper DOF is a problem for focusing, than brighter lenses should make focusing easier, not harder...

And above all - lighter/darker viewfinder is not affecting your DOF in ANY way.
[a href=\"index.php?act=findpost&pid=145691\"][{POST_SNAPBACK}][/a]

a couple of points, first I don't think the OP was raising any kind of issue re ff vs smaller sensor.

second, I take the point of his discussion of the article that a combination of factors accumulate to give you a smaller effective f stop than you would expect given the mechanics of the lens:  "The focusing screen, fresnel and eyepiece combination is reducing the exit pupil of the lens."   That's the point likely subject to being refuted - certainly doesn't make sense to me.  The physics of DOF depends on the f stop (amoung other factors) and doesn't include the amount of light effectively hitting the view finder.
Logged
DiaAzul
Sr. Member
****
Offline Offline

Posts: 777



WWW
« Reply #3 on: October 13, 2007, 07:46:14 AM »
ReplyReply

Quote
The focusing screen, fresnel and eyepiece combination is reducing the exit pupil of the lens. No wonder so many users - professionals included - are baffled by the unsharpness and limited depth of field present in so many digital shots with high speed lenses used wide open. The viewfinder has been fooling you into thinking more of your subject is sharp.

[a href=\"index.php?act=findpost&pid=145673\"][{POST_SNAPBACK}][/a]

Not sure that this isn't stating the obvious and limitations of trying to focus lenses with limited depth of field through the eyepiece. The image in the viewfinder is small and, therefore, it is harder to assess with such a small circle of confusion around the focal plane how much is in focus and out of focus when the image is enlarged to print size. What you are describing is that technology (in the form of live view focusing using the rear LCD) gives a much larger image and better ability to see the impact of the circle of confusion on the resulting image. I don't think this is related to one optical system showing a lens as f/4 versus another showing f/1.4, rather it is our ability to discern circle of confusion impact - i.e. there is a threshold below which which perceive the image in focus when the circle of confusion drops below a certain size. The question would be if the viewfinder image was larger would it be easier to see when things were in focus and out.

I probably haven't explained this particularly well, but in a nutshell what I am saying is that a small DSLR viewfinder is not the best tool for assessing focus with wide aperture/narrow depth of focus and that technology is helping to solve that problem (Live View on the LCD) by giving us a bigger image with which to assess focus.
Logged

David Plummer    http://photo.tanzo.org/
jing q
Sr. Member
****
Offline Offline

Posts: 594


WWW
« Reply #4 on: October 13, 2007, 10:57:19 AM »
ReplyReply

Quote
Not sure that this isn't stating the obvious and limitations of trying to focus lenses with limited depth of field through the eyepiece. The image in the viewfinder is small and, therefore, it is harder to assess with such a small circle of confusion around the focal plane how much is in focus and out of focus when the image is enlarged to print size. What you are describing is that technology (in the form of live view focusing using the rear LCD) gives a much larger image and better ability to see the impact of the circle of confusion on the resulting image. I don't think this is related to one optical system showing a lens as f/4 versus another showing f/1.4, rather it is our ability to discern circle of confusion impact - i.e. there is a threshold below which which perceive the image in focus when the circle of confusion drops below a certain size. The question would be if the viewfinder image was larger would it be easier to see when things were in focus and out.

I probably haven't explained this particularly well, but in a nutshell what I am saying is that a small DSLR viewfinder is not the best tool for assessing focus with wide aperture/narrow depth of focus and that technology is helping to solve that problem (Live View on the LCD) by giving us a bigger image with which to assess focus.
[a href=\"index.php?act=findpost&pid=145696\"][{POST_SNAPBACK}][/a]

however my old pentax camera gives more snap in the focusing when looking through the viewfinder (I can find my focus much more easily)
I don't really understand whether it's a focusing screen issue, all I know is that it used to be easier than what I have to live with nowadays
Logged
samirkharusi
Full Member
***
Offline Offline

