It shouldn't be 4/3 but "6x6"

Discussion in 'Digital Cameras' started by Guenter Fieblinger, Feb 15, 2004.

  1. Surely it's good that the industry is going to standardize the mount for
    interchangeable lenses!

    However, It is unfortunate that the ratio 4:3 is chosen.
    Instead of choosing the ratio of the now widely used monitor format it
    would be more convincing to go for the quadratic format that has been
    used in photography for a long time.

    BENEFITS
    The rationale behind the plea for a square format is that the lens is
    the most limiting factor in camera design today.
    If well designed the lens is very costly, and lenses are heavy. The
    priority should be to use this resource as well as possible.

    The added benefit would be that for taking the shot one wouldn't need to
    design a camera for the awkward vertical framing.
    If the photographer wants to compose 4/3 or 3/2 formats he could do this
    easily in a camera designed for rectangular sensor.
    For the most thrifty person a simple gadget could be implemented to
    restrict the field of view as well as restrict the utilisation of the
    sensor in order to save on memory.

    DISADVANTAGES if at all:
    The disadvantage of a quadratic format would be a slightly increased
    sensor area that would cost fractionally more than a rectangular one.
    But sensor area and pixel count would be larger for those who like to
    use the added area.
    In the future sensor cost will decrease even mor in comparision to the
    cost of 'glass'. Wasting sensor area for those cropping to rectangular
    formats would not amount to be significant at all.

    QUADRATIC frame?
    Would you like a square format at all?
    I bet you would! There is a long tradition of 6x6 photography that has
    proven the concept.
    True, the field of vision of the human perception is wide, even wider
    than 3:2. It makes sense to compose many shots for this format. But why
    limit the camera to this format?

    HISTORY
    Looking back in history the main reason for 3:2 was the cost of film.
    It would have been wastage not to utilise film to the maximum, eg.
    'cropping' a rectangular film format for the much utilised composing of
    a shot in the rectangular format.
    The populariziation of photography was actually furthered by the film
    saving LEICA format that Oskar Barnack developed 1913.

    The second historical reason for the growing following to the
    rectangular format 3:2 can be seen in the spread of the SLR. The design
    of the rapidly moving mirror is a major problem in SLR engineering. A
    square format mirror would per area be much more difficult to dampen
    than a rectangular one.

    RATIONALE
    Perhaps some major company, Canon springs to mind, could dare to go the
    rational way and implement a square format.
    For the photographer it would be much more satisfying to utilize a good
    lens fully than to waste the effort gone into the construction of the
    lens and into the effort of carrying unused glass along all the time.

    Of course, the square format will be more attractive for the non SLR
    digicam because there no moving mirror that seems to be a problem for
    the industry still. Lets go the whole way - the square way.

    Guenter
     
    Guenter Fieblinger, Feb 15, 2004
    #1
    1. Advertisements

  2. Guenter Fieblinger

    Don Stauffer Guest

    First of all, I hadn't heard of any new standard mount (other than
    T-mount), though the old Pentax screw mount was pretty close for awhile.

    Secondly, what has the mounting interface, which is a mechanical
    interface, got to do with the format aspect ratio?
     
    Don Stauffer, Feb 15, 2004
    #2
    1. Advertisements

  3. The 4/3 standard does not stnadardize the 4:3 aspect ratio.
    It only standardizes the maximum sensor diameter, i.e. diagonal.


    /Roland
     
    Roland Karlsson, Feb 15, 2004
    #3
  4. Guenter Fieblinger

    Tony Spadaro Guest

    The disadvantage would be the need for square prints.
     
    Tony Spadaro, Feb 15, 2004
    #4
  5. Guenter Fieblinger

    per Guest

    Do you talk about 6x6 inch or 6x6 mm?
    /per
     
    per, Feb 15, 2004
    #5
  6. I've owned two 6x6 cameras. I always viewed them as wasting a lot of
    film. A 6x6 and a 6x4.5 are equivalent in photographic resolution for
    my uses; and the 6x4.5 gets more shots per roll, is smaller, lighter,
    and the lenses may average a tad cheaper. It's a necessity with the
    TLR format, where turning sideways to shoot just doesn't work. But
    for SLRs and for digitals, this isn't much of a problem.

    I don't take square pictures, so a square-format camera is a bad thing
    for me.
     
