Sigma RAW format?

Discussion in 'Digital Photography' started by Roland Karlsson, Aug 1, 2003.

  1. Anyone knows the Sigma x3f RAW format?

    Anyone hacking RAW format converters for Sigma?

    Roland Karlsson, Aug 1, 2003
    1. Advertisements

  2. Roland Karlsson

    John Bean Guest

    Why bother when SPP is effective and free?

    THe Adobe RAW plugin for PS7 will load X3F files. But it costs $99 and
    is not as effective (by a long way) as SPP.
    John Bean, Aug 2, 2003
    1. Advertisements

  3. The same motivation as climbing Mount Everest,
    because it is there! :)

    Roland Karlsson, Aug 2, 2003
  4. I have just found it - and - Dave makes a fantastic job
    when it comes decoding. He is a decoding wizard.
    Unfortunately - he is not as good when it comes to
    interpolation. The Bayer interpolation he makes is not
    good and the SD9 one looks weird.

    Roland Karlsson, Aug 2, 2003
  5. Roland Karlsson

    Lionel Guest

    Uh. The SD9 doesn't need Bayer interpolation, & Dave's decoder doesn't
    attempt to provide one.
    Lionel, Aug 2, 2003
  6. Yes, thats right - no Bayer interpolation - of course.

    But ... have you looked at the code?

    If you look in "foveon_interpolate()" you see that it is rather large.
    A small part of it (20 lines or so) makes the color space conversion.
    The rest is (as far as I can understand) "interpolation" in order to
    minimize noise. I think this look by far too complicated. But I might
    be wrong - of course. I have a picture of how I should have made that
    "interpolation", and it partly matches David's code (I think). I
    should have done it this way

    * Create a monochrome low noise picture from all three channels.
    * Create a (noisy) color picture, converting the color space.
    * Optional: smooth the hue in the color picture (Don't know
    if this is neccessary).
    * Use the monochrome picture as luminosity to denoise the
    color picture.

    BTW - my comment that the Bayer interpolation code is
    not good comes from comparisons with Canon's RAW converter; it
    does a much better job. Looking at David's code for the Bayer
    interpolator, you can also see that it is somewhat ad hoc. If
    you read about research Bayer interpolation on the net you see that
    there are more advanced methods available.

    And the latter (being ad hoc) I think is David's strong side. Not
    many could have made a converter for 66 cameras with RAW format,
    and I am certain that only one in the world have done so - David.
    This is impressive! Kudos to David for this!

    Now - it would be nice if people with knowledge about picture
    conversion could help David with the part that makes optimal
    looking pictures. David has done the hard work to crack the

    Roland Karlsson, Aug 2, 2003
  7. Oops - I see that David has used a new algorithm from Ting Chen's
    overview over Bayer interpolations. So - this might be wrong now.

    Roland Karlsson, Aug 3, 2003
    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.