Digital "Veil" of Death

Discussion in 'Digital Darkroom' started by kevin_delson, Feb 21, 2010.

  1. I believe I have my own answer here, but wanted to post this for more opinions/answers.
    The two sample photos here intrigue me.
    One is untouched and the other has had only a black point and white point set.
    Disregard the blown out sky, I have no question with that as I was metering off the foreground.
    This image is quite obviously a rather low contrast scene.
    Analysis of the histogram seems pretty straight forward to me, so I don't need to hear "It's underexposed"..I know it is underexposed.
    Regardless of how I exposed this scene, I am unable to reproduce the (processed) look out of the camera.
    Of particular concern to me is the overall (Veil) or (fog) to the image when the two are compared.
    So, the question is:
    "Can the post processed image be produced out of camera w/o post processing?"
    I can certainly construct a custom tone curve, though that is obviously NOT the way to go around shooting.
    00VpOQ-222701584.jpg
     
  2. Processed
    00VpOT-222701684.jpg
     
  3. "Can the post processed image be produced out of camera w/o post processing?"​
    Yes, you need to experiment adjusting the camera's contrast and saturation settings as well as look into scene styles accessed within the camera's menu system and external rotary dial if it comes with one.
     
  4. Thanks Tim, but I already considered those options and tried them...contrast adjustments runs too high a risk that may clip blacks.
    The "Veil" as I like to call it really concerns me.
    Here's another image that may make it easier for all to see the difference.
    00VpOs-222707584.jpg
     
  5. With light that flat, you're going to just have to take your chances with in-camera processing, or make those few mouse clicks after the fact. Of course, you can have the best of both worlds, and shoot RAW+JPG.

    The main issue, though, is that you want more contrast than exists in the scene you're shooting. When you want to take control of the curve, you've got to take control of it. The camera isn't the best venue for that. The best bet, of course, is to shoot under better light. Or, bracket and combine.
     
  6. I believe I have my own answer here,​
    Ya..I pretty much came to that conclusion too Matt.
    I could as Tim mentioned, crank up the contrast in camera, but I am not a believer in molesting the original image.
    I was no doubt hoping against hope in this. :)
    JPEG? What's that? LOL
    It was interesting to take the image into Capture NX and design the perfect curve, but every lighting situation isn't the same obviously, so loading any particular curve into the camera is not the answer.
    OK....OK....Perhaps someday we will have camera's that record as we think. ;)
     
  7. This happens when the luminance range of the scene falls short of the dynamic range of the film or sensor. It's not unique to digital. In the film days one would shoot a higher contrast film and/or make a higher contrast print to try and eliminate the "veil". One of the great advantages of scanners, once they became widely available, was that one could scan film and precisely control the black and white points to clear any veiling.
    You cannot balance out the histogram as well in camera as in post, though you may be able to manipulate contrast, saturation, and styles to get close enough to satisfy your needs.
     
  8. One option that might help ---
    With cnx2 open, look in the edit list >camera settings and find picture control
    To the left of the reset button (above noise reduction section) is a little icon with a rectangle, a gear, and an arrow. click the arrow and you will be presented with a menu of picture control options. Choose (click) launch picture control utility, and you can create a custom curve. Build an S curve to your preference, save it, and it can the be loaded into your camera as an option that you may choose for flat light situations. Being an S curve it should increase the midtone contrast without affecting the absalute black or white clipping.
     
  9. deleted duplicate response
     
  10. Thanks Wayne,
    That's what I'll be doing just to save some time in post..Now all I need to do is remember to (un-load) the curve when I'm done shooting low contrast subjects. LOL
    I'm one of those who has acquired the bad habit of NOT looking at my shoot menu often enough. :)
     
  11. Here's a crop of your unprocessed image with an LCH curves adjustment. It's just an S-curve to compress highlights and shadows a bit and increase the contrast of the mid tones. Just an example that you'll have much greater flexibility in post.
    How much or little contrast is a personal preference. The sky in this shot is pretty much featureless.
    00VpYq-222795684.jpg
     
  12. In my opinion,camera can hardly get the image which is the same as one has been set some numeric values in the curve. Fixed algorithm is used to change the pixel point color level when the white and black point are defined in software such as photoshop,but when we set contrast out of camara,the algorithm is different
     
  13. "...but I am not a believer in molesting the original image."
    So when you shot print film you had the photos printed as negatives? :) All photos must be processed. You can use the crude in-camera processing software, or more sophisticated out-of-camera processing software. Placing importance on the physical location of where the software is stored and run doesn't make much sense to me.
     