Posts: 196


WWW
« Reply #5 on: October 14, 2007, 01:14:18 AM »
ReplyReply

I find that some of the above comments do not incorporate the fundamental thesis in the original article (that I have not read!) which is the issue of Exit Pupil. In case people are not clear as to what this is, let me put it this way. The only light (-cone)that matters in focusing manually is what gets into your eye, not what the lens collects or what the optical train projects. If the pupil of your eye in the ambient illumination is, say, 2mm diameter, but the Exit Pupil of the camera optics (lens+Fresnel screen+viewfinder optics) is 7mm, then your eye will simply not be receiving the full effective aperture of the imaging lens. You will see a DoF that applies to a much smaller f-stop than the nominal value of the f-stop of the lens. The write-up quoted states that the Exit Pupil of the viewfinder system, including the Fresnel screen is such that only around f4 actually comes out of the back end of the viewfinder, not the f1.2 of the imaging lens, never mind what your pupil has opened up to. Hence, with such a viewfinder system it would be impossible to fully appreciate the shallowness of the DoF of fast lenses with your eye, and your focusing will consequently be more sloppy than it should be. IMHO this argument would apply to both film cameras and DSLRs and the only way to focus such a fast lens properly would be to use Live-View, as is now available in the Canon 40D.

Last time I tested the autofocusing in my now ancient 1Ds (for astrophotography) I discovered that the autofocus was extremely reliable with lenses f2.8 or slower, but considerably less so with faster lenses. With a 1.4 lens there were random errors, but of course the focus was always within the DoF of f2.8. Now, in retrospect, perhaps the autofocus system of the 1Ds does have an effective exit pupil equivalent to f2.8? Food for thought.
Logged

Bored? Peruse my website: Samir's Home
richs
Newbie
*
Offline Offline

Posts: 14


« Reply #6 on: October 14, 2007, 10:16:20 PM »
ReplyReply

I've just come across this article, which alludes to the same effect - look at the "Focus Screens" section.

http://www.kenrockwell.com/nikon/50mm-f12.htm

Regards,

Richard
Logged
juicy
Sr. Member
****
Offline Offline

Posts: 254


« Reply #7 on: October 15, 2007, 05:25:56 AM »
ReplyReply

Hi!

My experiences are in line with the OP. I happened to notice this effect recently when comparing photos on the camera's own screen and the viewfinder when shooting with 1Ds and 50mm f1.4. Yesterday after reading this thread I did some informal tests and my conclusions are that the perceived dof in the viewfinder with this prticular camera-lens-combination is similar to the picture taken at f 3.2-f4. You can see lots of detail in the viewfinder that will be completely unrecognisable bokeh-blur in the photo when the lens is (nearly) wide open.

Cheers,
J
Logged
Jonathan Wienke
Sr. Member
****
Offline Offline

Posts: 5759



WWW
« Reply #8 on: October 15, 2007, 09:20:03 AM »
ReplyReply

The image you see in the viewfinder is a low-resolution preview of the captured image; your eye's visual acuity limit and the texture of the focusing screen reduce the detail of the viewfinder image considerably. Not to the same degree as an LCD finder, but you're still losing resolution. Whenever you reduce resolution, you are effectively increasing the CoC size and making it harder to accurately judge DoF and focus accuracy. There's nothing mysterious about it.
Logged

wilburdl
Full Member
***
Offline Offline

Posts: 106


WWW
« Reply #9 on: October 15, 2007, 10:49:21 PM »
ReplyReply

Quote
In a nutshell what I am saying is that a small DSLR viewfinder is not the best tool for assessing focus[a href=\"index.php?act=findpost&pid=145696\"][{POST_SNAPBACK}][/a]

I think that about sums it up.
Logged

Darnell
Editorial Photographer | Cartoonist
darnellwilburn.com
Rob C
Sr. Member
****
Offline Offline

Posts: 12215


« Reply #10 on: October 17, 2007, 03:55:58 PM »
ReplyReply

All this confirms my own opinion that a split-image viewfinder is the most accurate.

