kodachrome turning blue....

Discussion in 'Film and Processing' started by jason_inskeep, Nov 5, 2008.

  1. i just got back a roll of kodachrome from the lab a week or so ago. the mounted slides looked great from my slide
    viewer, with the exception of a couple of bad exposures. so i took some slides to Ritz camera to have them scan it
    so that i could see what they looked like a bit bigger and maybe play with them a bit in photo shop and every frame
    came back with a bluish cast.
    i know that some of the kodack films have a bluish cast to them and i know that shooting during certain parts of the
    day will give a more blue cast on film than what the eye sees. while i dont have any experience with slides other than
    this roll i have never noticed quite as much of a problem with any other film i have shot.
    Do you think this could just be part of the time of day and print film just handles it better or something else.
    the only other thing i can think of is the film was in my camera for quite some time. a couple of months because i
    cant afford to get film developed to often and really am more selective with film than digital shooting you know. would
    the time just sitting around cause this, or the scanning maybe?
     
  2. here is an example i messed up tryign to post it the first time.
    00RObu-85597784.jpg
     
  3. Cloudy lighting is blue in tone, and the Kodak slide films are known to go blue in the shadows. It could also be down to the (auto)corrections applied to the scans during the scanning process. Get a loupe and see if the scans match the actual slides in tone. To me, the posted example looks OK in terms of tones.
     
  4. Actually, I take that back, the highlights are a little bluer than I would expect, but the rest stands.
     
  5. That looks too blue to me.
     
  6. You could take them back or try to correct in Photoshop
     
  7. Are you saying that the actual slides are too blue or the scans are too blue?

    I personally find that K64 scans too blue as a default in Vuescan and they need extensive color balance and hue
    changes to make them right. So I think you need to be prepared for this. E6 is easy to scan but K14 is not so
    straightforward, as you must have heard.
     
  8. I got back some Kodachrome scans the other day. The slides themselves look fine, but the colors in the scans were indeed too blue.
     
  9. The way to know whether it is a problem with the scans or your slides is to look at the slides on color balanced (5500k) light table. If they look fine on the light table, they messed up the scans. If they look like this on the light table, well, then technically you should have used a warming filter that day. But practically, it doesnt matter, as slide projectors use warm tungsten light, as do many viewers also, so they will look great anyway. When scanning, it is very easy to fix the blue cast caused by the cold light source used by scanners. I would even go so far as to say the lab SHOULD have color corrected the scans - every time they scan negatives, the results are automatically color corrected by default (no, negs do have better color balance, they are just always corrected, so the results you see look better).
    But the easiest thing to do now is simply correct them yourself in PS or PS Elements. Try the "autocorrect" button in Elements, or try Image>Remove Color Cast, then clic on a white, black, or gray part of the image. Presto, it will remove the blue cast from the image. Hope that helps.
     
  10. i tried the remove color cast idea in photoshop. it seemed to help a bit. then i tried the auto color correction and that seemed to really help quite a lot. that is definitely a menue that i dont use very much because i have been trying to figure out how to do every thing with adjustment layers instead of using the auto settings. i do appreciate all the help, i was about ready to put away that particular film so i am glad there is a solution.
    00ROhu-85653584.jpg
     
  11. sorry i attached the wrong one.
    00ROiT-85659784.jpg
     
  12. It is an example of poor quality scanning.
     
  13. This is a known issue with Kodachrome- it can be very difficult to scan for a few reasons.

    Do a search for Kodachrome and "Cyan cast" or "cyan dye" and you'll find a lot more info about this.
    I wouldn't expect labs on autopilot to do a good job. If you must give them slide film, try something like Fuji Sensia or Kodak Ektachrome.
     
  14. sensia or ektachrome.... what about velvia or provia out of curiosity. at the moment i only have so much film in my fridge and those are the ones that i currently have in the camera.
    i have just started playing with slides so i am fairly new to all this.
    thank you again for all your help.
    jason
     
  15. Hi Jason,
    This is a well known artifact from scanning that occurs with Kodachrome. The best solution is to obtain a Kodak Q60 Kodachrome color calibration slide (available from B&H), and calibrate your own scanner with it. I realize that you had Ritz scan it. Just out of curiosity, I loaded your image into Capture NX, and applied my Kodachrome color profile which I generated for my Nikon Coolscan 5000. Because of the differences between scanners, the image then seemed a bit too magenta. Use of the white, black and gray eyedroppers (I used the sand for the gray tone) gave me this image, which I then converted back into your original Adobe RGB format.
    Is this anything like the original slide?
    At any rate, using Nikon Scan in RAW mode, or Vuescan, it is easy enough to encode the correct color profile so that the resulting scans match the original slide very closely in color and contrast, automatically.
    00ROrv-85729584.jpg
     
