When I take pictures of flowers, I often try to get the classic soft background around the in focus flower.

I have a canon t3i, and a lens that goes to f/2.8. I set that to wide open, take my photos, and get results I like.

However, whenever I am looking through the viewfinder, it always seems to appear with a much wider depth of field than the final images. Is that true, or am I just crazy? I thought that the whole point of the SLR type camera was so that what you see in the viewfinder is what you get. Is that just for framing and composition, or should it reflect the DOF accurately as well? Since I set the aperture to its max f/2.8, it shouldn't be any previewing behavior.

So am I really crazy, or is there a difference in DOF between the viewfinder and final image?

  • 1
    This is something I constantly hear but it actually wrong. DSLRs show the focus-plane and composition (if the viewfinder has 100% coverage) only. They are actually further from WYSIWYG than fixed lens cameras which show composition, exposure, white-balance and focus (with the latest EVFs which are sharp enough). – Itai Oct 22 '12 at 3:53
  • 2
  • 1
    I don't think this is a duplicate of the other question, that one is about inaccuracies while using the DOF preview button in a variety of apertures, and the way info from there applies to this isn't totally obvious. – whatsisname Oct 22 '12 at 14:16
  • 1
    When you're shooting wide open, the viewfinder is showing you DOF preview all the time because you're already looking through the final aperture. – Imre Oct 22 '12 at 16:54
  • 2
    Tagged the wrong duplicate. Try: Why can't I see bokeh in the viewfinder?. I swear there was another one even more close, but I can't find it. – mattdm Oct 22 '12 at 17:23

As Mike Johnston explains the viewfinders, the focusing screens used in modern cameras tend to be made super-bright in order to accommodate the slow zoom lenses that often get connected, but this brighter construction (like an array of miniature lenses, or a bunch of very short fiber-optic cables) also makes much of the image to appear in focus, much more so than on final image. So the apparent depth of field is larger than on resulting photo by design.

If you want a truer representation, use Live View or a custom focusing screen (tuned for manual focusing, usually with split prisms).


No. Depth-of-field is dependent on viewing size. Technically, the angle-of-view covered by the image when you see because visual acuity is determined by angular-resolution (Thanks to @Imre for pointing that out). Since the image in your viewfinder covers a smaller angle of view as your final image, there will always be a discrepancy.

Your T3i has a 0.85X magnification viewfinder which should cover about 27° diagonally. If you look at your final image on a print or screen, it usually covers a larger field-of-view.

Viewing DOF on the LCD in Live-View gives a better approximation. You are right that the viewfinder shows the image at maximum aperture (unless you press the DOF Preview button).

What is important to understand is that DOF is not a hard limit where things are in focus on one side and not on the other Things get gradually out of focus away from the focus plane. At one point, they become so out-of-focus that we consider them outside the DOF. However, this is based on perception which is dependent on viewing size. The larger the print, the less apparent DOF.

Even DOF field tables and calculators use a built-in factor to determine what to consider too much out-of-focus. Traditionally this has been an 8x10" print. Some software let you specify your own parameter for this.

EDIT For Completeness:

There is another factor which affects the difference in DOF between the OVF and final images and that is the focusing screen. This affects the perception at large apertures only but how large is dependent on several factors.

On a Pentax K-5 with default focusing screen, I can see differences in DOF down to F/2 and they correspond to my expectations when I see the image large. On a Nikon D600, I can see differences down to F/2.8. Those are just the two cameras on my desk right now based on experience, most DSLRs should fall withing those limits with their stock focus-screens.

  • 1
    While true, it may not necessarily apply. The simple fact is that (on most reasonably current SLRs) the depth of field you see in the viewfinder is radically different from what you're going to get in the picture (and that remains true even when/if you activate DoF preview). – Jerry Coffin Oct 22 '12 at 3:59
  • 1
    Yes, is it. I see there is another explanation for very bright lenses but the asker here is using an F/2.8 one. The difference in the view is most likely perceptual due to a much smaller view in the OVF than on the final print size. – Itai Oct 22 '12 at 12:53
  • Would it be rude to ask if you've ever printed an image the size of your viewfinder? :) On the only SLR for which I have a F/1.4 lens with me, I can see DOF differences to down to F/2. With another I can see what seems like the correct DOF at F/2.8. I could check the other 3 but the point would be that perception of DOF depends on the focusing screen, viewfinder and lens. – Itai Oct 22 '12 at 15:11
  • 1
    "Depth-of-field is dependent on viewing size" - shouldn't it be field of view occupied by the image (which depends on viewing distance, in addition to size)? The DOF calculators also include viewing distance, because angular resolution is what determines whether a viewer will perceive something sharp or not. Stamp-sizedness of an image is not such a big deal when viewing distance has been adjusted accordingly to be half an inch. – Imre Oct 23 '12 at 8:09
  • @Imre You are correct that angular resolution is what is used for DOF tables. However, I am not sure if it still applies at close focus distances? Informally trying it seems I can see much more details when things are at least a several inches away, compared to closer-but-still-in-focus. – Itai Oct 23 '12 at 13:24