For some Godforsaken reason, Nikon did not produce a split-image screen with grid lines which was suitable for fast lenses; the one they DID produce was for slow lenses only. As much of my work involved clear-cut horizons, there was no contest (pre-digital) because sloping horizons wasted transparencies and who would accept such images, regardless of how good the main subject was. So, the choice was absolutely crisp models or level horizons. In the interest of the the horizons, the models were more or less crisp depending on how my eyes felt when I was using the groundglass (instead of the split-image).

How silly not to produce a screen with the best of both worlds for fast optics!

That the split-image worked I have no doubts: other, horizon free shots were always spot-on crisp when I used it.

Rob C
Logged

wturber
Newbie
*
Offline Offline

Posts: 8


« Reply #11 on: October 17, 2007, 06:43:25 PM »
ReplyReply

Quote
That's the point likely subject to being refuted - certainly doesn't make sense to me.  The physics of DOF depends on the f stop (amoung other factors) and doesn't include the amount of light effectively hitting the view finder.
[{POST_SNAPBACK}][/a]

Sure.  But DoF does depend on things like whether all the light hitting the focus screen gets sent to your eye.  As nearly as I can tell (I'm still looking for a very good technical illustration) modern "bright" SLR focus screens are very efficient at sending the light received from the just a portion of the exit pupil of a lens.  That portion is usually either f/2.8 or slower, or f/4 or slower. These focus screens, however, are very inefficient at sending the light from larger (faster) exit pupils.

The light from the faster lens gets filtered out because it hits the focus screen at a steeper angle or because the microlenses have a limited FoV.  I'm not sure which. But ether way, the light from the faster portion of the exit pupil gets blocked or deflected and doesn't make it to your viewing eye.  And it is this light that shows the greatest amount of defocus.  If you don't see that light, you don't see the defocus.  In other words, you see greater DoF than the lens will deliver to the film or sensor.

Every focus screen has an f-number where it becomes efficient.  You can figure this out for yourself by using either a fast old manual lens or by using the DoF preview on your camera.  Note when the image in the finder gets noticably dimmer.  For most DSLRs, that's around f/4.  For more upmarket cameras, it might be around f/2.8 - problably because buyers of those cameras are more likely to purchase faster lenses.

Here are a couple links to two example images that show this clearly happening.  The top row are the images as seen through the viewfinder of an E-500 (Taken with a C7070 at much greater effective magnification than we can see with our eyes alone)   The bottom row are what the actual images look like when the picture is taken at that aperture.

[a href=\"http://www.jayandwanda.com/photography/ViewfinderDOF2.jpg]http://www.jayandwanda.com/photography/ViewfinderDOF2.jpg[/url]
http://www.jayandwanda.com/photography/viewfinderDOF2b.jpg

Here's a page I wrote up on the issue and how it relates to manual focusing.

Here's a graph put together by Larry J. Clark showing the non-linearity of the E-1 and the E-500 focus screens.  You can see that the E-1 is optimized for about f/2.8 and the E-500 for about f/4.  Neither screen gives you much extra brightness even when opening a lens up by a full two stops from f/2.8 to f/1.4

http://www.jayandwanda.com/photography/dsl...ewBright-3c.jpg
Logged
wturber
Newbie
*
Offline Offline

Posts: 8


« Reply #12 on: October 17, 2007, 06:46:29 PM »
ReplyReply

Quote
I probably haven't explained this particularly well, but in a nutshell what I am saying is that a small DSLR viewfinder is not the best tool for assessing focus with wide aperture/narrow depth of focus and that technology is helping to solve that problem (Live View on the LCD) by giving us a bigger image with which to assess focus.
[a href=\"index.php?act=findpost&pid=145696\"][{POST_SNAPBACK}][/a]

That's not it.  The issue is the use of microlenses in the so-called "focus screen" in the DSLR.  You can put a magnifying aid on your viewfinder and all you end up with is a magnified image that shows DoF greater than it should (when using fast lenses).
Logged
wturber
Newbie
*
Offline Offline

Posts: 8


« Reply #13 on: October 17, 2007, 06:52:34 PM »
ReplyReply

Quote
The image you see in the viewfinder is a low-resolution preview of the captured image; your eye's visual acuity limit and the texture of the focusing screen reduce the detail of the viewfinder image considerably. Not to the same degree as an LCD finder, but you're still losing resolution. Whenever you reduce resolution, you are effectively increasing the CoC size and making it harder to accurately judge DoF and focus accuracy. There's nothing mysterious about it.
[a href=\"index.php?act=findpost&pid=146106\"][{POST_SNAPBACK}][/a]

Then why do you see correct DoF representations when using an LCD and liveview or when reviewing the image on the LCD, but you see it wrong when using a focus screen?  I agree that there is nothing mysterious about it.  But your explanation is not correct.  The cause is the design of the focus screen which was introduced as a means to compensate for the dimmer relatively slow normal zooms that replaced the typical faster primes as cameras.  The advent of AF and the loss of some light to a partially silvered reflex mirror also added to the motivation to use a different kind of focus screen.  And with the use of AF, there wasn't as great a reliance on the "focus" screen as there once was.  And if you were using a slow lens anyway, the focus screen wasn't lying to you anyway.  So lots of people never really noticed the change.
Logged
wturber
Newbie
*
Offline Offline

Posts: 8


« Reply #14 on: October 17, 2007, 06:56:00 PM »
ReplyReply

Quote
All this confirms my own opinion that a split-image viewfinder is the most accurate.

<snip>

That the split-image worked I have no doubts: other, horizon free shots were always spot-on crisp when I used it.

Rob C
[a href=\"index.php?act=findpost&pid=146727\"][{POST_SNAPBACK}][/a]

Depending on how fast you are shooting, this is where liveview can really help out.  You get the most precise manual focus and the ability to overlay grid lines.

If you really need the optical finder, I'd bet you could get someone like KatzEye Optical to scribe some lines for you. Either that, or use a bubble level on the tripod or camera.
Logged
Rob C
Sr. Member
****
Offline Offline

Posts: 12215


« Reply #15 on: October 18, 2007, 03:08:39 AM »
ReplyReply

Quote
Depending on how fast you are shooting, this is where liveview can really help out.  You get the most precise manual focus and the ability to overlay grid lines.

If you really need the optical finder, I'd bet you could get someone like KatzEye Optical to scribe some lines for you. Either that, or use a bubble level on the tripod or camera.
[a href=\"index.php?act=findpost&pid=146769\"][{POST_SNAPBACK}][/a]

Well, not really: the beach shots were mostly hand-held ones of models bouncing about and a tripod only became possible/essential when using long lenses. I donīt know that ANY size of focussing screen can compete with a split image for accuracy - all screens depend on the viewerīs eyesight and the ability to say stop! no more backwards or forwards, thatīs it!  - but a split is just so obvious (to me).

Iīm sure that there is no argument with landscape and a focusing screen - tripods are obviously the way to go because there are few problems with the fleeting moment as far as setting up the camera goes; I do not, of course, include the effects of transient light here!

Rob C
Logged

Gregory
Full Member
***
Offline Offline

Posts: 191


WWW
« Reply #16 on: October 18, 2007, 05:48:18 AM »
ReplyReply

if this is true, the problem for me personally would be that f3.2 or f4 in the viewfinder doesn't allow me to see the image sharply enough to accurately judge when the focus is the best it can be rather than judging if the DOF is what I need in the image. a narrower DOF would allow me to more accurately judge the focus than a wider DOF.

(question. is the article saying that the current digital camera architecture defines a minimum effective aperture; eg, always ~f4 for a defined camera; or is it saying that the architecture increases the effective aperture of the lense in use at the time; eg, by one or two stops; regardless of the lense's largest aperture setting?)

