Skip to main content

⚠️ Please note that this topic or post has been archived. The information contained here may no longer be accurate or up-to-date. ⚠️

Black Tux's have A Blue Tint

Paul Spatafora
I have been struggling with black suits that have a blue tint to them for years. I've had this problem with all of the LR's, Bibble, Silkypix, and now CO 7. If I get the suits black in CO, the faces are too orange and they don't look good. If I add blue to the image, I get a more pleasing skin tone, but the tux's and black suits start to turn blue. It's been a problem with all my camera files and now with my D800. To get rid of it, I have to go into my photo editor Photoline and use the Color Balance tool and adjust the Shadows by adding yellow to get rid of the blue. With LR you could desaturate the magenta and purple to get rid of it, but sometimes it really screws up the files.

I sometimes find that the skin tones in CO are a bit too orange, too blue in LR and green in Silkypix. But I will say CO does have the best skin tones and they are easily adjustable.

Is there a technique in CO to get rid of the blue in the suits and less orange skin tones in the raw converter? It would be great to make the adjustments in CO instead of Photoline.

Thanks in advance.

Paul
0

Comments

19 comments

Date Votes
  • SFA
    If you don't have a fully colour managed workflow things might be a bit hit and miss but have a look at the color editor or, thinking of your tux problem in particular, you might find that popping on a few colour readout pins and playing with a curve on the blue channel get you close to where you want to be. A small adjustment at the dark end should probably be enough. If you need a large adjustment then there might be other issues to be identified.

    I'm sure there are many other approaches too.

    Grant Perkins
    0
  • Christian Gruner
    A hardware monitor calibration would seem needed here. Especially when it happens across many applications.
    0
  • Paul Spatafora
    Thanks for the replies. I tried the curves idea and it seems to work ok and I can copy and apply that technique to other images. The presets are a great find. Silkypix has similar features called film tones

    As for calibrating my monitor, I use a Datacolor Spyder Elite to periodically calibration my 30" Dell wide gamut screen. This issue of the blue suit goes back to 2004 with the Kodak SLRn, Fuji S5, D700 and now with the D800. I have spent countless hours calibrating my monitor in the hopes of eliminating this issue. This happens under certain lighting conditions. I use Auto WB in camera and I've noticed that CO 7, Silkypix, and LR4 all read the Auto WB differently. As I stated earlier, LR favors bluish skin tones, Silkypix greenish and CO orange. I have opened the same file side by side and have seen the difference and have had my wife confirm it just in case I was seeing something different. BTW, my lab complained about this phenomenon as well, and that's where I learned the Color Balance technique. I understand that WB is a starting point and subjective to each program.

    Under most circumstances, the suits are black when I open the file, until I try and remove the color cast I find unfavourable. In CO 7's case, by introducing blue to offset the orange bias (to make the skin tones more pinky) it introduces the blue suit. I can duplicate this with CO 7's presets too. I was hoping to see if the Skin tones feature in the WB tab would help with this, but as reported, it's currently broken in 7.1.4. In Silkypix, they use simulated film tones to try and correct the greenish tone and in most cases it works. LR automatically intros the blue suit right off the bat and you have to use the magenta and purple sliders to desaturate the blue.

    So why am I going to great lengths to describe this issue? Simple, it's been an ongoing matter for almost 10 years and I want to see if there is a solution out there other than calibrate your monitor.

    I see skin tones as pinky and not greenish, orangey or bluish and my quest is to achieve this with CO 7.

    Any help in this matter would be greatly appreciated.

    Paul
    0
  • Christian Gruner
    [quote="NNN635018233259472077" wrote:
    Thanks for the replies. I tried the curves idea and it seems to work ok and I can copy and apply that technique to other images. The presets are a great find. Silkypix has similar features called film tones

    As for calibrating my monitor, I use a Datacolor Spyder Elite to periodically calibration my 30" Dell wide gamut screen. This issue of the blue suit goes back to 2004 with the Kodak SLRn, Fuji S5, D700 and now with the D800. I have spent countless hours calibrating my monitor in the hopes of eliminating this issue. This happens under certain lighting conditions. I use Auto WB in camera and I've noticed that CO 7, Silkypix, and LR4 all read the Auto WB differently. As I stated earlier, LR favors bluish skin tones, Silkypix greenish and CO orange. I have opened the same file side by side and have seen the difference and have had my wife confirm it just in case I was seeing something different. BTW, my lab complained about this phenomenon as well, and that's where I learned the Color Balance technique. I understand that WB is a starting point and subjective to each program.

