1. Reminder: Please use our affiliate links for holiday shopping!

14-Bit vs. 12-Bit RAW

Discussion in 'Open Discussion' started by Biro, Apr 4, 2016.

  1. Biro

    Biro Mu-43 All-Pro

    May 8, 2011
    Jersey Shore
    Steve
    • Like Like x 1
    • Informative Informative x 1
  2. Ross the fiddler

    Ross the fiddler Mu-43 Hall of Famer

    That just shows that as Olympus shooters we shouldn't be concerned that we only use 12bit.
     
  3. MarkRyan

    MarkRyan Instagram: @MRSallee

    772
    May 3, 2013
    California
    I can't tell a difference in these samples, and I certainly don't know enough about digital imaging to understand the benefits of 12-bit vs. 14-bit. I've read Ming Thein's reviews of the Sony A7 cameras and he often mentions that their use of 12-bit files produces some noticeable artifacts in shadows. I believe Sony released a firmware update to allow the cameras to save 14-bit files. Curious if it's placebo, or perhaps the difference is camera-specific. (Or perhaps the issue attributed to the color depth is actually an effect of some other variable.)
     
  4. wjiang

    wjiang Mu-43 Hall of Famer

    The problem was that they were using lossy compression on the RAW, which isn't necessarily a bad way to quantize down, except that the way their algorithm worked created artifacts.
     
    • Informative Informative x 1
  5. pellicle

    pellicle Mu-43 All-Pro

    Feb 10, 2010
    Southport, OzTrailEYa
    pellicle
    Hi
    I also felt that there was a lack of clarity on how far they exposed to the RIGHT and there seemed to be no discussion on how delicate highlights were handled. Essentially it was either dull soft light (which does not usually press the issue) or full scale blow out material.

    (inappropriate example deleted to save space)
     
    Last edited: Apr 5, 2016
  6. b_rubenstein

    b_rubenstein Mu-43 All-Pro

    Mar 20, 2012
    Melbourne, FL
    The difference in bit depth has nothing to do with dynamic range. An A-to-D converter has an output voltage range. The number of discrete voltage steps in the range is broken into is the bit depth. Lots of bit depth lots of little steps, a few bits bigger steps; the total range doesn't change. The granularity of the steps is smaller.

    There's a reason hire people that went to schools to learn this stuff, rather than people who just read internet forums.
     
  7. pellicle

    pellicle Mu-43 All-Pro

    Feb 10, 2010
    Southport, OzTrailEYa
    pellicle
    true, my above is incorrectly assuming that the quanta steps equate to similar counts (which they may not) ... but I think its also clear that the SuperCCD produces a greater voltage saturation level as well from the pictures supplied (not just finer quantization of the same).

    Also I think you may have missed something from your point:
    perhaps the word to

    There is no reason to resort to denigration of peoples education because they make a simple error in thought. That sort of thing should be on DPReview forums not here. (although they'd be less oblique than you have).
     
    Last edited: Apr 4, 2016
    • Like Like x 2