I also agree that the graininess of the focus planes would decrease resolution and hence make it even harder to focus manually accurately.

there have been many times when I've focused to the best of my ability feeling that it was accurate in the viewfinder only to find that it wasn't that accurate in the processed image on screen. I never had that problem with my Canon 620 (although I had 20-20 vision back then ;-). I also find myself frequently rocking the focus back and forth multiple times in an attempt to find the exact focus point because it simply isn't clear in the viewfinder.

split-screen focus screens wouldn't help me. I prefer to frame and focus at the same time. a split-screen requires the subject to intersect the split.

does Canon or another manufacturer have clearer; i.e., not grainy; focus screens for manual focus use with the Mark III ?  this wouldn't solve the problem entirely but I feel it would be a big step forward.

regards,
Gregory
« Last Edit: October 18, 2007, 05:53:11 AM by Gregory » Logged

Gregory's Blog: An Aussie in HK
Equipment: Canon EOS 1D Mark III, 17-40L, 24-105L, 70-300 DO
Rob C
Sr. Member
****
Offline Offline

Posts: 12215


« Reply #17 on: October 18, 2007, 10:59:41 AM »
ReplyReply

Greg - you might have a theoretical point about split-screen focussing, but in practical terms it works just fine for me!

Neither am I at all certain that finer grained glass makes for easier focus; I think that the grain seems to act in a sort of unintended, but overall micro-prism way. But Iīm no expert on screens, the only experience I have of them is with Exakta, Nikon, Rollei, Hassleblad, Mamiya and Pentax for roll-film/35mm and Sinar and MPP on 4x5. How people can use LF of their own free will I fail to understand. I got off that bus the moment I became self-employed! Naturally, that worked for me because I then had no interest to speak of in camera movements... other than to prevent them, of course!

If I may just go back to the split-screen v. ground-glass thing for a moment: my F3 has both the split-screen as well as the grid screens available in the bag; the D200 comes with the possibility of a permanent grid if you want it. I do. However, the debit side of this is that I do not find it very comfortable using the combination of grid and little light to indicate best focus (I do not use AF lenses) and it certainly isnīt very fast, just very distracting.

Rob C
Logged

wturber
Newbie
*
Offline Offline

Posts: 8


« Reply #18 on: October 18, 2007, 01:01:46 PM »
ReplyReply

Quote
Iīm sure that there is no argument with landscape and a focusing screen - tripods are obviously the way to go because there are few problems with the fleeting moment as far as setting up the camera goes; I do not, of course, include the effects of transient light here!

Rob C
[a href=\"index.php?act=findpost&pid=146866\"][{POST_SNAPBACK}][/a]


For what you are describing, yes.  For instance, I use a split when shooting birds with a long manual focus telephoto.  For that, the spit is the best compromise.

But for some subjects, the split doesn't always work so well.  If you have the time, nothing I've found beats liveview from the camera sensor at 10x on a DSLR.  At 10x, the differences in user's eyesight becomes essentially a non-issue.  But that wouldn't be any good for your beach shots.
Logged
wturber
Newbie
*
Offline Offline

Posts: 8


« Reply #19 on: October 18, 2007, 01:06:48 PM »
ReplyReply

Quote
If I may just go back to the split-screen v. ground-glass thing for a moment: my F3 has both the split-screen as well as the grid screens available in the bag; the D200 comes with the possibility of a permanent grid if you want it. I do. However, the debit side of this is that I do not find it very comfortable using the combination of grid and little light to indicate best focus (I do not use AF lenses) and it certainly isnīt very fast, just very distracting.

Rob C
[{POST_SNAPBACK}][/a]

KatzEye Optics may have a solution for you.

[a href=\"http://www.katzeyeoptics.com/cat--Nikon-DSLRs--cat_nikon.html]http://www.katzeyeoptics.com/cat--Nikon-DS...-cat_nikon.html[/url]

They are an excellent company that is service oriented.  If they have what you need, I can easily recommend them.
Logged
Pages: [1]   Top of Page
Print
Jump to:  

Ad
Ad
Ad