Thread: Gen 4 discussion

Page 3 of 8 FirstFirst 12345 ... LastLast
Results 21 to 30 of 73
  1. #21  
    Senior Member Asyndeton's Avatar
    Join Date
    Feb 2016
    Location
    Portland, OR
    Posts
    292
    Quote Originally Posted by CaptainHook View Post
    Oh yes, I forgot the Pocket 4K Film was also added which will be the same as the Broadcast Film setting since they're both Gen 4. By those settings, do you mean you're also using CST to transform to Rec.709? Sorry this has gone so off topic, I just wanted to help shijan. Perhaps a mod will shift this discussion.
    Yeah, I used the CST to get it to Rec 709 and thought it looked pretty good, better than transforming the 4K Film v3 to Rec 709. I think I wound up playing with it just keeping it in LOG afterwards. Can’t wait to have the P4K in hand.
    Reply With Quote  
     

  2. #22  
    Senior Member shijan's Avatar
    Join Date
    Apr 2012
    Location
    Odesa, UA
    Posts
    803
    Quote Originally Posted by CaptainHook View Post
    You need to use a separate gamma in the timeline setting. You're telling the CST plugin that it's getting pocket 4K film (ISO400), but it's actually getting BMCC/Pocket film (ISO800) which is a different log curve. So set the gamut/space to Pocket 4K but gamma to original BMD Film so that it matches the RAW decode settings. Or, just manually set that in the CST plugin instead of using timeline settings and then leave timeline as the default.

    Also, I wouldn't be surprised to see those lights clipping in the small gamut of Rec.709. It's not a great gamut for grading, I prefer working in a wider gamut and pulling into 709 manually for output.
    I always do all grade in LOG timeline and Rec 709 Color Space Transform node only in the end, so it is not Rec 709 clippng. It clips in LOG even if i disable Rec 709 Color Space Transform node.
    Changing gamma to BMDfilm don't help. I still see clipped saturation.

    Wonder why color science in Resolve implemented in so overcomplicated way. Why don't make it simpler and more user friendly? It may be just an option in CST or separate color science OFX plugin. It is really strange that we need to match color science version with proper timeline color space to make it work correctly.

    Even more - all those different color manage options in project settings are kind of rudimental because they all the same. (ACES use input color profile presets but not too flexible because it is kind of separate color transform project integrated to Resolve. RCM just use some other kind of in-out presets and support all possible workflow options in Resolve). It seems current Color Space Transform node use simple matrix-like input and output color spaces. So if you want more complicated input color corrections you need to build LUTs or use Color Match tool.
    My guess that "Color Science" is bad term, too abstract. There are terms input color profile and output color profile.

    Color Space Transform node use simple matrix-like input and output color spaces and can not support more complicated LUT-based color corrections. And Color Science seems like just a more complicated LUT-like input color profile engine. It corrects colors and gamma in more complicated way, but that kind of correction is not universal and may be not too perfect in some conditions.
    All my custom made accessories for BMMCC/BMMSC now available here https://lavky.com/radioproektor/
    Reply With Quote  
     

  3. #23  
    Senior Member CaptainHook's Avatar
    Join Date
    Aug 2012
    Location
    Melbourne, Australia
    Posts
    2,148
    Quote Originally Posted by shijan View Post
    I always do all grade in LOG timeline and Rec 709 Color Space Transform node only in the end, so it is not Rec 709 clippng. It clips in LOG even if i disable Rec 709 Color Space Transform node.
    Changing gamma to BMDfilm don't help. I still see clipped saturation.
    There could still be strong saturated LEDs that clip the Gen 4 gamut, yes. That is expected. It should happen less than it did with Gen 3 though.

    Quote Originally Posted by shijan View Post
    It is really strange that we need to match color science version with proper timeline color space to make it work correctly.
    I don't work in the Resolve team, but you don't need to adjust the timeline settings at all as I mentioned. Just setup the CST plugin correctly. Which means telling it which gamut/gamma you are giving it as a source, and what you want to target it as.
    Blackmagic Design
    My BMD LUTs.

    **Any post by me prior to Aug 2014 was before i started working for Blackmagic**
    Reply With Quote  
     

  4. #24  
    Senior Member shijan's Avatar
    Join Date
    Apr 2012
    Location
    Odesa, UA
    Posts
    803
    So the main rule - Color Science version in DNG settings should always match Timeline color space? And also should match the camera it was designed for (UM4.6K -> Color Science v3 -> BMD 4.6K film timeline)?

    Maybe Color Science v4 will work better with Pocket4K DNGs? Maybe it was just not designed for other cameras?

    Usually input color profile should match precise. And probably it is bad idea to attach Color Science v4 designed for Pocket4K Sony sensor to DNG taken from completely different 4.6K or HD sensor.

    Color Science v4 clips with 4.6K footage as well. Clipping occurs mostly in red-orange colors.
    All my custom made accessories for BMMCC/BMMSC now available here https://lavky.com/radioproektor/
    Reply With Quote  
     

  5. #25  
    Senior Member CaptainHook's Avatar
    Join Date
    Aug 2012
    Location
    Melbourne, Australia
    Posts
    2,148
    Quote Originally Posted by shijan View Post
    So the main rule - Color Science version in DNG settings should always match Timeline color space?
    No, timeline settings can be left alone in general (unless you have a specific need for it). But you were using the CST plugin and had the input set to timeline. In that case, the timeline needs to reflect what the clips are being decoded as, or what was baked into them if there were ProRes. I don't recommend setting the CST plugin input to timeline unless you're very sure what each clip is outputting. If you have mixed clips on a timeline, better to leave timeline as default and set CST plugin per clip for the right input.

    Quote Originally Posted by shijan View Post
    Color Science v4 clips with 4.6K footage as well.
    You would likely find they clip with Gen 3 too, and likely would clip faster as mentioned. Gen 4 works as intended for all cameras and was designed that way.
    Blackmagic Design
    My BMD LUTs.

    **Any post by me prior to Aug 2014 was before i started working for Blackmagic**
    Reply With Quote  
     

  6. #26  
    Senior Member
    Join Date
    May 2012
    Location
    Germany
    Posts
    1,004
    Hook, noone I trust more about color, still using Hook's original BMCC LUTs consistently.
    Reply With Quote  
     

  7. #27  
    Senior Member
    Join Date
    Sep 2013
    Posts
    821
    Just installed the newest Beta download and I don't see v4 anywhere in the CST settings. Should it be underneath Blackmagic 4.6K Film to Rec 709 v3?
    Reply With Quote  
     

  8. #28  
    Senior Member
    Join Date
    Aug 2012
    Posts
    616
    Seems like the problem is defining what camera you are on and how it is interpreted. I am clueless and haven't tries any of this, just hoping there is ultimately going to be an easy way to match the various BM sensors as well as other stuff. It would be nice to be able to concentrate on grading rather than struggling with matching.
    Reply With Quote  
     

  9. #29  
    Senior Member LochnessDigital's Avatar
    Join Date
    Jul 2014
    Location
    Tucson, AZ
    Posts
    585
    Quote Originally Posted by AaronChicago View Post
    Just installed the newest Beta download and I don't see v4 anywhere in the CST settings. Should it be underneath Blackmagic 4.6K Film to Rec 709 v3?
    Camera settings:


    CST Settings:


    So, input color space on CST needs to be version 4 color, the only camera model that has that in the CST is the broadcast. The input gamma needs to be the specific camera's gamma. So if you're on Ursa Mini 4.6K or UMP, then set that to BMD 4.6K Film. If you're on BMCC, BMPCC, BMMCC, set it to BMD FIlm. If you're on BMPC or Ursa Mini 4K, set it to BMD 4K Film.

    You can also, if for some reason you're doing a Color Managed workflow, set the Timeline to the exact settings as shown above in the CST (check the box "use separate color space and gamma") and then set the CST to "use timeline". That will give identical results. Also, in the Timeline color space, you can select either the Broadcast color space or the Pocket 4K color space -- same results either way.
    Last edited by LochnessDigital; 08-09-2018 at 03:49 PM.
    Aaron Lochert
    Reply With Quote  
     

  10. #30  
    Senior Member
    Join Date
    Sep 2013
    Posts
    821
    Thanks @LochnessDigital
    Reply With Quote  
     

Similar Threads

  1. Daredevil Season 2 Discussion
    By innerspark in forum Off Topic
    Replies: 8
    Last Post: 03-28-2016, 04:19 PM
  2. Zeiss C/Y Discussion
    By maxsims in forum Lens Discussion
    Replies: 17
    Last Post: 08-26-2014, 12:29 PM
  3. 4K - Now what? Codecs discussion...
    By dmetz in forum Workflow
    Replies: 15
    Last Post: 02-22-2014, 05:28 PM
  4. BMCC as a discussion piece
    By scorsesefan in forum General Discussion
    Replies: 14
    Last Post: 08-31-2013, 03:38 AM
  5. Adaptors for EF discussion
    By CineMac in forum Lens Discussion
    Replies: 5
    Last Post: 08-18-2012, 04:13 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
  •