  16. Whoops! I just realized that although my Safari browser can view Adobe RGB encoded images properly, for most other people, the images are going to look funny. Here's the image again in the usual sRGB web profile.
    00ROs1-85729684.jpg
     
  17. Absolutely standard problem scanning any Kodachrome. The dyes used don't "read" well with the tri-color readout of a scanner. Our eyes are much more continuous tone, and can deal with those dyes, but the scanner sees them wrong.
     
  18. you got bad scans. Do not try to correct their mistakes: have them rescan them. Kodachrome is not difficult to scan unless an idiot is doing the scanning.
     
  19. Robert,

    You have some bizarre purple shades in the sky....

    In my experience it is not so difficult to adjust Kodachrome landscapes (add a lot of yellow and remove some blue)
    but to get any flesh tones like the original you will have to alter the hue of the reds perhaps reducing the saturation
    overall too. Also if the slide is composed of very subtle colors (e.g. cloudscapes) that you just have to
    have right - then it can be the very devil. It is not, perhaps, rocket science to scan Kodachrome, but it does take
    some degree of experience and I am not sure I would expect your average scanning service to have this expertise,
    especially given how often they see Kodachromes (rarely). In my experience most professional scans of
    Kodachrome ultimately result in "close, but no cigar".

    It is also almost inevitable too that you will get problems with noisy shadows and/or blown highlights.

    For all these reasons Kodachrome is not an ideal choice for scanning - fantastic for projection, poor for
    digital printing.
     
  20. "sensia or ektachrome.... what about velvia or provia out of curiosity"

    You'll get better scans with less contrasty transparency films unless you do the scans yourself (and know what you're doing) or use a high-end scanner. I would not recommend Velvia or Provia if your purpose is to scan versus project. You can always increase contrast and saturation with a film like Astia/Sensia/Elitechrome100, etc.

    The newer Nikon scanners have a software Kodachrome setting which seem to yield better scans than the older scanners- again, there have been very long threads about cyan casts with Kodachrome over the years on the comp.periphs.scanners newsgroup http://groups.google.com/group/comp.periphs.scanners/search?q=kodachrome+cyan&start=0&

    Anyone who thinks it is easy hasn't used a scanner and scanner software that wasn't made for the task.
     
  21. Hi Robin,<p>
    Yes, I acknowledge that there are wierd magenta casts in my "corrected" scan. But, that's the problem with applying a color correction
    profile meant for RAW scans from my Nikon Coolscan 5000 to the scans from the Ritz machine which already have an Adobe RGB profile
    applied. However, my intent was to show the
    concept of a corrective color profile in application.<p>Ultimately, if you intend to scan Kodachrome, you need to do it yourself with your
    own
    equipment.
     
  22. It's weird...I have shot plenty of Kodak slide film(Kodachrome,Ektachrome and Elite Chrome) and I have never had any slides come back with a blue tint.
     
  23. I recently posted a topic regarding Kodachrome scans that I had done at Dwayne's. Same problem.
    It seems to always happens in lower light.
     
  24. Patrick, it's not the slides that look blue. The problem is that a straight scan of Kodachrome looks little like the original (comes out blue), whereas a straight scan of an Ektachrome will be almost correct.
     
  25. The bluish cast is a characteristic of the film itself. If you look at the H&D curves for Kodachrome, you can see the differences in contrast among R, G, and B. Unless special compensation is made for the cast, ICC profiles (even made with the Kodachrome target) will only help assure that the blue cast is accurately preserved in the scan. The blue bias is there to compensate for the effects of viewing slides in a dark room, projected by the yellowish light from a tungsten bulb. Under those viewing conditions the eye does not fully adapt to the yellow projector light. One way to correct the blue bias in scans is to increase the contrast of the blue channel, while decreasing the contrast of the red channel. Try setting the middle slider in Levels to 1.15 for Red, 0.93 for Blue for a ballpark correction.
    00RTRc-88015684.jpg
     

Share This Page

1111