    David Dyer-Bennet, Feb 15, 2004
    #6
  7. Thank you all for clearing up a lot of confusion that had boggled my mind.
    My underlying interest was unnecessarily referring to the efforts of the 4/3
    initiative that I misinterpreted. I thought that it did actually refer to
    rectangular sensor size in some of its specifications.
    I will try now to clarify my original issue that still intrigues me: why do
    we have rectangular sensors instead of quadratic ones in digital cameras?

    What I am pleading for is to optimise the usage of a scarce resource and, in
    fact, rather waste a cheap resource instead.

    Scarce resources in the epoch of digital cameras are the
    - carrying capacity (weight of equipment)
    - cost of lenses
    - handling of camera (ergonomics) when shooting, due to manipulating buttons
    and knobs to accomodate the functionality of the camera.

    Cheap resources are nowadays
    - the size and pixel count of the sensor
    - the capacity of memory
    - the effort for cropping a shot.

    Optimisation of a camera design should therefore go for a quadratic size of
    sensor (its dimensions being a matter of technical advances).
    This design for quadratic format will waste in fact for most of the time
    'film', that is sensor area and memory capacity of the memory chip, because
    for many takes the user will crop to a rectangular image. 'Cropping' by
    either presetting a mask when shooting thus saving memory space, or by
    postprocessing doing the framing in leisure or when viewing thus keeping the
    original overkill of information outside his intentions of composition.

    The benefit of the quadratric format of the sensor would be
    - optimal usage of the lens
    - better ergonomics in camera handling as the awkward vertical position
    would not be neccessary
    - availability of peripheral information outside the format of interest to
    be decided upon when postprocessing or viewing.

    What used to be 'wasting a lot of film' as David rightly pointed out in his
    post wouldn't anymore.

    I am hoping for the appearance of quadratic format sensors in the near
    future.

    Guenter
     
    Guenter Fieblinger, Feb 16, 2004
    #7
  8. Guenter Fieblinger

    Don Stauffer Guest

    Ah, you mean as a 4/3 CCD diagonal? Okay, missed that. Boy, that is a
    long way from 6cm x 6 cm!
     
    Don Stauffer, Feb 16, 2004
    #8
  9. Guenter Fieblinger

    John Navas Guest

    [POSTED TO rec.photo.digital - REPLY ON USENET PLEASE]

    In <> on Mon, 16 Feb 2004 09:36:13 +0100,
    Only at the very high end of the range.
    High-resolution sensors are still relatively expensive, and cost goes up
    rapidly as the area increases (due to less chips per wafer multiplied by lower
    yields).
    Memory is also still relatively costly.
    It's a tradeoff with convenience -- many people prefer in-camera cropping.
    With all due respect, I don't think that follows.
    Drawbacks include:
    - substantially more expensive sensors
    - significant increase in memory cost
    - reduced convenience due to increased need for masking and/or post processing
    Instead it would be wasting a lot of silicon.
    I wouldn't hold my breath. :)

    For that matter, why not a round sensor? ;-)
     
    John Navas, Feb 16, 2004
    #9
  10. Guenter Fieblinger

    Drifter Guest

    First of all, I hadn't heard of any new standard mount (other than
    http://www.four-thirds.org/en/index.html
    The PDF on that page makes for some interesting reading.

    Drifter
     
    Drifter, Feb 16, 2004
    #10
  11. Actually, square works better than rectangles in books and magazines: it
    sits on the page with space for titles and commentary below it. Horizontal
    rectangles are either tiny or cross two pages trashing the center of the
    image, and vertical rectangles leave no room for titles/commentary.
    Square is a bit harder to compose to than rectangles, but it does have its
    own beauty.

    Square sensors have the problem, though, that for wide angle, they lose
    width: a 56x56mm square fits inside an 80mm circle, but so does a 40x69mm
    rectangle. For wide angle work, there's a big difference between a 56mm
    width and a 69mm width.

    David J. Littleboy
    Tokyo, Japan
     
    David J. Littleboy, Feb 16, 2004
    #11
  12. Guenter Fieblinger

    Alan Browne Guest


    The industy has missed a golden opportunity to go Golden Section
    (SQRT(5)+1/2):1

    or even better: ISO-216 where the ratios are SQRT(2):1 ...eg: the size
    used in most of the world for standard paper stocks.

