Which free software could acquire 48 bits color depth pictures from a scanner ?

Discussion in 'Scanners' started by Guilbert STABILO, Nov 16, 2008.

  1. Guilbert STABILO

    Guest Guest

    photoshop is not limited to 4 channels, although the vast majority of
    images will be 3 or 4 channels,
    it's false.
    that's exactly what it does.
     
    Guest, Nov 17, 2008
    #21
    1. Advertisements

  2. Guilbert STABILO

    Ray Fischer Guest

    And I would like to point out that the MINIMUM precision the PS allows
    is 24 bits, or 8 bits per channel. And the fact that one can print
    16-bit/channel images doesn't mean that the device provides anywhere
    near that big a range of colors.
     
    Ray Fischer, Nov 17, 2008
    #22
    1. Advertisements

  3. Guilbert STABILO

    Ray Fischer Guest

    Of course it is. I've CS4 running right in front of me. One can
    easily do operations on single channels or combine them.
    Photoshop allows for many channels. Not just the common R, G, &B.
    No kidding?
    Sometimes. What about 64-bit color? Would you know that that refers
    to CMYK at 16-bits per channel? Or LAB color at 16-bits per channel?
    You've never used Photoshop, have you?
    If there's 32-bits to be processed with a math operation then
    it's the math precision.
    Ah, so you really don't know what you're blathering about.
    Did you forget that you don't know what you're blathering about?
     
    Ray Fischer, Nov 17, 2008
    #23
  4. It's been obviouis since your diatribes against gay-marriages of what you really
    enjoy. It's nice to see your follow-up comments reaffirm what everyone's already
    deduced about you being an insecure closet-case that's only trying to appear
    macho online with the screen-name of "Hemi-Powered". Overcompensate much?
     
    Brandon B. Gaston, Nov 17, 2008
    #24
  5. Guilbert STABILO

    Eric Stevens Guest

    People seem to be becoming confused. Whether a computer is an 8, 16,
    32, 64 or 80 bit machine is important from the point of view of the
    accuracy of the calculations its calculating engine is asked to carry
    out.

    This is NOT the same as the number of bits used in the RGB/CMYK data
    channels used to describe the images.

    When an image is manipulated in any way the new RGB/CMYK data is the
    product of the computer's calculating engine. It is at this point that
    the accuracy of the calculations become important.



    Eric Stevens
     
    Eric Stevens, Nov 17, 2008
    #25
  6. Guilbert STABILO

    Eric Stevens Guest

    He's correct.
    Not necessarily so.
    I suspect the problem is that you don't understand what he his
    blathering about. Math precision and colour depth are two entirely
    different things.



    Eric Stevens
     
    Eric Stevens, Nov 17, 2008
    #26
  7. Guilbert STABILO

    Eric Stevens Guest

    Don't forget that Motorola made an 80 bit Arithmetic Processor Unit
    (APU) for use with the 68000 series of CPUs well before the first IBM
    PC was built.



    Eric Stevens
     
    Eric Stevens, Nov 17, 2008
    #27
  8. Guilbert STABILO

    ChaseGoldwin Guest

    Not confused at all. What you fail to realize is that the larger the values of
    data the more math accuracy is needed to manipulate that data without losing
    information from that data. Educate yourself.

    (Why do these trolls even bother trying to catch up?)
     
    ChaseGoldwin, Nov 17, 2008
    #28
  9. Guilbert STABILO

    Ray Fischer Guest

    No it isn't. Even an 8-bit processor can do 128-bit math.
    What do you think the math is working on?
     
    Ray Fischer, Nov 17, 2008
    #29
  10. Guilbert STABILO

    Ray Fischer Guest

    He's still an idiot.
    When not?
    Just because you say so?

    Heres a clue: Any processor can handle any precision (within the
    limits of time and memory).
     
    Ray Fischer, Nov 17, 2008
    #30
  11. Guilbert STABILO

    Ray Fischer Guest

    You're not smart enough to be condescending.
    If there's a 32-bit channel then the math is 32 bits. They might do
    extended precision in intermediate steps for some bit depths, but I
    doubt it. I suppose I could ask some people at Adobe who I once
    worked with, but I don't think it's worth wasting their time on.
    If you're an idiot you're going to pretend that the obvious is deeply
    significant.

    sometimes:
    adverb
    occasionally, rather than all of the time
    Unless it doesn't.
    Unless it does.
    RGB+Alpha. 64 bits per pixel.
    Not since I was a graduate student.
    It is not possile to have a graduate degree in computer science
    without knowing more about computer precision in doing math
    calculations than you possibly understand.
     
    Ray Fischer, Nov 17, 2008
    #31
  12. Guilbert STABILO

    Eric Stevens Guest

    Because I'm ahead of you. But having an APU with the ability to handle
    lots of bits in its calculations doesn't mean that you suddenly have
    colour channels with gazillion bit capabilities.



    Eric Stevens
     
    Eric Stevens, Nov 17, 2008
    #32
  13. Guilbert STABILO

    Eric Stevens Guest

    True to a point. But an 8-bit floating point algorithm to emulate an
    80 bit APU is likely to incorporate an accumulation of rounding off
    errors.
    That was really what I was trying to tell him.



    Eric Stevens
     
    Eric Stevens, Nov 17, 2008
    #33
  14. Guilbert STABILO

    Eric Stevens Guest

    Of course the math is working on the individual data channels but
    there is no need for number of bits in the channels to match the
    number of bits in the arithmetic processors.



    Eric Stevens
     
    Eric Stevens, Nov 17, 2008
    #34
  15. Guilbert STABILO

    Eric Stevens Guest

    How many bits does a math processor need to handle for it to be able
    to deal with the Encyclopaedia Brittanica?
    Yes, especially when I talk about my own suspicions.
    And what has that got to do with the number of bits used to code an
    image data channel?



    Eric Stevens
     
    Eric Stevens, Nov 17, 2008
    #35
  16. Guilbert STABILO

    Steve Guest

    *IF* the math engine is implemented correctly (and that's a big if
    since many are not) then you only need one more significant bit to
    carry through the calculations than you have in the data.

    The 32 bit IEEE 754 floating point format that most computers use only
    has 24 significant bits. But one of them is a sign bit. Since image
    data is unsigned, you can only represent 23 bit integer image data.
    Which means you can use 32 bit floating point math on 22 bit or less
    unsigned integer image data if you want to keep your precision through
    multiple calculations.

    If you break up your 48 bits of color depth pictures into individual
    channels, 32 bit floating point math should be enough if the math
    engine is implemented correctly. I.e., it does proper rounding at
    each intermediate stage and any time you want to convert back to
    integer. But carrying through the extra couple of bits is a good
    thing.

    Going to double precision (64 bit floating point math) is a waste of
    time (literally) in this instance.

    Now for the caveat I mentioned above: I've seen quite a few
    implementations of math routines where order of processing, incorrect
    rounding, etc., all take their toll on the precision/accuracy of the
    result when theoretically it should not. I've seen signal processing
    routines that work on 16 bit signed integer input data using 32 bit
    floating point math that give a different result if you change the
    math to 64 bit floating point. Just compiling the routine using 2
    different compilers changes the results. In a perfect world that
    should not happen. But it does.

    Steve
     
    Steve, Nov 17, 2008
    #36
  17. Guilbert STABILO

    Steve Guest

    Which, of course, is meaningless unless you define the format of the
    32 bits. I.e., if you have 32 bit integer data, you cannot do 32
    floating point math on it and expect to keep the level of precision
    through even the first calculation that is present in the input data.

    Steve
     
    Steve, Nov 17, 2008
    #37
  18. Guilbert STABILO

    Ray Fischer Guest

    Nope. It will produce the exact same result.
     
    Ray Fischer, Nov 17, 2008
    #38
  19. Guilbert STABILO

    Steve Guest

    Well, not quite. The 80 bit coprocessor for the 68000 series (68881)
    wasn't even anounced until a year after the first IBM PC's were
    already in use.

    Steve
     
    Steve, Nov 17, 2008
    #39
  20. Guilbert STABILO

    Ray Fischer Guest

    Sigh. You guys have very little idea how computers do math.

    Did you know that a lot of image processing these days is done by the
    graphics processor in the video card? Did you know that SIMD
    instructions are used to process multiple chunks of data concurrently?
     
    Ray Fischer, Nov 17, 2008
    #40
    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.