Olympus OM enthusiasts' digital prayers have been answered ...

Discussion in 'Digital Photography' started by Bruce, Feb 4, 2012.

  1. Bruce

    Trevor Guest

    So show me where you previously mentioned the use of Kodak sensors which my
    Canon doesn't have, and possibly explains why I don't have the problems you
    are sure exist with the Olympus cameras.

    Trevor.
     
    Trevor, Feb 9, 2012
    #41
    1. Advertisements

  2. Bruce

    Bruce Guest


    This is like trying to have a conversation with a wasp.
     
    Bruce, Feb 9, 2012
    #42
    1. Advertisements

  3. Bruce

    RichA Guest

    I think there is more to this than just telecentricity. I tested an
    85mm f2.0 OM lens on a m4/3rds camera and it vignetted, right down to
    f5.6.
    If an 18mm OM doesn't vignette horribly on a FF, then something else
    is at work.
     
    RichA, Feb 9, 2012
    #43
  4. Bruce

    Bruce Guest


    I find this discussion highly entertaining because it is the blind
    leading the blind.

    The suggestion that the least telecentric lenses are always those with
    tiny rear elements is completely risible. In some cases, they are
    among the *most* telecentric!

    The implied converse, that a larger rear element means the lens is
    closer to telecentric is *complete nonsense*. It is extremely naive
    to assume that because the size of the rear element is closer to the
    size of the sensor, that the rays must be emerging almost parallel to
    each other and therefore perpendicular to the sensor. Once again, the
    truth can be the exact opposite, because a larger rear element can
    allow light rays to emerge at *more* oblique angles!

    That might appear counter-intuitive to some, but that merely
    demonstrates that intuition is highly subjective, and often wrong.
     
    Bruce, Feb 9, 2012
    #44
  5. Nobody said that is *always* the case.
    I gave an example, which is certainly the most common situation, where
    it *IS* the case.

    That is a completely ridiculous statement - we can all restrict our
    comparisons to the "some cases" which *can* be worse. In most cases
    that is NOT the case. I referred to the entire OM range.

    There are many aspects of the OM 18mm which demonstrate it's poor
    telecentricity. However poor rear telecentricity is guaranteed
    *because* it has a small rear element. That, together with the proximity
    of the rear element to the focal plane (it projects into the lens mount
    and just clears the mirror!), restricts the angle of incidence of the
    principle rays at the corners of the focal plane to very oblique angles:
    the very problem that Olympus claim makes such a non-telecentric lens
    unsuitable for digital. Yet it works extremely well and is one of the
    most sought after OM Zuikos for FF dSLRs!
    That is *not* the implied converse. Only an idiot thinks that "all cats
    are furry animals" implies "all furry animals are cats"!

    The converse is that telecentricity, more specifically
    rear-telecentricity which is the version under discussion here,
    *requires* a large rear element!

    Should Bruce Almighty dispute that, I am sure he can point us to a ray
    diagram which shows otherwise. (That doesn't mean a diagram showing that
    telecentricity *can* be worse on a large rear element design, that is
    obvious with the standard textbook design of a front-telecentric lens
    being such an example.)
     
    Kennedy McEwen, Feb 9, 2012
    #45
  6. Bruce

    Pete A Guest

    Indeed, the principle ray contains no light whatsoever because it's
    infinitely thin. Yes, the aperture affects the cone of peripheral rays.

    A pinhole camera exhibits cos^4 light falloff at the film. Having a
    very small aperture results in a correspondingly small cone therefore
    it is primarily the angle of the principle ray that causes the light
    falloff in such a camera. At 33 degrees the falloff is 1 f-stop.

    With an image-side (rear) telecentric lens, it is only the angle of the
    cone that causes light falloff. The cone does not vary across the image
    plane therefore the image doesn't suffer from corner vignetting. If the
    lens is badly designed then mechanical vignetting may cause light
    falloff in the image corners.

    Very wide-angle lenses suffer from light falloff on the object side,
    but this is a separate issue.

    In any mirror-less system, the lens designer is free to position the
    exit pupil very close to the image plane, which places an unrealistic
    demand on the design of image sensors; especially Bayer CFA sensors.
    Specifying "near-telecentric" simply means placing the exit pupil
    _reasonably_ far away from the sensor. So, it is a big issue and
    perfectly warranted in a mirror-less system specification.

    Literally, you are correct. In essence, stopping down reduces the rays
    that are problematic to the sensor.

    The tiny exit pupil you mentioned is not the problem. The distance of
    the exit pupil from the image plane is the problem.

    There is no paradox, you misstated cause and effect. That isn't a
    criticism - I frequently write things that make sense only to me.

    That sounds like a very interesting lens.
     
    Pete A, Feb 9, 2012
    #46
  7. Bruce

    Mike Guest

    Why would a user of the 24x36mm OM-1/2/3/4 be anxious for a 1/4 frame
    (17.3 x 13 mm) look-a-like?
     
    Mike, Feb 9, 2012
    #47
  8.  
    Kennedy McEwen, Feb 9, 2012
    #48
  9. Bruce

    Rol_Lei Nut Guest

    Not anxious, but there are things to like: "Free film" (well, once
    you've more or less heavily invested in equipment and any necessary
    peripherals), "instant developing" (though if you shoot raw that also
    takes some time & most images can use some adjusting before printing or
    viewing), some quite decent lenses (some of the very few WA zooms I've
    found to be good - I normally use Zeiss & Leica with film) and very
    compact and light (even compared to the original OM series).

    What I like less it that the "form follows function" idea got lost in
    translation and the useless false pentraprism adds quite a bit to the
    effective size of the camera.
    Also the hand grips seem to be a lame attempt to label the camera as a
    "pro" model (marketing predominating over content).
    Perhaps Olympus should introduce some oversized white telephoto lenses
    which can be parked at the side of sport fields, then they'd have
    succeeded in emulating the ultimate marketing brand....

    That said, if the image quality is as good as it should be and the
    camera seems tough enough, I'll probably end up buying one when they get
    affordable. A water resistant, tough (hopefully) camera with decent (or
    hopefully more than decent) quality and tiny but good lenses is
    something to like.

    But I'll still use my film Leicas and Rolleiflexes when I want to do
    something special...
     
    Rol_Lei Nut, Feb 9, 2012
    #49
  10. Bruce

    Pete A Guest

     
    Pete A, Feb 9, 2012
    #50
  11. Bruce

    Pete A Guest

    Thanks Eric.

    Some may find these interesting:

    <http://www.pco.de/fileadmin/user_upload/db/download/pco_cooKe_kb_shading_0603_s.pdf>


    <http://en.wikipedia.org/wiki/Vignetting>


    I totally agree with Bruce. A true bi-telecentric (afocal) lens demands
    a front element somewhat larger than the object and a rear element
    somewhat larger than the sensor; a near-telecentric image-space lens
    has no such demands. Very few people can reverse-engineer a lens just
    by looking at it and jumping to intuitive conclusions :)
     
    Pete A, Feb 11, 2012
    #51
    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.