  14. You mis-understood me Matt.
    I meant I do as little in camera processing as possible.
     
  15. I would just be grateful that the entire scene fits on the histogram! This provides a situation where you can easily "fill out" the data to fit the dynamic range of your presentation medium. This gives you an easy way to make a boring, plain scene into something dynamic and vibrant.
    The only way to do this in-camera is to only shoot subjects that are perfectly balanced, contrasty, dynamic, and vibrant, with a perfect balance of highlights and shadows. If your subject is anything else, that's not what you're going to record in camera.
    By the way, I believe Curves is almost 100% of the time a more subtle and effective way of achieving this in Post-processing than Levels. Levels always leaves me cringing a little, and it doesn't give enough control in the midrange.
     
  16. Let's take a different approach; what is your final output? A print? Screen? and if screen for projection or direct view? As people shoot more digital images, and get access to useful yet cumbersome things like RAW files and advanced editing software, they get lost in the "Forest" of technology and fail to perceive the "tree" or final print.
    If you're serious about photography, the final goal should be as good a print as you can make. Whether you exhibit in a gallery, publish in a book, or just hang on your own wall for enjoyment. The printed image is the great leveller of all photographs. It's rare to get more than about 5 stops of tonal scale on the paper.
     
  17. In LR, I would lower Exposure, increase Brightness or Contrast, add Clarity, depending. This is why I don't use Capture NX.
    By the way, if you are shooting NEFs, use all the in-camera processing you want... it's not permanent... t
     
  18. This is why I don't use Capture NX.​
    Huh?
     
  19. I'm confused by your post, Kevin. Do I have this right - you're shooting RAW (NEF) and wondering why you can't get the same look in-camera as you get from post-processing the RAW image...that the unprocessed RAW looks relatively flat and dull?
     
  20. D.B....
    You have it partially correct.
    Yes, I shoot RAW 90% + of the time.
    No, I don't expect out of camera images to look anything like PP images.
    I'm not a newbie at this either..been shooting professionally for 26 years.
    My concern (and I know it's probably been difficult for some to follow this thread) is the "Veil" or "Foggy" cast over the entire image in low contrast scenes.
    I don't get this in studio or in higher contrast scenes; and I know why.
    In the days when I shot film, I would simply select a film that would produce a higher contrast and print on high contrast papers.
    It is/was more of a curiosity to me that the digital sensor seems to struggle in the above scenario (low contrast) as did film.
    I compensated in the film days and fully realize digital appears to be less forgiving in these low contrast scenes.
     
  21. I don't think it's a case of the film (of the non-high-contrast variety) or the default recording mode of the digital camera "struggling" with this situation. The problem is that the scene you're shooting looks that way in the first place. You're dealing with a low-contrast scene, and it's being recorded as such. No struggle, just recording what's there (or not) in terms of contrast. I think you might be over-thinking this, or just annoyed that it takes a few more steps to do the equivalent of changing your film on the fly.
     
  22. or just annoyed that it takes a few more steps to do the equivalent of changing your film on the fly.​
    An astute observation Matt; and quite correct.
    'nuff said.
     
  23. No worries then...RAW is supposed to look flat and dull without post...more so if the scene is flat and dull - I think that's your 'fog'. To me, it would be the expected result of the scene conditions and image file format. As you know, with RAW, no contrast, color, saturation, sharpening, or WB are applied. RAW is intended for post...they go hand-in-hand with a computer. The reasons to shoot RAW are to get more working latitude for post via the increased data available (in this case, including 'fog'), or getting something that can't be otherwise gotten with in-camera processing and settings. The sensor saw the scene in its own way, and not as our brain processes a scene. In this case, unprocessed, it looks 'foggy'.
     
  24. Kevin, about your contrast comparison to film I found out recently that digital camera sensor's RGB Bayer filters are made from organic dyes which very closely captures and renders a spectral response human vision is accustom too seeing meaning the camera's response is quite accurate.
    Film couldn't do this that well but no one wanted accuracy from film anyway.
    I took some abstract shots of water lit by the sun low in the sky and it looked quite ordinary but I metered so the data landed right in the middle of the histogram because it's a low contrast scene. Viewing the image out of the context surround of a wide dynamic range outdoor environment will tend to make those shots look flat on display viewed in a darkened low dynamic range environment.
    00VpxE-222953584.jpg
     
  25. YES. In ACR simply select Auto when converting from raw. You will be close. Your 'Levels" adjustment is terrible. If you are asking if the camera can do this in JPG the answer is a flat NO.
    This such an easy image to correct in PS CS4. 2 seconds. Done.
     
  26. The OP's concern is that digital is incorrectly recording the scene, creating a 'veil'.
    That isn't true. Digital IS correctly recording the (lack of) contrast in the scene. Digital is NOT creating veiled mid-tones. It's your eye/brain image processing that 'focuses' on a small part of the scene when you're standing there, and showing you high contrast in a very tiny area of the whole 180 degree scene that your peripheral vision takes in.
    And the correct response of the photographer is to alter the tone curve to increase the local contrast so as to make an enjoyable image.
     
  27. Here's a single tone-curve adjustment as a quick example:
    [​IMG]
     

Share This Page