Page 2 of 2 FirstFirst 12
Results 21 to 32 of 32

Thread: sensor size and DOF

  1. #21

    Default

    Ok, here's a very good link which explains why comparing DOF across formats is purely academic and serves little useful purpose for real world shootning. The conclusion has an excellent summary of all the factors involved.

    http://www.rags-int-inc.com/PhotoTechStuff/DoF/

  2. #22

    Default

    Am i right to say that DOF and FOV are being confused/interchanged here? esp. on the explaination for the 1.6x crop value

  3. #23
    Senior Member denniskee's Avatar
    Join Date
    Oct 2003
    Location
    bukit batok
    Posts
    5,468

    Default

    If I use a good old M42 screw mount manual focus lens on DLSR with crop factor, set it to hyperfocus, is the marking on the lens it still valid

    eg
    if i set aperture f4, turn focusing ring untill 1m marking on the of the f4 marking, 3m marking is on the other f4 marking, does it means i still have DOF of 2m?
    photography makes one sees things from all angles.

  4. #24

    Default

    denniskee -----
    it should-more or less-depending on the COC value used for calculation, as the digital body may be able to squeeze a bit more from the same lens and thus smaller DOF, otherwise the hyperfocal settings stand valid.
    ----------------------------------------------------------------

    When saying "other parameters being equal", we should really mean ALL. We are too often assuming certain conditions can be equal, without even trying them in the equation as variables. Some of the parameters simply contradict with others and should be included in the conditions explicitly.

    I'm thinking maybe the "diameter of bokeh from dot light source at infinity" serves better purpose in the context of portrait photography (and it's what the initial poster was asking I believe). Pure DOF calculation only tells you within what range of distance is your subject "reasonably sharp", but not how the bokeh turns out to be and thus how the final images may compare.

    From a practical point of view, it makes more sense to compare the bokeh size between different sensors using the same: ---
    FOV (a.k.a equivalent focal length);
    Subject distance;
    F value;
    Print size.
    and no cropping is done during post-processing
    ----makes more sense because it's as if we were shooting using the same perspective, same framing, same aperture value and different formats. With these conditions set, I guess I can say that larger formats give bigger bokeh circles(or hexagons, pentagons..whatever).

    edited: Noticed that COC is kicked out of the scene? That's why DOF isn't really of our interest.
    Last edited by grantyale; 22nd April 2005 at 07:13 PM.

  5. #25

    Default

    Wow, that was a whole lot of technical details... Let me try to put it in the real world context and in layman terms....

    After reading all these, I think what we're trying to prove here is that given a specific scenerio, a person holding a Canon A95 (Point and shoot) and a person holding a Nikon D70 (DSLR) standing at the

    same distance from the subject
    using the same shutter speed
    same aperture
    same focal length
    focusing on the same object

    The result will be that the D70 will have a shallow depth of field as compared to the A95 because the D70 has a larger sensor size.
    Pictures always should have a special story behind it or a special meaning to the photographer.

  6. #26

    Default

    Hitman --
    "Same focal length" as in FOV, sure you know that.
    And yes, shallow depth of field is technically correct but not directly of our interest - DOF and Bokeh size are very much related but are different after all, and it's the later that's usually of our interest, especially when it comes to portrait shots.
    --------

  7. #27
    Member
    Join Date
    May 2003
    Location
    Singapore
    Posts
    334

    Default

    Quote Originally Posted by roygoh
    I think there is some misconception here.
    Sorry, but can you tell me what was my misconception. HydroPoP's question was "Is there a relationship between sensor size and the depth of field given all other parameters constant?" If all other parameters are constant, why are we introducing final image size and viewing distance into the equation?

    For example, to get the 28mm equivalent field of view on a full frame sensor would require a 7.1mm focal length on a 2/3" sensor. If all other parameters are identical, are you saying the shorter focal length alone does not account for the greater depth of field with the smaller sensor?

  8. #28

    Default

    Try reading #24 and #26. I believe DOF is not the concern here, but rather bokeh size.
    If you are still looking at DOF though, watch out for the conditions you've probably overlooked.

  9. #29
    Senior Member
    Join Date
    Jan 2002
    Location
    Northwest
    Posts
    5,011

    Default

    Quote Originally Posted by Zplus
    Let me try again....

    Given the same lens, set at a focal length, aperture and focus distance... the different sensor sizes will result in different crops. The DOF will be the same... but on a smaller sensor... you'd get less of the blurred background becoz its been cropped away... It would APPEAR to have more DOF because more of the image is in focus...

    So while the DOF does not change in each case... the composition is affected by the sensor size.

    Thought the diagram shows that... the DOF is the same in both the full frame and 1.6x crop....
    I think Zerstorer has given much better explanation than I have.

    Basically DOF only makes sense on the final image output, because DOF is about what appears as sharp or not sharp to the person viewing the image. It nakes no sense to compare the COC or DOF on different image sensors. So all your statements that goes like "the DOF is the same on both sensors" is simply not applicable.
    As complexity rises, precise statements lose meaning and meaningful statements lose precision.

  10. #30
    Senior Member
    Join Date
    Jan 2002
    Location
    Northwest
    Posts
    5,011

    Default

    Quote Originally Posted by linse
    Sorry, but can you tell me what was my misconception. HydroPoP's question was "Is there a relationship between sensor size and the depth of field given all other parameters constant?" If all other parameters are constant, why are we introducing final image size and viewing distance into the equation?