    Cheers,
    Alan
     
    Alan Browne, Feb 16, 2004
    #12
  13. Guenter Fieblinger

    Alan Browne Guest

    Guenter Fieblinger wrote:

    The scarce resource that needs preserving is trees. Therefore a common
    paper format aspect ratio (ISO-216 / DIN 416) would have been ideal.
    "A" sizes are all SQRT(2):1 in aspect ratio. Printing systems, editing
    systems, paper stocks, boxes to store the stock, systems to maange the
    stocks are available in abundance... any change in output size is always
    at the same ratio.

    It is also reasonably close to 1:1 and hence the lens is well employed.

    Cheers,
    Alan

    PS:If you want to optimize lens use, photos would need to be round.
     
    Alan Browne, Feb 16, 2004
    #13
  14. But the "Thanks to this lens mount design, most light can strike the image
    sensor from nearly straight ahead..." bit is a complete lie. Lenses don't
    work like that: they take a cone of light from a point in the scene and
    focus it back to a point: the light emerging from the back of the lens is a
    cone, in all photographic lenses. Wide angle lenses on SLRs are already
    fairly radical retrofocus designs, so unlike the wide angles for rangefinder
    cameras, they are positioned well away from the sensor plane, and the angles
    involved aren't a problem at all. They've solved a non-problem.

    Also, the smaller sensor means higher noise, so for the same image quality,
    you have to shoot at a wider aperture. Thus you need lenses a full stop
    wider on the E-1 than on the 10D, completely negating the advantage of the
    larger crop factor.

    So far the _only_ advantage of the E-1 is that the zoom lenses are designed
    for more sensible focal length ranges. But with the various 18-<whatever>
    lenses appearing now, that's no longer an issue.

    David J. Littleboy
    Tokyo, Japan
     
    David J. Littleboy, Feb 16, 2004
    #14
  15. Guenter Fieblinger

    Alan Browne Guest

    I hope when you looked at the ISO216 you noticed that while paper comes
    in many sizes, they are all consistently at a ratio of SQRT(2):1.

    It is the smart way to size paper for enlarging or reducing... and
    unfortunately the camera industry did not pick up on it for digital...
     
    Alan Browne, Feb 16, 2004
    #15
  16. Thank you all for your response.

    Your points are well taken, so time will tell, eventually.
    At present I think I would be the only buyer for such a device if I scan the
    response. So I'd rather not hold my breath for this to happen.

    The round sensor passed my mind, but there I thought in your (John) lines: waste
    of silicon, it's not that cheap, because in production the packaging would be
    drastically reduced. Getting the sensor round would be a major effort anyway.
    (Yes, you were pulling my leg for sure).

    Guenter
     
    Guenter Fieblinger, Feb 16, 2004
    #16
  17. It's not 4/3 inch either. The sensor has about the same useful diagonal
    as a TV pickup tube with a 4/3 inch outside diameter.

    A "one inch" sensor has a diagonal of about 16 mm, so a 4/3 sensor ought
    to be about 21 mm diagonal. That's just about half the size of a 35 mm
    camera frame with 43 mm diagonal.

    Dave
     
    Dave Martindale, Feb 16, 2004
    #17
  18. Guenter Fieblinger

    John Navas Guest

    [POSTED TO rec.photo.digital - REPLY ON USENET PLEASE]

    In <c0qs62$3d2$> on Mon, 16 Feb 2004 16:46:58 +0000 (UTC),
    4/3 has a diagonal of 22.5 mm, or slightly more than half the 43.3 mm diagonal
    of 35 mm.
     
    John Navas, Feb 16, 2004
    #18
  19. Guenter Fieblinger

    Mark Roberts Guest

    The reason that digital cameras have rectangular sensors is to optimize
    usage of an expensive resource: the senor itself, still by far the most
    expensive part of a digital camera. Since the overwhelming majority of
    printed photos are rectangular, making the sensor square would entail
    additional expense that would not be used much of the time.
     
    Mark Roberts, Feb 16, 2004
    #19
  20. Mark,
    if you are right about sensor price with regard to the lens (well I mean the
    whole array that constitutes what in Germany is called the 'Objektiv', eg. all
    the glass in its mounting) I will retract from my position. But frankly I
    would be surprised.
    Guenter
     
    Guenter Fieblinger, Feb 16, 2004
    #20
    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.