    Under most circumstances, the suits are black when I open the file, until I try and remove the color cast I find unfavourable. In CO 7's case, by introducing blue to offset the orange bias (to make the skin tones more pinky) it introduces the blue suit. I can duplicate this with CO 7's presets too. I was hoping to see if the Skin tones feature in the WB tab would help with this, but as reported, it's currently broken in 7.1.4. In Silkypix, they use simulated film tones to try and correct the greenish tone and in most cases it works. LR automatically intros the blue suit right off the bat and you have to use the magenta and purple sliders to desaturate the blue.

    So why am I going to great lengths to describe this issue? Simple, it's been an ongoing matter for almost 10 years and I want to see if there is a solution out there other than calibrate your monitor.

    I see skin tones as pinky and not greenish, orangey or bluish and my quest is to achieve this with CO 7.

    Any help in this matter would be greatly appreciated.

    Paul

    Personally I have bad experiences with that calibrator, I would suggest something like this: http://www.xrite.com/i1display-pro

    If that doesn't help, I would suggest you start a support case, and we will try to help you.
    0
  • Paul Spatafora
    Okay, I'll send you a file for your review and see what magic can happen.

    I have tried an Eye One and my lab uses one too with the same results.

    Thanks for the help.

    Paul
    0
  • SFA
    [quote="NNN635018233259472077" wrote:
    Thanks for the replies. I tried the curves idea and it seems to work ok and I can copy and apply that technique to other images. The presets are a great find. Silkypix has similar features called film tones

    ......



    There are a number of way to approach this and of course you can combine them and then save your own "styles" for future use.

    However I'm interested to know if your on screen colours always match your lab printed colours (making allowance for the different medium and so on) based on the final adjustment as seen on screen.

    There seem to be so many lookup tables involved in the entire process as the file is shifted from one media system to another that the potential for something going awry seems large.

    I keep a small A4 printer at home and C1 seems to do an excellent job with it for my home office purposes without too much effort whereas by other photo editing application can also produce excellent results but not as easily and consistently. Most notably the printer driver and the application seem to constantly confuse each other about orientation, scale and, in fact most other things with the other application.

    I'm not usually attempting Fine Art quality levels with it nor do I need absolute Pantone accurate colours so it is good enough for its purpose without demanding hours of support time and the challenges of trying to get a profile for a Notebook screen (albeit a quality notebook screen) that is always going to be subject to less than stellar native colour balance before we start. Since much of my output is of outdoor events with many variations in light from minute to minute 100% colour accuracy is not too much of an issue most of the time and one has to be practical about how far to take the desire for it! It sounds like some of your regular shooting subjects can be shot in a more controlled situation so it makes sense to find an easy to apply way to deal with them.

    Grant Perkins
    0
  • Christian Gruner
    [quote="NNN635018233259472077" wrote:
    Okay, I'll send you a file for your review and see what magic can happen.

    I have tried an Eye One and my lab uses one too with the same results.

    Thanks for the help.

    Paul


    The Eye One is an older generation. Lots has happened since then 😊
    0
  • Paul Spatafora
    I was printing some images in Image Print using there built in profiles and I noticed that some of the magenta blue color cast was eliminated. I was speaking to a friend and he wanted to know what color space I was using. Duh, I don't know! Does the color space I'm editing have a meaningful implication to how the image looks like?

    What color space does CO 7 work in? Can I adjust the color space? I have it set to Adobe 1998, but I was wondering if I'm editing in Adobe 1998 or another color space. Any ideas would be great.

    Christian, I have to correct myself. It's not always a bluish cast. Most of the time it's a magenta cast in the black suits. I have a friend that has an I1 2 and he's going to come over in the next couple of weeks and calibrate the monitor, but heal\ready told me that it ain't going to help as the Spyder is a good tool.

    Any help to resolve this mystery would be greatly appreciated.

    Thanks in advance.

    Paul
    0
  • Christian Gruner
    [quote="NNN635018233259472077" wrote:
    I was printing some images in Image Print using there built in profiles and I noticed that some of the magenta blue color cast was eliminated. I was speaking to a friend and he wanted to know what color space I was using. Duh, I don't know! Does the color space I'm editing have a meaningful implication to how the image looks like?

    What color space does CO 7 work in? Can I adjust the color space? I have it set to Adobe 1998, but I was wondering if I'm editing in Adobe 1998 or another color space. Any ideas would be great.

    Christian, I have to correct myself. It's not always a bluish cast. Most of the time it's a magenta cast in the black suits. I have a friend that has an I1 2 and he's going to come over in the next couple of weeks and calibrate the monitor, but heal\ready told me that it ain't going to help as the Spyder is a good tool.

    Any help to resolve this mystery would be greatly appreciated.

    Thanks in advance.

    Paul


    If you have to trust your print output, you have to use a fully managed and calibrated workflow.

    That means that you have to work with ICC profile all the way through from editing on screen to print. An ICC can significantly impact both colors and contrast.

    Regarding CO's working profile, this is the one you choose in your Output recipe.
    Then when you print, you have to either make or find one that exactly matches the type of ink, paper and printer. (I.e. Epson has a lot of profiles for their printers, so does many other manufacturers).
    0
  • Paul Spatafora
    Hi Christian,

    I do have a managed workflow. Here's what I have. Let me know if I missed something in the process.

    1. 30" Dell Ultra wide gamut monitor calibrated with a Spyder 3 Elite puck and software version 4.02. I've been told by Datacolor that the software will calibrate the wide gamut monitor and it does.

    2. I set CO to Adobe 1998 as my output color space. I can soft proof in CO 7 and Photoline and I can confirm that the soft proof in Photoline looks identical to CO 7 soft proof. ( This is what started the whole thread for me)

    3. I output my files using CO 7 and adobe 1998 ICC.

    4. I edit in adobe 1998 in Photoline and soft proof with Image Print ICC profiles when I need to print.

    5. I always print in Image Print 9.0 using there supplied profiles and they have never let me down. Best investment I made.

    I can edit with the soft proof in CO so my results are predictable when printing. My problem is when I have to supply the files to the customer and I can only assign a standard color space and I've picked Adobe 1998 because it is easily recognized by most software and has a wider gamut.

    As this is a 10 year mystery as to why I get blue, purple, magenta black suits, I think this mystery may go on a little longer.

    What color space is CO 7 developed at for optimal performance? I will setup a support case and provide you with a file, and an Image Print ICC profile for a paper that I use to illustrate the issue.

    Paul
    0
  • sizzlingbadger
    [quote="Christian Gr" wrote:

    Regarding CO's working profile, this is the one you choose in your Output recipe.
    Then when you print, you have to either make or find one that exactly matches the type of ink, paper and printer. (I.e. Epson has a lot of profiles for their printers, so does many other manufacturers).


    This doesn't make sense, what if you have multiple recipes selected ?

    Surely the editing space is something huge like ProPhoto ?
    0
  • Photocor
    Dear,
    I think your problem cmes from Spyder.
    Spyder 3 and 4 are made with gelatin filters that wear over time. It is common to replace the sensor every three years.
    This is one reason why the type of solution Xrite soultion pro 2 is better (glasses and spectrophotometer).
    For your screen I would suggest the following parameters: Brightness: 80 to 100cd, and color 5500K (D55) .
    When you export your photos to print I suggest you use the workspace eciRGB v2 ICC v4 if you are in Europe: it is quite larger than Adobe and close with printing profile.
    The second problem could come from your subcontractor (modification in their process without information to their customers). The result is clearly better when you can print by yourself with a personnal printing profile.

    Regards.
    0
  • Paul Spatafora
    [quote="Photocor" wrote:
    Dear,
    I think your problem cmes from Spyder.
    Spyder 3 and 4 are made with gelatin filters that wear over time. It is common to replace the sensor every three years.
    This is one reason why the type of solution Xrite soultion pro 2 is better (glasses and spectrophotometer).
    For your screen I would suggest the following parameters: Brightness: 80 to 100cd, and color 5500K (D55) .
    When you export your photos to print I suggest you use the workspace eciRGB v2 ICC v4 if you are in Europe: it is quite larger than Adobe and close with printing profile.
    The second problem could come from your subcontractor (modification in their process without information to their customers). The result is clearly better when you can print by yourself with a personnal printing profile.

    Regards.

    I have my screen set to 110 and at 6500K at 2.2 gamma. I'm not sure what the above setting will do for me, but I'll give it a try when I calibrate it again.

    The Image Print profiles work perfectly with my Epson 3880 printer and don't believe its a problem. I believe it have to do with how the software determines the proper WB. I find CO has the proper Blacks but sets the skin tones to a yellow / Orange tone. I have to adjust the skin tones by setting the WB to the blue side and this is where the issue starts.

    Is there anyway the skin tones can be tuned little bluer without affecting the blacks?

    Paul
    0
  • Photocor
    Dear,
    1) Somethings are not clear for me: when do you meet your problem: display or print? Magenta appears when you do the softproofing, or always ?
    2) Remember that when you are at 6500k,you are working your photos at tis value, realy cold and blue. It is offset from a daylight observation==> You move all your values ​​by amplifying yellow to compensate it.I don't think that your issue comes from the WB. But to be sure you can check it with a grey chart.
    3) In Capture One, when you work on your photos, what is the camera ICC profile used, and which curves?
    4) When you print on your 3880, is it with self made profile or the manufacturer profile?

    Regards
    0
  • nggalai
    Long shot, but have you tried an IR filter on your lenses in a situation where you expect this behaviour to happen?

    I don’t know the cameras in question, but I do remember that the D100’s sensor’s red channel was too IR sensitive, producing purple /blue / magenta “blacks†under certain lighting conditions. The same applied e.g. to the Leica M8.

    Best,
    -Sascha
    0
  • Paul Spatafora
    I used to use IR filters when I was shooting with the Kodak SLRn. It helped but it didn't resolved the problem entirely. That was one of the reasons I didn't get an D800E for fear of this issue.

    I opened a support case today to see if the problem can be resolved. I forwarded a processed LR4 file of the raw file I sent support that was processed at the default settings and the issue was not present. Every color except for black is perfect with CO.

    Thanks for the input.

    Paul
    0
  • Paul Spatafora
    The people at CO are fantastic about trying to resolve issues. After only one day, they were able to help me out with the phantom blue suit and orange skin issues. The solution for those that might encounter this issue is to use the "No color correction" ICC located in the Base Characteristic panel. It's found under the Effects pull out menu.

    With one click, the matter disappeared. You still have to play with the settings such as film curves to get your desired effect. The skin tones are now neutral and the orange cast has disappeared. This is for the D800 but the technique may work for you. It's been noted that the D800 does have a green yellow bias native to the RAW files. This has been a complaint by some when compared to the D700 that had a more bluish tone. Therefore, it would seem that CO tried to maintain a faithful representation of the RAW files intended look.

    Hope this helps.

    Paul
    0
  • sankos
    I've encountered a similar issue with the files from my Sony RX100 -- too much magenta (apparent in daylight skin tones) and bluish magenta blacks for a night/candlelight photography. I had to counter this with the WB sliders (favouring the greens to get rid of the magenta) for every single picture, so it started being a pain and something I wasn't used to when dealing with the beautifully rendered colours of the Pentax K10D.

    Thanks for the hint with "no colour correction" icc as I wouldn't have thought of trying it. It seems to work OK for the night photos I've tried it on, though you need to work with contrast and saturation afterwards. So it looks like a case of bad profiling for some individual camera models, or am I wrong? My 60 day trial is almost over and I'm not sure if I want to upgrade or not at this point.
    0
  • Paul Spatafora
    Hi Sanko,

    Yes, that technique is awesome. I suffered this issue in LR4 also, so I went to Silkypix and that solved the problem but the skin tones were a bit greenish but Silkypix did have some film tones that removed it.

    CO is the best raw converter of all the ones I have used, and I've used them all. The colors, details, noise controls and output is great. I couldn't believe that CO doesn't have a red eye reduction tool and a real brush tool. The spotting tool is useless too. But, strictly speaking as a raw converter, workflow, and custom ability it's the best one I've used in the last ten years. Even with these shortcomings, its still a great tool to have.

    Hope this helps.

    Paul
    0

Post is closed for comments.