Should 4/3 lenses be half the size of full frame lenses?

Discussion in 'Digital Cameras' started by bob, Jun 13, 2011.

  1. bob

    Guest Guest

    then it's not equivalent, is it?
    they do for you though.
    Guest, Jun 14, 2011
    1. Advertisements

  2. bob

    dj_nme Guest

    You obviously didn't read what I wrote about "equivalent f-stop".
    Are you being deliberately obtuse?
    No, I don't.
    It is all about the same Dof and NOT about the same exposure settings.
    Read what I wrote, not what you want to see.
    This is relevant, how?
    For the same DoF, not exposure value.
    No, that would give an over-exposed image.
    Wider aperture requires a shorter exposure time or lower ISO setting for
    the same exposure value.

    Are you sure that you even know which end of a camera to hold?
    I am beginning to doubt this, based on what you've written.
    No, it does not.
    I said nothing about raising ISO settings.
    you have created that from your own imagination.
    dj_nme, Jun 15, 2011
    1. Advertisements

  3. That is *exactly what you get* with f/2.8 or any other given f-number.

    Everything else you're saying here is either wrong or irrelevant.
    Neil Harrington, Jun 15, 2011
  4. bob

    dj_nme Guest

    In what version of reality?
    Either you want the same DoF (then aperture setting AND shutter speed
    must change) or you want the same aperture AND shutter speed setting
    (then DoF will change).
    You can't have it both ways.
    That is a surprise.
    The facts demonstrably contradict what you are trying to portray as
    How does that make me "wrong"?
    dj_nme, Jun 15, 2011
  5. bob

    Guest Guest

    to have the same dof, you need 2 stops wider on 4/3rds, which is what i
    it explains it.
    which means to be equivalent, it must be 2 stops wider.
    Guest, Jun 15, 2011
  6. bob

    Guest Guest

    raise the iso on the larger sensor so you can use a smaller f/stop,
    which will match noise and dof and keep the exposure the same.
    Guest, Jun 15, 2011
  7. Signal/noise ratio is not what we've been talking about, and irrelevant
    We're not talking about "image quality" either.
    We're talking only about whether f/2.8 is or is not the same thing in terms
    of exposure. The answer is Yes, it is. f/2.8 is f/2.8, under any and all
    circumstances. It is *never* "equivalent" to f/1.4 or any other f-number.
    On 1/4 the area. That's the same exposure.
    Changing the sensor area does not change the exposure. If you take a 50mm
    f/1.8 lens for a full frame camera and put it on a DX camera, it is still
    f/1.8. It will give the same exposure wide open as a 35mm f/1.8 DX lens on
    that DX camera, or any other f/1.8 lens.
    Neil Harrington, Jun 15, 2011
  8. bob

    Guest Guest

    it's very relevant. two photos are not equivalent if one is noisier
    than the other.
    yes we are. two photos are not equivalent if the image quality is
    except that exposure is not the only factor that goes into a photo.
    but with 4x the noise and greater depth of field. in other words,
    they're not equivalent.
    i didn't say it changed exposure. however, changing the sensor size
    changes other things, which must be compensated for if you want to call
    something equivalent.
    the dx camera will have more noise (assuming same sensor technology)
    and greater depth of field. that's not equivalent.

    by changing the f/stop and adjusting the iso to match, you have the
    same exposure *and* the same depth of field *and* the same noise.
    that's equivalent.

    it's simple physics, really.
    Guest, Jun 15, 2011
  9. bob

    dj_nme Guest

    Nospam ("bob") doesn't seem to understand this basic aspect of photography.
    I don't know why.

    Neil: What did you think about my explanation of "equivalent f-stop",
    does it make sense to you?
    I was attempting to make it as simple as possible.
    dj_nme, Jun 15, 2011
  10. bob

    dj_nme Guest

    You made no issue of ISO setting in your previous posts, so I was merely
    keeping the same variables as you.

    The problem with your idea is that to get the same DoF and shutter speed
    on a smaller format, the ISO must be lowered.
    This is not always possible, EG: how do you lower the ISO if the setting
    is already at the lowest when using the same aperture and shutter
    setting as the FF camera?
    Big problem, isn't it?
    What you do, use a 2-stop Neutral Density (ND) filter to match the
    exposure value of the FF camera?
    That would be your only option (using a ND filter).
    If you're already using the base ISO on the smaller format at the same
    aperture & shutter settings as the FF, then adjusting the aperture to
    match DoF the ISO can't be adjusted down and so out comes the ND filter
    to compensate for the same exposure value.
    dj_nme, Jun 15, 2011
  11. bob

    dj_nme Guest

    No, it does not.
    You don't seem to be wanting about the same thing that you're talking about.

