Reichman on EVF's and the future of optical viewfinders

Discussion in 'Digital Cameras' started by RichA, Feb 17, 2011.

  1. RichA

    RichA Guest

    http://www.luminous-landscape.com/reviews/cameras/panasonic_gh2_revisited.shtml

    The Viewfinder (Panasonic GH2)

    Potentially the most controversial aspect of the GH2 is its electronic
    viewfinder. The way that MFT cameras with viewfinders achieve their
    small size is by substituting the usual mirror box and prism assembly
    for an electronic viewfinder – essentially a small video screen
    similar to what one finds on a video camera.

    Digicam EVFs tend not to be of very high resolution or brightness, but
    the one on the GH2 (and the GH1 originally) is very good indeed. In
    fact, I would say that it is an many instances preferable to the
    dismal (meant literally) and small optical viewfinders found on the
    smaller and usually inexpensive DSLRs.

    The EVF on the Sony A55 is also very good (though I haven't had a
    chance to do a side-by-side comparison), and similarly I would prefer
    these any day to a small and dim optical reflex system, especially
    those on cheaper systems that use pentamirrors instead of true prisms.

    No – these EVFs are not as bright and clear as a good bright reflex
    system, especially one on a full-frame body. But, the trade-off in
    size and weight is considerable. Also, the ability to have display
    overlays, such as a live histogram, goes a long way to making this new
    alternative viewing system attractive.

    Frankly, the writing is on the wall. It won't be more than a few years
    until the vast majority of new camera model with viewfinders dispense
    with prisms and mirrors and replace them with EVFs.

    Get used to it. It's not that we as photographers are necessarily
    asking for this (though EVF display technology is getting better all
    the time). It's just the pressure of industry economics. Price
    competition is fierce. Moving mirror assemblies and glass prisms are
    expensive to manufacture and assemble. High quality EVFs are not
    exactly inexpensive at the moment, but as with all high volume silicon
    based products, prices will inevitably fall. When that happens mirrors
    and prisms will be relegated to only the high end, where users are
    willing to pay for special capabilities.

    About 90% of the time I am not displeased using the GH2's EVF. It's
    even possible to forget that there's anything different going on. But
    in low light things get a bit weird. With a fast lens (like the
    remarkable Nokton f/0.95) the view actually becomes brighter than
    reality and the final result not what one expects.
     
    RichA, Feb 17, 2011
    #1
    1. Advertisements

  2. RichA

    Ofnuts Guest

    So the viewfinder luminosity is a matter of mirrors
    So the viewfinder luminosity is a matter of lenses

    Which one is true? Could it be that the entry level DSLR are always
    tested with their not-so-fast kit lenses?

    And is there any measure anywhere of the actual difference between
    pentamirrors and pentaprisms?
     
    Ofnuts, Feb 17, 2011
    #2
    1. Advertisements

  3. RichA

    Alfred Molon Guest

    Actually an EVF is brighter in low light than an OVF, because the EVF
    can amplify the light, while the OVF cannot.
    That is certainly a factor.
    I wouldn't know why a pentaprism should be brighter than a pentamirror.
    Mirrors should reflect 100% (or close to) of the light, so it should not
    matter if it's a mirror or a prism.
     
    Alfred Molon, Feb 17, 2011
    #3
  4. It may be coincidental: high end (d)SLRs tend to have big heavy pentaprisms,
    and are designed with 100% coverage and exit pupils large enough that even
    people with thick glasses and long noses can use them comfortably, whereas
    low-end cameras are made with mirror assemblies, and are designed to be too
    small, have less coverage, and tiny scrunched exit pupils that make using
    them unpleasant without contacts and major plastic surgery. So in real life,
    pentaprisms are way better than mirror viewfinders. But the big and easy to
    use vs. small, scrunched, yucky, and dim may have more to do with the
    designs than the technology used to implement the designs. Maybe.

    Whatever. Look through a high-end Nikon or Canon at a store some time and
    you'll be unhappy with you midrange dSLR for the rest of your life. Sigh.
     
    David J. Littleboy, Feb 18, 2011
    #4
  5. RichA

    Dudley Hanks Guest


    Once you learn to "point-and-shoot" like a blind guy, it doesn'
    t matter which you use ...

    If you want lessons, just let me know ;)

    Take Care,
    Dudley
     
    Dudley Hanks, Feb 18, 2011
    #5
  6. RichA

    Paul J Gans Guest

    There is a very slight light loss each time the light goes from
    air to glass and glass to air. Mirrors have at least six such
    transitions. A pentaprism has only two.

    That said, the loss is VERY small and I doubt it is detectable.
     
    Paul J Gans, Feb 18, 2011
    #6
  7. RichA

    Me Guest

    These mirrors don't have a glass/air transition, as the glass is
    "silvered" on the reflective side. Light loss is probably significant.
    How do you come up with six transitions for pentamirror?
    I count 3 - from one side to the other side of the angled roof of mirror
    assembly, to front mirror.

    I'm not sure that modern pro DSLRs with high eyepoint and focus screens
    optimised for AF systems and "100% VF" are as "accurate" as some wishful
    thinkers may like to believe. Put such a camera on a tripod, look
    through the VF at the edge of the image and move your head from side to
    side to see what I mean. An EVF would actually be better.
     
    Me, Feb 18, 2011
    #7
  8. RichA

    RichA Guest

    Odd thing is, mirror systems can be superior to prism systems and are,
    in different realms than cameras. Prisms induce chromatic aberration
    (something mirrors don't do) and absorb more light than the newest
    reflective surfaces of mirrors.
     
    RichA, Feb 18, 2011
    #8
  9. RichA

    Alfred Molon Guest

    How much exactly?
     
    Alfred Molon, Feb 18, 2011
    #9
  10. RichA

    PeterN Guest

    Another question you will evade:
    Which camera prism system are you spouting about?
    Where is the scientific backup for your statement?
     
    PeterN, Feb 18, 2011
    #10
  11. True. But it can only amplify what the sensor delivers,
    and delivers quickly (can't expose for, say, a modest 1/15s,
    or you'll have lots of fun following the action) while the eye
    has all the evolution behind it to cope with the dark at night.
    Including longer integration times without impeding following
    the action. (Or rather, if your eye cannot follow the action
    any more, you cannot blame the camera.)

    And of course you loose resolution due to the rather limited
    number of subpixels in an EVF.

    -Wolfgang
     
    Wolfgang Weisselberg, Feb 18, 2011
    #11
  12. RichA

    Me Guest

    FWIW, visible spectrum light loss from front "silvered" mirrors seems to
    be quoted (google for sources) as being in the range from 5-12%, (X3, as
    it's reflected 3 times).
    They use prisms in periscopes and binoculars etc for the same reason,
    there are much lower light losses through internal reflection in a prism
    than with mirrors.
     
    Me, Feb 18, 2011
    #12
  13. RichA

    Paul Furman Guest


    Interesting point though. Mirror telescopes do not have chromatic
    aberrations; glass lens systems do. But yeah, cheap DSLRs have mirrors,
    high end models have glass. I have no answer to that contradiction.
     
    Paul Furman, Feb 18, 2011
    #13
  14. RichA

    PeterN Guest


    The mirror and prism are part of the viewfinder. The exposure plane in
    both SLRs and DSLRs is from the lens to the medium. (Either sensor or
    film.)



    Peter
     
    PeterN, Feb 18, 2011
    #14
  15. RichA

    Ofnuts Guest

    To put that in photographic perspective .88**3 is .68 (roughly
    1/sqrt(2)) so that's half a diaphragm stop. To emulate that, set your
    lens at max opening minus half a stop, and depress the DOF check button.
     
    Ofnuts, Feb 18, 2011
    #15
  16. RichA

    Me Guest

    Except it might not work like that...
    With an f1.4 lens, there's no visible difference in DOF preview VF
    brightness unless stopped down to f2.8 or smaller. The reason for that
    has been posted here before, but I've forgotten it. The maximum
    brightness (on Nikon anyway) seems to be just below f2.8.
     
    Me, Feb 19, 2011
    #16
  17. RichA

    Paul Furman Guest

    Right, it doesn't effect the image but prisms presumably provide a
    brighter viewfinder since that's what you see in more expensive DSLRs. I
    guess it's as simple as; the cheap mirrors used in entry level DSLRs
    lose some light vs glass prisms.

    http://en.wikipedia.org/wiki/Pentamirror
    "(more precisely) referred to as roof pentamirror because of the
    roof-like ridge."
    http://en.wikipedia.org/wiki/Pentaprism
    They both have mirrored surfaces... <shrug>

    Here's a weird one I have from an old microscope:
    http://www.flickr.com/photos/edgehill/4819890710/
    -it makes a 45 degree angle rather than 90 and I don't think it has any
    mirrored surfaces...
     
    Paul Furman, Feb 19, 2011
    #17
  18. Prisms don't cause chromatic aberration if the light enters and exits
    them at near vertical to the glass-air surface transition.
     
    Chris Malcolm, Feb 19, 2011
    #18
  19. RichA

    Alfred Molon Guest

    Resolutions of EVFs are going to increase in the future and for manual
    focus EVFs can show an enlarged section of the image, down to pixel
    level if you choose so, allowing precise manual focus, much more
    accurate than possible with an OVF.
     
    Alfred Molon, Feb 19, 2011
    #19
  20. Resolutions of EVFs are going to increase in the future and for manual
    But according to Rich, if you need magnification you are blind.

    David
     
    David J Taylor, Feb 19, 2011
    #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.