How much more bokeh do longer focal lengths produce at the same field of view?

Discussion in 'Digital Photography' started by Chris7, Apr 11, 2010.

  1. Chris7 macrumors 6502

    Joined:
    Aug 8, 2008
    Location:
    Lost in Thought
    #1
    Thanks to those who recommended a couple of Brian Peterson’s books to me. I have found them very interesting.

    In Learning to See Creatively, Brian Peterson instructs that one will obtain more background and foreground blur (bokeh?) outside the focal plane by using (in his example) a zoom when set at 135mm verses the same zoom when set at 60mm, at the same aperture and the same field of view (p. 41). (Obviously, the depth of field narrows the more you zoom, but he is talking about maintaining exactly the same field of view, backing up from the focal plane as you zoom in).

    As I play with depth of field calculators, it appears that the depth of field at a particular aperture is almost exactly the same when shooting wide and up close, or zoomed in and far away, so long as the field of view is the same.

    This seems to imply that one could stop down a little and get the same bokeh but with a deeper depth of field, simply by using longer focal lengths and backing up from the focal plane. (I imagine some of it has to do with lens construction, but Brian Peterson is talking about using the same zoom lens set to different focal lengths).

    So my question is, how much could I stop down and get the same bokeh when using longer focal lengths at the same field of view?

    For example, going from a EF 50mm f/1.4 USM to a EF 85mm f/1.8 USM (on a 1.6x crop sensor camera). Could I get the same bokeh with the 85mm at f/2 as the 50mm at f/1.4, while increasing the DOF slightly (at the same FOV)?

    How about going from an EF 50mm f/1.4 USM to a EF 24-105mm f/4L IS USM @ 100mm (50mm equivalent:160mm)?

    BTW, I’m not so interested in the “quality” of the bokeh. I just want the background and foreground to look out of focus.

    Thanks in advance. I'll be watching this thread, but may not have time to reply for a couple of days.
     
  2. toxic macrumors 68000

    Joined:
    Nov 9, 2008
    #2
    "bokeh," by definition, is the quality of out-of-focus blur. it is a lens characteristic, like resolution or flare resistance or distortion.

    what you are asking is how much more blur there is, or how much more diffuse the background is. this has to do with the size of the blur disk. the size of the blur disk depends only on aperture, the size of the diaphragm opening.

    aperture = focal length / f-number. a 50mm lens at f/1 yields the same amount of blur as a 100mm lens at f/2, or a 200mm lens at f/4.

    to answer your examples, 85mm at f/2 has about the same amount of blur as 50mm at f/1.2, and 100mm at f/4 is the same as 50mm at f/2.
     
  3. alebar14 macrumors regular

    alebar14

    Joined:
    Jul 14, 2007
    Location:
    Auckland CBD, New Zealand
    #3
    I was just thinking about buying Canon 85 mm f/1.8 lens. I am still unsure,because I have 50 mm f/1.4 for my Canon 40D (this lens will be equal -> 80 mm on my 1.6x crop factor). And as I use the prime lens for portrait shoot, will 85 1.8 be any useful (considering distance of shooting area) ?

     
  4. gødspeed macrumors regular

    gødspeed

    Joined:
    Jun 11, 2009
    Location:
    Oregon
    #4
    The 85mm f/1.8 is a great lens, but you're right that it's pretty long on an ASPC sensor -- equivalent to 136mm on a 35mm sensor. The further you are from your subject, the deeper the depth of field. I'm not sure what the exact equation is, but it is an exponential function. Another concern would be that to get a full portrait figure in the frame, you'd need to be standing quite a ways away, which is a limitation in many circumstances (indoors, urban environments, etc).

    I'd say stick with the 50mm. It's substantially faster, and the focal length is more common for portrait shots.
     
  5. Chris7 thread starter macrumors 6502

    Joined:
    Aug 8, 2008
    Location:
    Lost in Thought
    #5
    I am a total beginner and so usually do not answer questions here. But one thing I have found useful is to figure out what field of view and depth of field I want at a particular distance from the subject. I have not found a field of view calculator (please post if you find one), but I have found it useful to use a DOF calculator, keeping in mind that, at 4.3 ft. from the focal plane the FOV is about 3 ft. You can find the rest of what you want from there. Toxic's post is making me lean towards a prime with a longer focal length and backing up a bit from the subject for portraits, but that's just me.
     
  6. toxic macrumors 68000

    Joined:
    Nov 9, 2008
    #6
    first, forget about crop factors and "equivalent" focal length. they are useful only when you want to take the same photograph with a different format. for example, let's say you want to retake a photograph that was taken with a 80mm lens on medium format film. you know the exact location for where to stand, you just need to know what FL will give you the same FoV. if you are on 35mm, you divide by 1.6 and end up with a 50mm lens. if you are on APS-C you divide by 1.6 again and end up with 30mm.

    another important distinction is that perspective varies only with distance. you cannot fill the frame with a 50mm lens on a 1.6 crop body and have a flattering headshot. you must either stand farther away and crop later, or use an 85mm lens or longer to force you to the proper subject distance.

    for a flattering perspective, you must stand at least ~6ft/2m away (something like that, the point is there is a minimum distance for a given size subject). once you know how far away you'd like or need to work, you can pick your lens to give you the field of view (and thereby composition) you want. if you want a head-&-shoulders, you'd probably use something around 50mm. for a half-body, probably 30mm, etc.

    FoV (the area covered) can be calculated pretty easily if you just look up the AoV (angle covered) of a different FL's...but no, I don't know of a convenient calculator.

    and just in case you don't know, DoF calculators are calculated based on some (relatively low) standard for visual acuity and a print size of 8x10, so they're not really that useful in practical applications. the only way to know what settings you should use is to take photos, find which have the amount of DoF you want, and note your settings and distance in a table.

    just keep in mind that the perspective will change.

    I should add that for my previous post, you can only compare amount of blur in the same format...so no 35mm vs crop comparisons just based on aperture.
     
  7. Chris7 thread starter macrumors 6502

    Joined:
    Aug 8, 2008
    Location:
    Lost in Thought
    #7
    Out of focus blur on a cops sensor vs. full size

    So the pattern is, every time you double the focal length, you can stop down by two and get equivalent out of focus blur (while also conveniently doubling your DOF)?
    I was thinking about this. Is seems like you’re saying that it the focal length of the actual lens only that follows this pattern, which would make intuitive sense. Let me see if I’ve got this right…

    For example, a body with a full sized sensor with a 100mm lens at f/4 will yield a DOF of about .5 ft at 8.6 ft. from the focal plane.

    A 1.6x crop body with a 62.5mm lens at f/2.4 will also yield a DoF of about .5 ft. at 8.6 ft. from the focal plane (35mm equivalent of 100mm, same FoV).

    But as far as out of focus blur on the 1.6x crop body, I would have to open up the aperture to about f/1.4 to get the equivalent out of focus blur as the full sized sensor body at f/4 (I subtracted another 1 ½ stop to compensate for the shorter focal length).

    Is this about right?

    BTW, what book would you recommend that would explain this stuff?
     
  8. toxic macrumors 68000

    Joined:
    Nov 9, 2008
    #8
    never thought of it that way, but I guess...

    right.

    no. the blur between the first condition, 62.5mm at f/2.5 and 100mm at f/4, would yield similar results. I'm just not sure if it's exactly the same...maybe it is.

    on second thought, I think the amount of blur is exactly the same, or at least it should be. DoF is a little more complicated, but there is essentially the same DoF between 1.6 crop and 35mm if the 35mm camera is stopped down 1 1/3 stops, which you can verify on a DoF calculator.

    no idea, unforunately. I figured this out from the equations in Wikipedia articles and based on some knowledgeable posters on POTN.
     
  9. gnd macrumors 6502a

    gnd

    Joined:
    Jun 2, 2008
    Location:
    At my cat's house
    #9
    There is a great article on bokeh in the April 2010 issue of Carl Zeiss newsletter: "Bokeh - Nine rounded iris blades" (warning, PDF document in the link).
     
  10. Chris7 thread starter macrumors 6502

    Joined:
    Aug 8, 2008
    Location:
    Lost in Thought
    #10
    It seems there a two ways to get more out of focus blur at a particular FoV -- [edit] opening up (and decreasing DoF), and using longer focal lengths.

    In the above example, I started with f/4 on the full sized sensor (36mm) body, and I subtracted 1 1/3 stop from the 1.6x crop body to compensate for the fact that a crop body produces a deeper DoF than a full sized sensor body at the same aperture and at the same FoV. This gave me about f/2.5 (actually 2.55, but whatever).

    Then I subtracted another 1.5 stops to compensate for the lower focal length needed on a crop sensor body, based on the pattern you mentioned above (this number is more approximate). This gives me about f/1.4 or f/1.5 on a crop body to equal the same out of focus blur as the full sensor body at a f/4 at 100mm at 8.6 ft., in the above example.

    Where did I go wrong?
     
  11. toxic macrumors 68000

    Joined:
    Nov 9, 2008
    #11
    huh? doing both - stopping down AND using a longer FL - can yield more blur, but using the same lens and stopping down decreases the aperture, so there is less blue (and more DoF).

    I think you have your terms mixed up - "to stop down" is to use a smaller aperture, i.e. f/2.8 to f/4.

    the 1 1/3 stop difference in DoF between 35mm and 1.6 crop is a coincidence, not a mathematical rule. to calculate the maximum amount of blur, you still go by aperture. this puts the 1.6 crop camera at a 1.6x shorter FL and 1.5 stops wider.

    f/2.5 to f/4 is 1.5 stops.
    62.5mm / 2.5 = 25mm, and 100mm / 4 = 25mm.

    DoF would also be 1.5 stops smaller for 1.6 crop, but DoF varies with CoC, which is a function of enlargement factor (and therefore sensor size). what ends up happening is that the smaller enlargement for the 35mm camera makes the DoF larger when it's stopped down 1.5 stops. to make the DoF the same, you need to open it back up slightly (decreasing DoF), which puts you closer to 1.33 stops of DoF difference than 1.5.

    in the end, though, a 0.1 stop difference is negligible, so just think 1 1/3 stops wider to achieve the same DoF and blur on 1.6 crop as 35mm.
     
  12. mrkgoo macrumors 65816

    Joined:
    Aug 18, 2005
    #12
  13. toxic macrumors 68000

    Joined:
    Nov 9, 2008
    #13
    that's because the aperture is larger with the longer lens. using a wider lens at a much wider f-stop will produce the same degree of blur, though of course the background will look different.
     
  14. gnd macrumors 6502a

    gnd

    Joined:
    Jun 2, 2008
    Location:
    At my cat's house
    #14
  15. funkboy macrumors regular

    Joined:
    Apr 25, 2008
    Location:
    elsewhere
    #15
    If you've got the cash, just get it. The 85 1.8 is a fantastic lens. The AF is faster than the 50 1.4 and the 85 has more resolving power than the 50 when used wide-open or close to it.

    I have both with me when I'm shooting portraits (well, my 50 says Zeiss on it but it's still a 50mm f/1.4). The 85 gets used when I'm outdoors, in a larger room, or really want to fill the frame with the subject's face, and the 50 is for smaller rooms and "portrait Americain". I've been in situations where I used my Zeiss 35mm f/1.4 for portraits as well as the room was just too small for anything else.
     
  16. Chris7 thread starter macrumors 6502

    Joined:
    Aug 8, 2008
    Location:
    Lost in Thought
    #16
    Yes, I meant "open up."
    I think I follow you here. I'll paraphrase to make sure I understand... You have to open up 1 1/3 stops on a 1.6x crop sensor to get the same DoF as on a 35mm (36mm) sensor at the same FoV. Shooting f/2.5 on a 1.6x crop sensor will produce the same out of focus blur as f/4 on a 35mm sensor at the same FoV, because the aperture will be the same on the lens, even though the required focal length to attain the same FoV will be different (f/stop being a ratio of focal length divided by aperture), and it is the aperture, not the the focal length that is providing more out of focus blur at the same DoF.

    So there is not the additive effect I mentioned above (of shallower DoF plus more out of focus blur from having to use a longer focal length on a 35mm sensor).

    The only thing I'm left wondering is pretty minor. f/4 minus f/1.333 on this f/stop calculator produces almost exactly f/2.5 (f/2.52). So I'm not sure where the 0.1 stop difference comes from, even though it doesn't matter. Is this calculator accurate?

    Anyway, many thanks for your time in explaining this. And I admit am a little relieved to know there is still only about a 1 1/3 difference when going to a 35mm sensor, as I'll probably never have one.:)
     
  17. Chris7 thread starter macrumors 6502

    Joined:
    Aug 8, 2008
    Location:
    Lost in Thought
    #17
    Reason why full sized vs. crop sensors have shallower DoF at same f/stop and FoF?

    Edit: I meant "FoV," not "FoF" above.

    Toxic's comparison (quoted in part below) got me thinking... It the is primarily (only?) the greater aperture that results from the longer focal length required to attain the same FoV that makes the DoF on a full sized vs. crop sensor shallower at a given f/stop?

    It looks like, basically, the longer focal length required on the full sized sensor means that the same f/stop on a crop will mean a larger aperture on a full sized.

    When I look up the question on line, I keep getting answers answers that do not explicitly say this (maybe I haven't looked hard enough, though).

    Thanks.
     
  18. toxic macrumors 68000

    Joined:
    Nov 9, 2008
    #18
    almost. first, film ("FF") is 35mm - this is the height of the film + sprockets. subtract the sprockets and you have the 24mm recording surface.

    anyways, don't confuse amount of blur with DoF. amount of blur depends only on aperture. DoF depends on focal length, f-stop, circle of confusion, and subject distance. the circle of conclusion is determined based on enlargement factor (and therefore sensor size and final viewing size), viewing distance, and a standard for visual acuity.

    as I said before, the .1 stop difference comes from the difference in sensor size.

    1 1/3 is actually a noticeable difference. whether it's significant is, of course, subjective.

    as I said above, don't get DoF confused with blur.

    I'll try to break it down again for you:

    when using the equivalent FL on 1.6 and 35mm 1.5 stops apart, you get the same amount of blur.

    if DoF did not depend on sensor size, there would also be a 1.5 stop difference in DoF between 1.6 and 35mm. however, it does.

    the DoF is larger on 35mm stopped down 1.5 stops than on 1.6 because of the less enlargement required for the same viewing size.

    for example:
    10mm at f/2.8 at 5' on 1.6: 24.1' DoF
    16mm at f/4.8 at 5' on 35mm: 28.7' DoF

    f/4.8 is 1.5 stops narrower than f/2.8

    16mm at f/4.5 at 5' on 35mm: 21.4' DoF. f/4.5 is 1.333 stops narrower than f/2.8.

    21.4' is closer to 24.1' than 28.7', so we say 1 1/3 stop difference. the difference in amount of blur due to different aperture is negligible, so we ignore it.
     
  19. Chris7 thread starter macrumors 6502

    Joined:
    Aug 8, 2008
    Location:
    Lost in Thought
    #19
    Thanks

    EDIT: actually, looking back over the wik quotes (below) makes me think. Wik says, same aperture, same DoF. Except when using a crop sensor, the image must be magnified more when printing to the same print size, meaning smaller circle of confusion and smaller DoF. So you subtract 1 1/3 stops from from a crop sensor to get the same absolute aperture and at the same FoV, and thus the same DoF (if the two pictures were magnified the same amount in the final print with the full size sensor to a larger print). But if you print both to the same size, it would seem that you actually need to stop down (not open up) slightly from the minus 1 1/3 on the crop sensor to attain the same DoF at equal print size. E.g. you may need to open up only one stop rather than the commonly said 1 1/3 to 1.5 stops to compensate (I made up up the number 1 stop just to show the principle).

    No one else is saying this so I'm sure I'm wrong. What am I missing?

    I wrote this (below), then looked back over your last post, and it looks like you’re saying basically the same thing. The only thing I notice different is that it appears that 1 1/3 stops is enough to attain the same absolute aperture diameter at the different focal lengths required to get the same FOV on crop vs. full sized sensors.
    But think I see how compensating with 1.5 stops might be more accurate to compensate for the magnification factor (I just leaned what is meant by this, BTW).

    Anyway, I think I got most what I was missing.

    According to wik here,“Adjusting the f-number to the camera format is equivalent to maintaining the same absolute aperture diameter; when set to the same absolute aperture diameters, both formats have the same DOF."
    But, wik here explains, “Cropping an image and enlarging to the same size final image as an uncropped image taken under the same conditions is equivalent to using a smaller format; the cropped image requires greater enlargement and consequently has a smaller circle of confusion. The cropped image has less DOF than the uncropped image.”
    By enlargement factor, you’re saying that, when the image from the cropped sensor and full sized sensor is enlarged to the same dimensions, your eye will have less tolerance for what is slightly out of focus because the what is out of focus in the image is magnified more (the “circle of confusion” will appear [EDIT] smaller). This effectively narrows the DOF.

    I wonder if this DOF calculator takes this into account when it gives me these numbers?:

    10 divided by 2.8 = 3.57 (aperture), with a DOF at 5’ of 24.1
    16 divided by 4.44 (I added 1.333 stops here) = 3.6 (aperture), with a DOF at 5’ of 21.4 (rounded to f/4.5 for the calculator).

    If I add 1.5 stops and round to f/4.8, I get a DOF of 28.7 ft (full sized sensor, 16mm focal length, 5’ from focal plane).

    So maybe somewhere between these two numbers is where you get the 1.5 rather than 1 1/3 stops?
    The aperture is the same when stopping down 1 1/3 on a crop sensor, but perhaps the extra 0.15 stops takes into account the magnification factor? It seems like I should think open up 1.5 stops to get equivalent out of focus blur on a crop sensor, which I think is what you're saying.
     

Share This Page