    Either you are talking about matching the image taken with a FF camera,
    or you are wildly changing what you are talking about every single time.

    Either you want the same DoF or the same exposure settings.
    Neither is possible at the same time.
    So, how do you set the ISO lower than the lowest setting on the camera
    to maintain the same exposure value?
    It can't be done.
    Likewise, when an FF lens is set to f/1.4 there is no way that any 4/3
    lens aperture could be opened wide enough to match DoF.
    Where are the f/0.8 4/3 lenses?
    dj_nme, Jun 15, 2011
  12. bob

    RichA Guest

    RichA, Jun 15, 2011
  13. bob

    Guest Guest

    nope. you raise the iso of the larger format.
    Guest, Jun 15, 2011
  14. bob

    Guest Guest

    raise the iso of the larger format.
    exactly why 4/3rds is limiting.
    Guest, Jun 15, 2011
  15. []
    It contradicts what you would expect, yes. The total number of photons
    gathered is less, so if the number of pixels if the same, there are fewer
    photons per pixel, and hence noisier images.
    As the aspect ratio is different, "half-frame" is a nominal term here.

    Full-frame: 36 x 24mm
    Four-thirds: 17.8 x 13mm (from:

    so "half" is near enough (it's 0.26 of the area, 0.51 of the mean linear
    dimension, 21.6mm diagonal).

    David J Taylor, Jun 15, 2011
  16. bob

    J. Clarke Guest

    You're missing something here. What matters is not the number of
    photons passing through the lens, but the number that both pass through
    the lens and strike the sensor. The ones that pass through the image
    circle but do not strike the sensor do not count.

    So the crucial figure is not the total number of photons passing through
    the lens but the photons per unit area in the image plane.

    And for a given aperture measured in f numbers, and leaving aside issues
    such as clarity of glass and quality of optical coatings, that number
    will not vary from one photographic lens to another regardless of focal

    People who hear astronomers talk about "light gathering" often try to
    transfer this principle to photographic lenses. This effort is futile
    because astronomers deal with a different situation.

    A photographic lens is used with the film or sensor at the prime focus.
    A longer lens will produce greater magnification.

    An astronomical telescope generally uses auxiliary lenses to achieve a
    desired magnification. The 15,000 mm Hale telescope, the 57,000mm
    Hubble, and my 1250mm Celestron might all be used to produce an image of
    Jupiter at the same magnification. And when you're doing _that_ then a
    larger diameter objective will result in more photons per unit area on
    the sensor.
    J. Clarke, Jun 15, 2011
  17. []
    Photons per unit area is what affects the exposure required, but the total
    photons in the image is what affects total image quality.
    But if the sensor is half the linear dimensions, only a quarter of the
    photons are captured compared to the full-size sensor. Hence the poorer
    high-ISO performance of small-sensor cameras.
    Imaging Jupiter is just the same - photons per unit area will depend on
    the T/number of the optical system (remembering that telescopes have
    mirrors causing partial obstructions). Given the same T/number, the
    longer focal lengths will produce a larger image, and therefore more
    photons in total.

    David J Taylor, Jun 15, 2011
  18. bob

    dj_nme Guest

    Yes it is.

    I am asking you how a smaller format could lower it's ISO when you have
    opened up the aperture to match the exposure value of a FF camera set at
    the same ISO.
    If you don't know how, then just say so.

    I know how to get around the problem (assuming that the smaller format
    lens can open wide enough to match DoF): use ND filters to cut the
    amount of light and prevent over-exposure (keeping the same exposure
    value as on the FF camera).
    Without the ND filters, you smaller format can't do that and you are SOL
    (so out of luck).
    dj_nme, Jun 15, 2011
  19. bob

    dj_nme Guest

    That isn't what I was asking you.
    I am asking how you would make the 4/3 camera match the FF camera.
    Not the other way around.
    So this is all just sort of long, drawn-out 4/3 camera bash?
    It would have been a whole more concise and not be wasting any-one
    else's time.
    You first post in in this thread should have been "4/3 cameras are
    shit!" and been done with it.
    dj_nme, Jun 15, 2011
  20. []
    It's a discussion of what is the "equivalent" lens for 4/3 compared to
    full-frame, the question I asked in the first place. If you want
    equivalent exposure times for a given light level, then look for the same
    f/number. If you want equivalence in other areas such as the same
    depth-of-focus or the same image quality for a given light level, you need
    double the aperture, half the f/number.

    Statements of physics as applied to different sensor sizes.

    David J Taylor, Jun 15, 2011
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.