Yes, it's quite common for the picture to have substantially less depth of field than you see in the viewfinder. I'm not sure specifically about the T3i, but for at least some cameras you can get a different viewfinder screen that diffuses the light more, so it shows depth of field more accurately. For what it's worth, that also tends to make manual focusing easier (things that are out of focus look out of focus). The downside is that the view through the viewfinder gets somewhat dimmer.


Here's a demonstration of what I'm talking about. These are taken with a Sony Alpha A900, but the principle is the same. Both are taken with an f/2.8 lens at f/2.8. In both cases, I focused so that in the viewfinder, the 5 in 135 was somewhat blurry, but still readable. Here's what I got from doing that with the stock ("type G") screen in the viewfinder:

enter image description here

Here's what I got when I used a viewfinder screen with more diffusion (what Sony calls "type L"):

enter image description here

I'll repeat: In both cases, I used manual focusing, adjusted so that the 5 in 135 (just before the big 36) was (somewhat) blurred but still readable when viewed in the viewfinder. In both cases, I was using an f/2.8 lens at f/2.8. In both cases, we're dealing with what I saw through the viewfinder, with an f/2.8 lens at f/2.8, not a faster lens, nor using DoF preview, nor comparing the view through the viewfinder to a (larger) print.

So, one more time, all together on the count of three now: yes, the viewfinder screen does affect the apparent depth of field, even with an f/2.8 lens, even when it's wide open.

  • 2
    With the T3i and most other Canon DSLRs I know of changing focusing screen only has an effect on lenses wider than f/2.8 – Matt Grum Oct 22 '12 at 8:45
  • 1
    @whatsisname: If I hadn't changed the focus, then yes, they'd be identical. What I did, however, what changed the focus so that in both cases, what I saw in the viewfinder was the 5 in 135 being visibly blurred, but still readable. – Jerry Coffin Oct 22 '12 at 16:25
  • 1
    Any chance you have a second camera that could see into the viewfinder of the first? – DHall Oct 22 '12 at 19:49
  • 1
    Jerry, just to spell things out here: The exercise is to start from focused to infinity and then slowly dial the the focus in until the edge of the apparent depth of field hits a certain point in the image. In your photos, one can see that this happens at different places, when the only factor changed is the viewfinder screen. – mattdm Oct 22 '12 at 20:45
  • 2
    @mattdm: In each case I focused so the near end of the DoF (as it appeared in the viewfinder) just about included the 5 of 135 (I.e., it was visibly blurred, but still readable). With the "L" screen, the apparent DoF matched that in the picture pretty closely (the 5 is blurred, but readable). With the stock ("G") screen, that 5 was readable in the viewfinder, but isn't even close in the picture (and I'm confident that reducing the reproduction size won't make it readable either). – Jerry Coffin Oct 22 '12 at 21:07

I believe the issue you're seeing is that the viewfinder cannot "stop down" as far as the lens installed on your camera. I know from experience that the minimum aperture of my viewfinder ( Nikon D90 ) is around f/2.8. Therefore, any time I install a lens with a minimum aperture of less than f/2.8, I'm going to have to guess the DOF. This can be easily determined by installing a fast lens on your camera, and progressively stopping down the aperture and hitting the DOF preview button, while framing something in the viewfinder. The viewfinder will register no change in brightness or DOF until I hit f/2.8 on the aperture. If your camera has the same limitation, then you will never see any DOF preview changes until you hit at least f/2.8 on the lens. As mentioned in other answers, your aperture is wide open until you release the shutter. The light is simply being reduced through your viewfinder, with the same effect as stopping down the aperture. @Ital and @mattdm's link explains this much more succinctly.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.