Moon shots

Discussion in 'Digital SLR' started by Sheldon, Feb 26, 2005.

  1. Sheldon

    JPS Guest

    In message <>,
    Well, I suppose a camera could have a really bad gain circuit that goes
    against this trend. Don't know how true it is, but Steve Giovanella
    (George Preddy) claimed that his SD9 gave cleaner results pushing ISO
    100 to 400 than shooting at 400. I really find that hard to believe,
    though.
    Yep. Rounding off "throws" the noise and signal to wider extremes. ISO
    100 would have a lot less noise than it has now, if it were digitized to
    more bit depth.
    Not really. People complain that they want cameras in the future with
    more dynamic range. My point is that the sensors are already there; it
    is the ADC circuits that are not delivering.
    --
     
    JPS, Mar 1, 2005
    #41
    1. Advertisements

  2. Sheldon

    Alan Browne Guest

    Alan Browne, Mar 1, 2005
    #42
    1. Advertisements

  3. Sheldon

    JPS Guest

    In message <d02uhp$mns$>,
    There might be some cameras that do have different gains for different
    channels. I don't think it is common, though. I think it would hard to
    be consistent with all the switching, and it would probably be slower.
    If it could be done without problems, though, images would probably be
    better than they are now. Even under sunlight, most cameras have a
    difference in sensitivity between the channels that makes them quite
    unbalanced, even with greyscale subjects (like a white statue).
    --
     
    JPS, Mar 1, 2005
    #43
  4. I think JPS's point is that some people put the cart before the
    horse and think that lower ISO means less noise no matter what the
    exposure. Certainly, if you can get all the light you need and
    expose at a low ISO, that's better.

    But for someone like me who shoots at night in available light a
    lot, JPS's comments are useful and welcome. Not that I was running
    around underexposing at ISO 100 before ... but now I can explain
    *why not* in more-or-less everyday English.
     
    Ben Rosengart, Mar 1, 2005
    #44
  5. Are you're saying that ISO 100 (or 200 on a D70?) *never* uses the
    most significant bits in its 12-bit "words"?
     
    Ben Rosengart, Mar 2, 2005
    #45
  6. Sheldon

    JPS Guest

    In message <>,
    No. I'm saying that 12 bits is not enough for the kind of dynamic range
    people want, and the better sensors are already ready to give more depth
    than 12 bits. If this were not true, ISO 800 and above would be totally
    worthless garbage, even on DSLRs.
    --
     
    JPS, Mar 2, 2005
    #46
  7. Sheldon

    JPS Guest

    In message <d02veq$on4$>,
    That's a "Canon Digital" camera strap on top of a Sam Ash Guitar bag, in
    native RAW color balance for incandescent light.
    --
     
    JPS, Mar 2, 2005
    #47
  8. Sheldon

    JPS Guest

    In message <>,
    No, and I didn't imply anything like that. My point is that in choosing
    parameters in a compromised photon budget, going to higher ISO is not
    necessarily a compromise in quality. Of course an ISO 100 shot at 0 EC
    is much less noisy than an ISO 1600 shot of the same thing with an EC of
    0. A low-contrast subject, however, may record better at ISO 400 and +2
    EC than at ISO 100 with "proper exposure".
    You understand incompletely. Very little of the noise of higher ISOs is
    caused by higher amplification. Most of it is the result of a poor
    signal-to-noise *RATIO*.
    --
     
    JPS, Mar 2, 2005
    #48
  9. Sheldon

    JPS Guest

    In message <>,
    Besides the noise issues, gravitating towards the lowest ISO can also
    compromise shutter speed, depth of field, and make the sensor more
    likely to bloom where there are specular highlights.

    It is good to know what you are really compromising when you make your
    choices; it is good to know that it is low signal-to-noise ratios and
    bit-depth utilization, not ISOs per se, that make images noisy.
    --
     
    JPS, Mar 2, 2005
    #49
  10. Sheldon

    paul Guest


    A high contrast subject risks blowing the highlights. If you are willing
    to do more work, two exposures can be merged.