    For example, to get the 28mm equivalent field of view on a full frame sensor would require a 7.1mm focal length on a 2/3" sensor. If all other parameters are identical, are you saying the shorter focal length alone does not account for the greater depth of field with the smaller sensor?
    OK, in the first place, "given all other parameters constant" is not possible even if it seems so.

    All other parameters simply cannot remain constant when you change sensor size, in the context of comparing DOF.

    1. When you change sensor size but use the same focal length, then the field of view changes. To maintain same size of the subject on the sensor then the lens to subject distance has to change.

    2. If you want to maintain same field of view (and thus keep subject distance unchanged for the same image size of the subject ont he sensor) then the focal length has to change.

    So you see, you simply cannot state simply that "all other parameters remain constant".

    In your second paragraph you assumed the second case (constant field of view). I am not saying that is wrong, I am jusy saying that it was not clearly stated in the original question. In fact I totally agree that it makes more sense to assume a constant field of view and thus a constant camera-to-subject distance (=perspective) when we want to compare DOF between images taken from different sized sensors. Bear in mind that the focal length is thus changed between the 2 scenarios, so not "ALL" parameter remained the same.

    So assuming the following are kept unchanged:
    1. FOV (note: focal length has to change to maintain same FOV on different sized sensors)
    2. Subject distance
    3. f number (note: the physical diameter of the aperture has to change to maintain the same f number)
    4. final output image size (note: the magnification from sensor to final image has to change to produce same output size from sifferent sized sensors)

    The COC has to be determined on the final image and then translated to the COC on the sensor, based on the magnification. That parameter (different for both sensors) is then thrown in to the DOF formula with the other parameters to calculate the DOF.

    The result will be that the image taken from the smaller sensor has a larger DOF.

    So yes, I am saying that the smaller size of the sensor alone does not account for the larger DOF. You have to be very specific about which of the other parameter are kept constant between the 2 scenarios.

    Quote Originally Posted by linse
    Don't look at the 35mm equivalent focal length bullshit. Same true focal length, aperture and subject distance means same depth of field. However, the smaller sensor will be more cropped. If you alter the subject distance to get equivalent field of view, then the smaller sensor will give you higher depth of field.
    35mm equivalent focal length is not entirely BS when understood correctly. When you assumed that FOV is kept constant then the 35mm equivalent focal length is also kept constant. The only problem is that the DOF calculation cannot rely on the 35mm equivalent focal length.

    Second sentence is where I feel your "misconception" lies. DOF only apllies when referring to a final output image that is viewable. While the amount of absolute "blurness" on the sensor remains the same when the focal length, subject distance, focus distance setting and aperture are kept the same, the amount of magnification required to deliver the same sized viewable output image is different. As such the DOF of the image from the smaller sensor actually have lesser DOF because the "blurness" has to be amplified more.

    So if FOV is kept constant, then a smaller sensor yields larger DOF. If focal length is kept constant, then a smaller sensor yields smaller DOF. (Note: assuming output image size, f number, subject distance and focus distance setting remains unchanged)
    Last edited by roygoh; 23rd April 2005 at 02:24 AM.
    As complexity rises, precise statements lose meaning and meaningful statements lose precision.

  11. #31
    Senior Member
    Join Date
    Mar 2002
    Location
    Some equitorial, tropical isle
    Posts
    2,398

    Default

    Quote Originally Posted by roygoh
    I think Zerstorer has given much better explanation than I have.

    Basically DOF only makes sense on the final image output, because DOF is about what appears as sharp or not sharp to the person viewing the image. It nakes no sense to compare the COC or DOF on different image sensors. So all your statements that goes like "the DOF is the same on both sensors" is simply not applicable.
    I made that assumption because the original post stated:
    Is there a relationship between sensor size and the depth of field given all other parameters constant?

    Which I took it to mean same fstop and focusing distance for the same lens... Basically, if you took a lens designed for a 35mm DSLR and focused it at a particular fstop and focusing distance... the image rendered on the film plan will be the same regardless of recording medium. Of course that lens is designed for a 35mm film frame and a smaller recording medium sensor will not show you the whole picture... the "crop factor" effect.

    Since the image is the same, I concluded that the DOF is the same in each case, just that the smaller sensor, eleminated much of the peripheral...thus making the resulting image appear to be sharper throughout the entire image.


    Of course pixel size and coc does come into play in the real world. And as I have read above, you have maintained FOV as the constant parameter instead of focusing distance as well. In which case, if we factor in these variables as well, I would totally agree with you and zestorer's comment.

  12. #32
    Senior Member
    Join Date
    Jan 2002
    Location
    Northwest
    Posts
    5,011

    Default

    Quote Originally Posted by vince123123
    so you're agreeing with me that MF has shallower DOF than 35mm? Cos that's what I seem to be getting and it is now beckoning me to enter the world of expensive MF backs LOL (like real)
    If you keep the angle of view, subject distance and final print size constant, then yes, DOF decreases when the sensor size increases.
    As complexity rises, precise statements lose meaning and meaningful statements lose precision.

Page 2 of 2 FirstFirst 12

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •