メインコンテンツへスキップ

14.4 Bright Christmas lights dull down after about 1 sec of making any adjustment

コメント

21件のコメント

  • Keith R

    Capture One initially loads the preview jpeg that's embedded in the RAW file, while it demosaics the Raw data.

    Then the preview is replaced by Capture One's default rendering - which you can change to your heart's content.

     

    0
  • merlin braun

    Thanks. I had assumed it was something like that. I cant get the lights to glow and turn yellow no matter what I've tried...  Any tutorials or suggestions on that? Although I'm not an expert on Capture one, I haven't had this problem before in years of using it so am in the dark on how to solve this. 

    0
  • LVS

    Did you try the Clarity slider ?

    0
  • merlin braun

    Yes I have. A small amount of difference... really like the jpeg preview much better so I "should' be able to get the Christmas lights to look like that; just not sure what process I need to do!  

    0
  • LVS

    How about the Color Editor tool ? Use the color picker to pick up one of the lights and tweak the sliders to see if it has any impact.

    0
  • merlin braun

    Great idea. Maybe makes a little difference... Will keep playing around.  (The Raw image looks similar when opened in other programs except Apple Photos and Affinity photo are more like the JPEG preview in Capture one)  

    0
  • Robert Farhi

    I would try to enlighten a little the shadows and blacks in the exposure tool.

    0
  • BeO

    It's also the sky is different.

    Base characteristics tool, ICC profile and curve could be interesting, and the curves tool.

    0
  • Eric

    Another thing is that C1 tends to over sharpen things, depending on camera/lens used. With my Fuji X-T4 and most lenses, this is often the case. This can affect colors when you have fine details, like in tiny lights or fireworks. If you turn off sharpening completely and then only selectively sharpen other areas with masks, you can often recover some seemingly lost colors in the details.

    0
  • BeO

    merlin, can you tell why the exposure adjustment is different in both screenshots, is this because you want to demonstrate what you see but cannot get a screenshot in the split second before it changes? If so, then your adjustment setting of the first screenshot (higher exposure) solves the problem, no?

    0
  • merlin braun

    I will try some suggestion tonight.  AS to BeO, to clarify. I took a quick screenshot when C1 first displays (or even if I make an adjustment, it displays the night bright Christmas lights (apparently the JPEG version) and then kicks in the RAW version in the 2nd screen capture.  I haven't been able to match the night brilliance of the jpeg version yet! (not even close)

    Eric.  Tried knocking off all sharpening. Great suggestion. Did help but maybe 1/2 the way there? 

    Robert: Doesnt seem to help. When I make any adjustment, I get a second of the lights looking great, and then it drops back to dull and unsaturated.  

     

    Be);  I looked at ICC and it shows nikon z6ii and "pro standard"  NO matter which drop down (landscape, portrait etc. the JPEG shows for a second and lights look great and then it changes back to dull christmas lights.) Same things with "curves".  

     

    0
  • BeO

    I still wonder why the exposure slider is -0.83 vs. -1.3, this is quite a coase adjustment, but if the exposure does not cause this then it doesn't matter.

    If Eric's idea with the sharpening was good, what about the structure slider in the clarity tool?

    If you like and have the possibility, you could also share the raw file.

    0
  • merlin braun

    The exposure doesnt seem to matter much. I'm surprised I had changed it; just messed around with so many settings.  I dont think I can upload RAW here, but remembered I have a dropbox account. I dont know if you can only view or download as I haven't paid for the upgrade.  Let me know if this works or other ideas on how to share this file so you can play around with it.   Here is the link:

    https://www.dropbox.com/s/qshogl4jjnu8int/DSC_6220.NEF?dl=0

    0
  • Pavel Hubík

    Merlin,

    I have downloaded your NEF file and looked at it in my Capture One Pro 22. I can see the effects you have described above. However, when the image is zoomed to 100% or more it appears that Christmas lights are nearly colorless (i.e.  R, G and B values are nearly the same). Playing with highlights, saturation, sharpness, color noise, white balance, etc. doesn't seem to change their color substantially.

    Yellowish hue appearing, for a moment, after import or some adjustments is, in my opinion, an artefact caused by averaging of neighbor pixels. Capture One uses a blurred (averaged) image to show its changes during adjustments - before full calculations are completed. Note that the lights are very small - they are typically formed by about 10 pixels - so their color is easily influenced by pixel averaging.

    This explanation is consistent with the fact that on my screen I can also see yellowish lights if the image is viewed at the smallest zoom (fit to viewer size) and in browser thumbnails.

    Regards,

    Pavel

     

    0
  • BeO

    hi merlin,

    I agree with Pavel, "Capture One uses a blurred (averaged) image to show its changes during adjustments - before full calculations are completed. "

    and you can see this very well when you zoom in to 1600% and adjust e.g. exposure.

    I exported the image full size jpg and in Firefox you see the similar behavior, yellow/orange when zoomed to fit, more desaturated when zoomed in. So, pretty much the same averaging effect when zoomed to fit.

    Btw., when opened in Nikon NX Studio and further reduced the window size in order to meet the 2MB limit for posting here, zoomed to fit is rather desaturared:

    One can draw the conclusion that NX Studio shows a better (more correct) preview than C1.

    Which is backed by the fact that when I export the image with C1 to a smaller jpg with the export recipe for web use (but 1920px, the resulution of my monitor), then FireFox shows the lights more desaturated than C1 preview.

    Interesting to note, also the sky is different in Firefox depending on the exported jpg size, it is notebably dulled when exporting to a smaller size. Don't know if the jpg algorithm in C1 has a problem,
    EDIT: or the preview/softproof!

    0
  • merlin braun

    Thanks guys. Not sure what this means in terms of me brightening them up... probably a lost cause!  Sure looks better when they have more glow in the first C1 preview.  I took more photos later in the evening and they look better with that regard.  Just couldn't figure out what c1 was doing with teh rendering; hadn't see that happen before. 

    0
  • LVS

    Do you have C1 preview set to match the current Export Recipe, and if so what is that recipe set to ?

    0
  • merlin braun

    This is what I have currently set to see if that was the problem: Im no expert so probably missed something somewhere.  What you're suggesting totally makes sense.   In Apple Aperture, it was so easy to see your proof based on your output but I havent figured that out in C1.  

    Proof     SRGB

    Lab Readout      off

    Export      SRGB

    ICC   SRGB   

    0
  • Eric

    @merlin, you could definitely take the (un)sharpening approach further if you don't mind the lights being a bit unsharp. Thanks for providing the file, I did a quick & dirty edit on it. As mentioned, I first:

    • turned off the overall sharpening
    • in the advanced color editor, sampled between lights where you can see some color
    • expanded the yellow-orange range to as much possible (got a bit of green as well, try to avoid)
    • selected "create masked layer from selection"
    • on that layer, turned down the Structure slider all the way (brings back a bunch more color)
    • bumped up Exposure and Saturation

    That should get you close to what the preview looked like, of course the lighs are now a bit soft but that's usually what I'm going for when I photograph Xmas lights. I try to find something in the foreground to focus on so that the lights become bigger and go out of focus.

    Before:

    After:

    Of course, this small view shows the pixel binning as well, so open these images in their own tab to see the difference better.

    0
  • merlin braun

    I did not know one could make a selection using the colour picker!  That is pretty handy!   I"ll try the same thing.  Now If I could only stop 21 from crashing. I think turning off hardware acceleration might be stopping it (though I've been using the same computer for 5 years and havent had a problem til about 6 months ago...) 

    0
  • SFA

    Merlin,

    If turning off Hardware acceleration solves the crashing problem there is a possibility that the drivers for the GPU have some issues with either the OS or the Application software. Or both. 

    As a Mac user you, as I understand things, very much at the mercy of what Apple choose to do about hardware support and the drivers for various components in the system. However, there may be some information on the internet about such problems for individual GPU products and OS releases which might lead you to a resolution for such problems. 

    As Windows user when such problems have appeared in the past it has been possible to seek on-line advice and, where the information suggests that the GPU driver may be the source of the problem, take steps to identify a better driver option (either revert to the previous driver or find another that is a later version that claims to resolve some problems) and change the system files.  My understanding is that is not a path that one can easily take with the Mac environment unless Apple have released something to help. However it might be that there is a solution that has not yet arrived at your system via an update.

     

    For the original question, you may also want to consider the effects of the Proofing tool and the influence on the results of using different output settings in the Output Recipes.  In particular, use a common recipe with one version of it set for a smaller size output (for example to fit into your available viewer dimensions) and one for full size. By alternating between the two recipes with proofing turned on you should be able to see the effect of any compression applied when the system reduces the number of pixels used to make the images fit into the smaller area. 

    Or, thinking the other way around, when your edits that have been applied for the image on the reduced number of pixels in the viewer are then seen in their full effect at the 100% scale with all pixels available.

    It may be that the optimal approach is to have slightly different edit values set for each requirement. For example a different variant for some images rather than just different values for things like Output Sharpening in the Output recipe.

    0

サインインしてコメントを残してください。