Thread: X-rite color checker for ursa mini pro

Page 3 of 4 FirstFirst 1234 LastLast
Results 21 to 30 of 34
  1. #21  
    Senior Member
    Join Date
    May 2012
    Posts
    1,127
    Example of CIE XYZ to REC709 gamm 2.4 conversion with basic one light correction.
    Reply With Quote  
     

  2. #22  
    Senior Member
    Join Date
    Feb 2013
    Posts
    158
    Quote Originally Posted by razz16mm View Post
    Example of CIE XYZ to REC709 gamm 2.4 conversion with basic one light correction.
    Do you have a before/after shot?
    Reply With Quote  
     

  3. #23  
    Senior Member shijan's Avatar
    Join Date
    Apr 2012
    Location
    Odesa, UA
    Posts
    439
    Quote Originally Posted by razz16mm View Post
    You can get a smaller version of the Oneshot from SMPTE for $99. https://www.smpte.org/store/product/...2-pocket-chart
    Resolve can correct to either the Macbeth style chart like the X-rite or the Oneshot. The X-rite is a more universal chart applicable to multiple color spaces. The Oneshot is specifically calibrated to hit BT709 standard vector scope targets like a video color bar test pattern when white balance, gamma, and exposure are correct.
    In Davinci YRGB color managed workflow, your source color space and gamma should match the camera and your target color space and gamma should match your display or delivery format for a mathematically correct conversion. That is independent of any raw or downstream color corrections. You would match the same I/O color spaces on the chart correction tool too for accurate results
    For example my D16 CDNG raw uses CIE XYZ linear input color space. I normally work in REC709 2.4 gamma for broadcast output color space, or SRGB for web video graded on a computer monitor. I find this preserves the vibrant color depth the D16 is capable of without chroma clipping in REC709.
    But Resolve workflow is usually more complicated. There is timeline color space between camera and your target color spaces. For example Source clip in RAW or ProRes uses BMDfilm gamma and gamut as input color space. Than it transformed to timeline color space which can be BMDfilm or rec709 or Cineon or any other desired color space and gamma encoded. All main color corerction done in this color space. And in the end you add final color space transform node to convert it to Rec709 or any other desired output. This workflow allows you to manipulate image in wide gamut and log gamma and same time view result in realtime in Rec709 or any other desired color space.

    BTW there is every year Black Friday sale on B&H and X-rite ColorChecker can be purchased at half price.
    Reply With Quote  
     

  4. #24  
    Senior Member
    Join Date
    May 2012
    Posts
    1,127
    Quote Originally Posted by shijan View Post
    But Resolve workflow is usually more complicated. There is timeline color space between camera and your target color spaces. For example Source clip in RAW or ProRes uses BMDfilm gamma and gamut as input color space. Than it transformed to timeline color space which can be BMDfilm or rec709 or Cineon or any other desired color space and gamma encoded. All main color corerction done in this color space. And in the end you add final color space transform node to convert it to Rec709 or any other desired output. This workflow allows you to manipulate image in wide gamut and log gamma and same time view result in realtime in Rec709 or any other desired color space.

    BTW there is every year Black Friday sale on B&H and X-rite ColorChecker can be purchased at half price.
    Timeline color space is what you see on your reference monitor screen and should match the color space your monitor is calibrated for. This allows visual color grading with accurate output conversion to other delivery color spaces But a mathematically correct color managed workflow is basically conversion from whatever camera color space to a display color space. I think you sacrifice too much grading a log timeline without correct conversion. In any case your grading color space is always your reference monitor color space. WSIWYG. If your workflow breaks that, then your output won't match the grade you see on your reference monitor.
    In my D16 example I have full access to the uncompressed wide gamut, HDR linear data inherent in the camera file. There is no limitation on the ability to manipulate the image. Scopes accurately reflect limits of the display color space, so I can push the limits and see any form of value or chroma clipping on the scopes. Internally is is still a 32 bit YRGB workflow. I have experimented with log-c timeline conversions, but the value and chroma compression are excessive and I can never recover the richness inherent in the linear data. I use highlight and shadow recovery and color tools in the DNG raw menu to bring values within the legal limits for REC709, working from wider values to lesser values. Log reverses this process, working from narrow value ranges to wider value ranges. but something is always lost with this process.
    Last edited by razz16mm; 06-25-2017 at 09:41 AM.
    Reply With Quote  
     

  5. #25  
    Senior Member
    Join Date
    May 2012
    Posts
    1,127
    Quote Originally Posted by polaroid22 View Post
    Do you have a before/after shot?
    Unvonverted XYZ linear frame



    REC709 2.4 gamma conversion ungraded
    Reply With Quote  
     

  6. #26  
    Senior Member shijan's Avatar
    Join Date
    Apr 2012
    Location
    Odesa, UA
    Posts
    439
    Quote Originally Posted by razz16mm View Post
    Timeline color space is what you see on your reference monitor screen and should match the color space your monitor is calibrated for. This allows visual color grading with accurate output conversion to other delivery color spaces But a mathematically correct color managed workflow is basically conversion from whatever camera color space to a display color space. I think you sacrifice too much grading a log timeline without correct conversion. In any case your grading color space is always your reference monitor color space. WSIWYG. If your workflow breaks that, then your output won't match the grade you see on your reference monitor.
    In my D16 example I have full access to the uncompressed wide gamut, HDR linear data inherent in the camera file. There is no limitation on the ability to manipulate the image. Scopes accurately reflect limits of the display color space, so I can push the limits and see any form of value or chroma clipping on the scopes. Internally is is still a 32 bit YRGB workflow. I have experimented with log-c timeline conversions, but the value and chroma compression are excessive and I can never recover the richness inherent in the linear data. I use highlight and shadow recovery and color tools in the DNG raw menu to bring values within the legal limits for REC709, working from wider values to lesser values. Log reverses this process, working from narrow value ranges to wider value ranges. but something is always lost with this process.
    I don't familiar with D16 workflow and files but you are free to set Timeline to Rec709 or monitor color space as well. This may simplify some things because you don't need to do secondary color space transform but same time this limit you with poor Rec709 gamut and color saturation will clip very fast when you do a color correction (Overall colors are looking not too crisp and natural, reds shifts to orange very fast). Also buildin LOG to REC conversion in Resolve is very poor and produces unnatural hard clipping results as well, so it is better to generate 1D transformation lut in LUTCalc online lut generator for this (Cineon to Amira709 for example)
    Why edit image in limited color space when you have wide gamut source from camera?
    Reply With Quote  
     

  7. #27  
    Senior Member
    Join Date
    May 2012
    Posts
    1,127
    Quote Originally Posted by shijan View Post
    I don't familiar with D16 workflow and files but you are free to set Timeline to Rec709 or monitor color space as well. This may simplify some things because you don't need to do secondary color space transform but same time this limit you with poor Rec709 gamut and color saturation will clip very fast when you do a color correction (Overall colors are looking not too crisp and natural, reds shifts to orange very fast). Also buildin LOG to REC conversion in Resolve is very poor and produces unnatural hard clipping results as well, so it is better to generate 1D transformation lut in LUTCalc online lut generator for this (Cineon to Amira709 for example)
    Why edit image in limited color space when you have wide gamut source from camera?
    If you are delivering to an HD video codec in any form you are by definition limited to REC709 color space. Fpr non-raw cameras, log curves compress sensor values to fit within REC709 color space without clipping. You can't escape REC709. It is what you see on your monitor or TV unless they are specifically capable and calibrated for some other color space. For linear raw footage, you have full access to the data to control how it is expressed in REC709 without log compression. Clipping is only an issue with incorrect conversions. But you have to limit DR and gamma to REC709 one way or another if that is what you are delivering. The color gamut of my D16 far exceeds DCI-P3 and is close to REC2020. But there is no point in grading for a color space you can't see on your monitor or that can't be reproduced by your delivery format.
    Reply With Quote  
     

  8. #28  
    Senior Member shijan's Avatar
    Join Date
    Apr 2012
    Location
    Odesa, UA
    Posts
    439
    Just few more tests in Resolve 14beta to explain the situation:

    1. Timeline set to Panasonic V-Gamut gamut/Cineon gamma. Contrast 1.2. Color Match tool applied. Cineon to Amira709 1D LUT and Color Space Transform Node V-Gamut to Rec709 applied in the end. (All dynamic range is visible, colors are natural)
    P.S. I use Panasonic V-Gamut because it is the only one color space in addition to Rec2020 available in Color Match tool list. Overall i find that Sony S-Gamut3.Cine is very nice wide gamut working space because it is primaries triangle designed proportional to Rec709 and so it transforms more even to Rec709 and produces less color shifts.



    2. Timeline set to Rec709 gamut/Cineon gamma. Color Match tool applied. Contrast 1.2. Cineon to Amira709 1D LUT applied in the end. (All dynamic range is visible, but some colors are very bright slightly oversaturated) The colors limits and saturation clipping will be way more visible if you do additional color correction or apply some kind of film emulation lut.



    3. Timeline set to Rec709 gamut/Rec709 gamma. Color Match tool applied. No contrast added, No LUT applied in the end. (Poor dynamic range, colors are oversaturated and destroyed) Sure you can add highlights recovery and it will back clipped areas but it will never look same natural and filmic as samples encoded with Cineon to Amira709 1DLUT.
    Last edited by shijan; 06-25-2017 at 10:34 PM.
    Reply With Quote  
     

  9. #29  
    Senior Member
    Join Date
    Feb 2013
    Posts
    158
    Can you use the x-rite colorchecker to change the colors to the Alexa colors? (without owning/renting an alexa?)
    Reply With Quote  
     

  10. #30  
    Senior Member shijan's Avatar
    Join Date
    Apr 2012
    Location
    Odesa, UA
    Posts
    439
    ColorChecker is ColorChecker it can only match different cameras to look similar. But i experimented with Arri color tool and with the luts generated in that application. It is only my own current conclusion and i may be wrong, but seems that Arri looks are not suited well for non Arri footage. It seems that Arri looks itself have some kind of additional background matrix color correction that puts in place colors from Arri cameras (Actually this is what Blackmagic should do for their cameras instead burn in that correction to camera ProRes files)
    I find that when you set timeline to Arri WideGamut/LogC and apply Arri looks to Blackmagic footage it looks more less ok. But when you apply same Arri look to Blackmagic footage that was corrected with ColorChecker - reds are shifted into deep dark magenta and some other colors became very strange and overloaded.
    Reply With Quote  
     

Similar Threads

  1. Replies: 8
    Last Post: 08-31-2016, 09:46 AM
  2. Ursa mini 4.6k False color?
    By david evans in forum General Discussion
    Replies: 6
    Last Post: 06-11-2016, 10:53 PM
  3. Replies: 5
    Last Post: 04-05-2016, 12:24 AM
  4. FS: SmallHD DP7 & X-Rite Display Pro
    By Note Suwanchote in forum For Sale / Want to Buy
    Replies: 0
    Last Post: 02-18-2015, 01:35 PM
  5. Help using color checker in Davinci Resolve 11!
    By deezid in forum DaVinci Resolve
    Replies: 4
    Last Post: 08-12-2014, 03:04 PM
Bookmarks
Bookmarks
Posting Permissions
  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •