How does ISO setting work?

Discussion in 'Digital SLR' started by David, Mar 5, 2010.

  1. David

    David Guest

    I know relatively little about the technical workings of digital
    photography, and from what I have read here, some of you know a great deal.

    I would like to understand what happens in my camera when I change the
    ISO setting.

    I am hoping that once I understand that, I will know the (potential)
    disadvantages of always shooting at high ISO, if there are any.

    Yes, I am aware that some shots can benefit from blurred motion (streams
    & waterfalls), and a smaller lens aperture can yield greater dept of
    field. But are there other factors, such as noise, that change when I
    shoot at ISO 3200 rather than ISO 400?

    If you know of a web site that explains this, referring me to it rather
    than spelling it all out for a newbie is just fine with me.

    Thank you.

    Dave S.
     
    David, Mar 5, 2010
    #1
    1. Advertisements

  2. David

    user Guest

    Le 05/03/10 22:11, David a écrit :
    Changing the ISO settings just changes the gain in the analogic
    electronic chain before quantization (conversion to digital format).

    The main problem is that it also increases noise
     
    user, Mar 5, 2010
    #2
    1. Advertisements

  3. David

    Ron Recer Guest

    Google ISO and you should find more information than you ever want to know
    about the subject.

    You might also want to experiment with your camera at various ISO settings
    to determine which ones are acceptable for the size prints you normally
    make. The higher the ISO the more noise the pic will have. The larger the
    print the more the noise shows.

    I take a lot of wildlife photos and higher ISO settings can be a real help
    when the subject is in deep shadow and I am hand holding a long telephoto
    lens!

    Ron
     
    Ron Recer, Mar 5, 2010
    #3
  4. David

    NameHere Guest

    It doesn't increase noise per se, it only amplifies it along with the
    available useful image data. There is also an interesting aspect to image
    noise which the trolls in these newsgroups fail to understand. In many
    subjects and compositions noise can greatly enhance the quality of a photo.
    Noise is never a problem in the hands of an artist and a pro. They can use
    it to their advantage whenever it presents itself. It seems to only be an
    incessant problem in the minds of resident-trolls who don't know anything
    about real photography. Their own minds' noise is preventing them from
    understanding anything.

    There is also an uncanny ability of the human mind and sensory system. If
    you inject a known noise into an unknown noise source, it can actually
    increase the perception of real detail and useful information. Using a
    known noise to cancel an unknown noise. This works with many of your sense
    systems. Sight, sound, and touch (it probably works with the sense of smell
    and taste too). You can easily prove it to yourself the next time you have
    an itch and scratch it to make it go away. This is precisely how it works.
    If you have a sound signal in which there is too much noise to comprehend
    what is being said, masked by all the noise, you can inject some
    white-noise into the path and make the audible information understandable
    again. The same also works for visual noise. One of my own discoveries
    based on a situation I encountered once.

    I was trying to listen to some distant shortwave radio stations one night,
    but the station I wanted to hear had noise so bad I couldn't understand
    what was being said. I decided to go to the kitchen to cook something but
    turned up the volume first to see if the sound signal would clear up while
    I was busy cooking. While in the kitchen I still couldn't understand what
    was being said on the radio, the noise being amplified just as much as the
    information it was masking. It was a little chilly in the house so I went
    back toward the living-room, where the radio was, to turn on the fireplace
    fan. A good source of white-noise. The fireplace fan being part-way between
    the radio and the kitchen. Then headed back into the kitchen. Suddenly, the
    sound from the radio was intelligible from the kitchen, the discussion from
    the radio station clear as anything. I went back to sit by the radio
    thinking the signal had cleared up. But no, the noise was still just as bad
    as before. I went back to the kitchen and could hear the radio broadcast
    clearly again. On a hunch I turned off the fireplace fan and went back to
    the kitchen. The noise from the radio made the sound unintelligible again.
    I went back and turned on the fireplace fan. Then went back to the kitchen
    and could understand the radio broadcast clearly again. The noise levels
    from the radio never changing. The white-noise from the fireplace fan was
    masking the noise from the radio. I have since learned this works on images
    too.
     
    NameHere, Mar 5, 2010
    #4
  5. At least that explains why you failed to see the noise in your last image.
    The one you removed when you realised it had CA all of the place.
     
    Robert Spanjaard, Mar 5, 2010
    #5
  6. David

    NameHere Guest

    It doesn't increase noise per se, it only amplifies it along with the
    available useful image data. There is also an interesting aspect to image
    noise which the trolls in these newsgroups fail to understand. In many
    subjects and compositions noise can greatly enhance the quality of a photo.
    Noise is never a problem in the hands of an artist and a pro. They can use
    it to their advantage whenever it presents itself. It seems to only be an
    incessant problem in the minds of resident-trolls who don't know anything
    about real photography. Their own minds' noise is preventing them from
    understanding anything.

    There is also an uncanny ability of the human mind and sensory system. If
    you inject a known noise into an unknown noise source, it can actually
    increase the perception of real detail and useful information. Using a
    known noise to cancel an unknown noise. This works with many of your sense
    systems. Sight, sound, and touch (it probably works with the sense of smell
    and taste too). You can easily prove it to yourself the next time you have
    an itch and scratch it to make it go away. This is precisely how it works.
    If you have a sound signal in which there is too much noise to comprehend
    what is being said, masked by all the noise, you can inject some
    white-noise into the path and make the audible information understandable
    again. The same also works for visual noise. One of my own discoveries
    based on a situation I encountered once.

    I was trying to listen to some distant shortwave radio stations one night,
    but the station I wanted to hear had noise so bad I couldn't understand
    what was being said. I decided to go to the kitchen to cook something but
    turned up the volume first to see if the sound signal would clear up while
    I was busy cooking. While in the kitchen I still couldn't understand what
    was being said on the radio, the noise being amplified just as much as the
    information it was masking. It was a little chilly in the house so I went
    back toward the living-room, where the radio was, to turn on the fireplace
    fan. A good source of white-noise. The fireplace fan being part-way between
    the radio and the kitchen. Then headed back into the kitchen. Suddenly, the
    sound from the radio was intelligible from the kitchen, the discussion from
    the radio station clear as anything. I went back to sit by the radio
    thinking the signal had cleared up. But no, the noise was still just as bad
    as before. I went back to the kitchen and could hear the radio broadcast
    clearly again. On a hunch I turned off the fireplace fan and went back to
    the kitchen. The noise from the radio made the sound unintelligible again.
    I went back and turned on the fireplace fan. Then went back to the kitchen
    and could understand the radio broadcast clearly again. The noise levels
    from the radio never changing. The white-noise from the fireplace fan was
    masking the noise from the radio. I have since learned this principle works
    on images too.
     
    NameHere, Mar 5, 2010
    #6
  7. David

    NameHere Guest

    That's because you're an idiot and a troll. There's nothing that anyone can
    do about that.
     
    NameHere, Mar 5, 2010
    #7
  8. David

    NameHere Guest

    Oh look, it's the pretend-photographer off-topic troll again. What a
    surprise. The only noise in that image was in your head. You can't even
    tell what are and are not JPG artifacts in an image.
     
    NameHere, Mar 5, 2010
    #8
  9. So you admit you didn't see it.
     
    Robert Spanjaard, Mar 5, 2010
    #9
  10. David

    NameHere Guest

    Damn, you blind and useless trolls never quit do you.
     
    NameHere, Mar 6, 2010
    #10
  11. David

    NameHere Guest

    Good, then you can entertain the obvious troll.
     
    NameHere, Mar 6, 2010
    #11
  12. David

    MikeWhy Guest

    IOW: his trolling mixed with factual posts from others serves to increase
    our collective understanding by substituting a false impression of knowledge
    for actual fact. It doesn't work here, though, because the noise is not
    random at all, but strongly cyclical and colored. It tends to overwhelm what
    small signal is actually present. It is, in fact, QRM of the worst kind, an
    intentionally broadcast jamming signal.
    Of the two ways to solve the problem, I opted to install a better antenna
    with directional gain and better side lobe attenuation. At the height of the
    most recent solar cycle, I exchanged greetings from midwestern USA with a
    Moscovite on 3W of radiated RF. Outside a few isolated such instances,
    though, I personally don't find much pleasure in plumbing the depths of the
    equipment's performance. Crystal clear voice at 10W on a mid-summer's
    evening terminator bounce to Chile is more to my liking. The gear is far
    from exotic, but at the same time several steps above the pedestrian. To
    close the loop on this allegory of making images, my goals go far beyond
    simply making recognizable liknesses with minimal equipment.
     
    MikeWhy, Mar 6, 2010
    #12
  13. David

    Derp Deplume Guest

    Right.. and that leaves the third choice.. a idiot and a troll.

    What NameHere said made perfect sense. It's all about learning how
    your camera works and behaves.
     
    Derp Deplume, Mar 6, 2010
    #13
  14. David

    NameHere Guest

    Then you would have never discovered what I discovered. That you can use
    noise to cancel noise.

    Go read some more web-pages about antennas and reception, I'm sure your
    troll's imaginary radios and imaginary cameras will get along fine with
    each other.
     
    NameHere, Mar 6, 2010
    #14
  15. David

    John A. Guest

    PSA...

    He (or it) is the infamous P&S troll. His M.O. is to disparage DSLRs
    in favor of P&S cameras, or to otherwise claim that camera quality
    doesn't matter, and to call the regulars in the group(s) he invades,
    or anyone who disagrees with him, "trolls". He often backs up his
    claims or attempts to display expertise in photography by posting
    (often misinterpreted, perhaps deliberately) googled information.

    It also changes names a lot, often openly but no doubt also putting on
    a sock puppet or two.

    If you stick around you'll soon learn to recognize it.
     
    John A., Mar 6, 2010
    #15
  16. David

    Russ D Guest

    It's easier to recognize all the many dozens of resident-trolls, no matter
    what name they use or even if they use the same name all the time, by their
    always dragging things off-topic. No other information nor techniques are
    required to recognize who are and are not the real trolls in any newsgroup.
    Consider yourself one of them.
     
    Russ D, Mar 6, 2010
    #16
  17. David

    Ray Fischer Guest

    What a moron you are. Increasing noise doesn't increase noise?!?
     
    Ray Fischer, Mar 6, 2010
    #17
  18. David

    NameHere Guest

    What a moron you are. There's a difference between amplifying what already
    exists and increasing the amount of what already exists.

    Here's two bits of noise:

    . .

    Here's amplifying those two bits of noise:

    - -

    Here's increasing the amount of noise:

    . .. . .. ... .

    Here's amplifying and increasing the amount of noise:

    - -- - -- --- -

    You can now hit yourself in the head very hard with an axe while saying
    "Oh! I finally get it!" Or not, if you don't get it. If you get it hit
    yourself in the head repeatedly while saying, "Oh! I get it!", until I tell
    you to stop.

    ****, this is like trying to educate a downs-syndrome child who's not even
    in kindergarten and will probably never qualify for that level of
    education.
     
    NameHere, Mar 6, 2010
    #18
  19. David

    Ray Fischer Guest

    Apparently you're not smart enough to realize that "amplify" and
    "increase" are synonymous. But then stupid trolls never are very
    smart.
     
    Ray Fischer, Mar 6, 2010
    #19
  20. David

    Martin Brown Guest

    If you can do that then it is not really noise at all, but some system
    wide correlated defect. Scientific and astronomical imaging systems are
    hot pixel, cosmic ray and flat field corrected in a separate step. Now
    you could call that noise but it is very specific type of systematic
    error that can be calibrated out or ignored in small quantities. They
    are also actively cooled to avoid as much thermal noise as possible.

    Now that we have all given the P&S troll a good kicking it is time to
    introduce him and his sock puppets to the kill file. I suggest everyone
    else does the same and he will soon tire of posting garbage.

    Regards,
    Martin Brown
     
    Martin Brown, Mar 6, 2010
    #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.