Request for corroboration

Discussion in 'General Technical Discussion' started by hillrg, Dec 5, 2005.

  1. I have been pretty sceptical about shooting uncompressed raw. I have experimented in the past and not been able to see any difference, but those I respect keep saying it makes a difference.

    So I thought I would make another test tonight, and I did see a significant difference this time. In fact, it was so much of a difference that I would like others to see if they can duplicate my results to make sure I haven't screwed something up.

    I shot at ISO 100 and converted using identical settings in ACR to PSD in ProPhoto color space 16 bits. I made similar crops (1:1), converted to sRGB and 8 bit and saved to the web.

    The first photo is a light in our hallway. I exposed to make sure there was some blown pixels. Note the uncompressed image has smoother transitions and less contrast, making the image appear less sharp.

    53213446.

    The second image is a direct flash portrait of my son, which I deliberately made as harsh as possible. Again, the uncompressed image appears much softer. I suppose I could have missed the focus or moved the camera, although I doubt it.

    View attachment 20586

    So, I would like to see how others fare. So far, I like the compressed look better...
     
  2. Rory, good test, and an argument that I have with myself all the time :wink:

    My understanding is that "visually lossless" as stated by Nikon means that when things are done at "normal" sizes "normal" humans can't see a difference. Clearly you have shown the difference here which is the difference between 12 bits of data vs. 9 or 10 bits of data, which I believe is what you get with compressed.

    To date the best reason I have heard to shoot uncompressed is that future software may be able to take better advantage if you have thrown 0 data away. The advantage with uncompressed of course, is speed and storage space. I don't see this difference with my D2H, but then I only have 1/3 of the pixel density to work with as well.

    Hopefully folks like Iliah will jump in with a more technical description of this. I see what you mean regarding the intial "look", but what happens once you complete a full workflow, with USM in post-process? Does the uncompressed image look better at the end of the day?
     
  3. I don't know, good point. I'm waiting to see if anyone else gets the same results.
     
  4. I think that part of the reason I don't see this much difference is simply the MP difference with the D2H. You might want to PM Iliah, I'm sure he has more specifics on this as well.

    Glad you brought this up, however.
     
  5. lowlight_junkie

    lowlight_junkie

    353
    Nov 28, 2005
    Very interesting post, not one I am able to contribute to with any useful input unfortunatley, but one I will follow closely!
     
  6. Very interesting Rory, thanks for sharing this information.
     
  7. MontyDog

    MontyDog

    Jan 30, 2005
    #1064 - You have an error in your SQL syntax;
     
  8. Chris101

    Chris101

    Feb 2, 2005
    Arizona
    I can't see much difference in the first set Paul, but in the second set, I'm gonna guess that B is uncompressed, based on the appearance of Rory's examples.
     
  9. lowlight_junkie

    lowlight_junkie

    353
    Nov 28, 2005
    I'd say in 1: a = Compressed top right seems to have lost a little detail
    in 2: b = compressed whole top and bottom row seems to have lost a lot of detail
     
  10. cwilt

    cwilt

    Apr 24, 2005
    Denver, CO
    If time permits tonight I will see if the difference between compressed and uncompressed show up on MTF50 tests.

    Rory,
    Strange, I don't think I have ever seen that big of a difference before. Did you update the firmware? Just to make sure we are all starting from the same point.:wink: Can you replicate it?

    Kind regards,
    Charles
     
  11. Flew

    Flew

    994
    Jan 25, 2005
    Alabama
    I agree with Chris.

    Now you guys have got my interest up regarding this issue (again). I'm going to take some test shots with the D4H and see if I can see any differences. :rolleyes:
     
  12. MontyDog

    MontyDog

    Jan 30, 2005
    #1064 - You have an error in your SQL syntax;
     
  13. Gale

    Gale

    978
    Jan 26, 2005
    Viera Fl
    Flewy camera with the flewy-dewy lens Paul.
     
  14. Gale

    Gale

    978
    Jan 26, 2005
    Viera Fl
    Well since I am the typo Queen, you can be typo King :>)))))
     
  15. Iliah

    Iliah

    Jan 29, 2005
    nowhere
    Dear Rory,

    "Compression" visually acts similar to USM. Removing fine gradation perceptually is sharpening.
     
  16. MontyDog

    MontyDog

    Jan 30, 2005
    #1064 - You have an error in your SQL syntax;
     
  17. In keeping with 4H there may be some connection to farming, and the use of fertilizer, or some such BS :wink: .
     
  18. Iliah

    Iliah

    Jan 29, 2005
    nowhere
    Dear Paul,

    One of the "older" methods of sharpening was just printing internegatives on high contrast film, thus removing transitional gradations :) Another one was controlled posterization using litho type film/developer combinations (Kodalith and alike ).
     
  19. MontyDog

    MontyDog

    Jan 30, 2005
    #1064 - You have an error in your SQL syntax;
     
Loading...
Similar Threads Forum Date
Request your recommendations for photo processing software General Technical Discussion Jul 13, 2012
Camera problems ~ Advice requested General Technical Discussion Oct 19, 2008
Request for your thoughts on Battery Grip for D40 and D60 General Technical Discussion May 3, 2008
35 mm PC Pano guidance requested ( ? for Randy Forshey) General Technical Discussion Mar 12, 2006