    Hmm, that is significant because it assumes no adjustment on the ISO 100
    shot. It does need adjustment for the +2 ISO 400 shot.

    Thanks for that observation! Has it been tested though?
     
    paul, Mar 2, 2005
    #50
  11. It's a shame when cousins marry.
     
    Randall Ainsworth, Mar 2, 2005
    #51
  12. Sheldon

    Mister Guest

    Every time I buy a new camera, it isn't long before I try to shoot the moon...

    The D70 is the first camera I had good luck with... I too was surprised by how
    bright it is.... I started in manual at 1/2000 F10 and got the best shot at
    1/4000 F9... don't remember the ISO but may have been 400... I was up north so
    that may affect things too...

    I was hand holding a 300mm and it isn't quite enough!


    Could you be so kind as to explain that 'Sunny 16' ??

    Does that mean you set the shutter to the ISO at F16 in the sun??

    I seem to remember something about setting your shutter to 1/100 and then...
    something about green grass... dam I forget!
     
    Mister, Mar 2, 2005
    #52
  13. Sheldon

    Bill Guest

    Yes.

    The "sunny 16" rule means that on a bright and clear sunny day, you can
    set the camera aperture at f/16 and the shutter speed at 1/ISO or as
    close as you can get, and get a proper exposure. So with an ISO setting
    of 100, the ideal shutter speed would be 1/100 at f/16.

    It's a sliding scale, so it also means that if you want to stop down to
    f/22 for a landscape shot, you can shoot at 1/50 and get the same
    results. Or for less DOF, open up to f/11 and shoot faster at 1/200.

    This is photography 101. :)

    The fun part about a new digital camera, is you can practice all you
    want till you get it right...no need to worry about bad shots and film
    costs.
     
    Bill, Mar 2, 2005
    #53
  14. Sheldon

    Owamanga Guest

    Err, yes John, that's why we dial in a higher ISO when we need it.

    I'm not absolutely certain what the point of your test was.

    "Use a higher ISO rather than severely underexposing the shot."

    Not a bad piece of advice, but nobody in their right mind would ever
    shoot 4 stops under anyway - would they?
     
    Owamanga, Mar 2, 2005
    #54
  15. Sheldon

    Owamanga Guest

    Guys, a four stop underexposure is an insane thing to try if you want
    a good quality image. If other aspects prevent you from increasing
    shutter speed, opening aperture, or improving the lighting of the
    scene, you *have* to bump the ISO, it's second nature to any digital
    photographer.
     
    Owamanga, Mar 2, 2005
    #55
  16. Sheldon

    Owamanga Guest

    <g>
     
    Owamanga, Mar 2, 2005
    #56
  17. Sheldon

    Owamanga Guest

    In english - the scene is darker, which is why you needed to key in
    the higher ISO ?

    So you think people compare daylight photography at low ISOs with
    dimly lit candle-light photography at high ISOs and place all the
    noise blame on the ISO rather than the lack of 'signal'.
     
    Owamanga, Mar 2, 2005
    #57
  18. *whoosh*
     
    Ben Rosengart, Mar 2, 2005
    #58
  19. I understand it better than you think. I understand that there exists
    an inherent signal to noise ratio associated with every picture I
    take. I choose to optimize the signal to noise ratio by adjusting my
    combination of shutter speed and aperture so as to play nicely with my
    lowest ISO setting. If that's not an option, I'll jack up the ISO
    setting (turn up the amplifier) with the knowledge that whatever noise
    that already exists due to the signal to noise ratio will be
    exaggerated and added to by the amplification process. You started
    this whole discussion by asserting that, and I quote, "Higher ISO does
    *NOT* equal more noise." Well I'm sorry, but it does. That's a
    simple matter of physics. You seem to understand this stuff well
    enough to know that. What you might have asserted instead is that
    higher ISO settings don't contribute as much noise to the capture as
    the initial signal to noise ratio.
     
    Brett Wheeler, Mar 2, 2005
    #59
  20. You could trade dynamic range for a degree of posterization, I think.
    Huh?
     
    Ben Rosengart, Mar 2, 2005
    #60
    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.