Blog

http://aviddssupport.avid.com/pdf/v10.2/AvidDS_HiRes_Workflow_Color_Guide.pdf

Discovered:

• Best quality level to use for grading is FilmScan1.
  • The next one down, High, is only for final product, as intermediate totranscode from.
  • The next level up, FilmScan2, is “overkill”.
• Tech Blog (many subjects)
  • http://techblog.cineform.com/?page_id=1288
• HDLink
  • I should make more use of this in my workflow.  It does batch-conversion etc. and people like its deinterlacing quality.
  • http://www.dvinfo.net/forum/cineform-software-showcase/97451-how-does-hd-link-de-interlace-50i-25p.html
    • Q: How does HDLink perform its de-interlacing?
    • A: It doesn’t drop fields, it uses both to extract the best vertical resolution.
      • No elaboration given.  Could it be some kind of smart selection between interpolation (single-field) for dynamic scenes and blending (of each field-pair) for static scenes.  It would be interesting to compare against my standard motion-compensated approach (using AviSynth>TDeint) and indeed the new one explained on the Sony Vegas forum by Nick Hope http://www.sonycreativesoftware.com/forums/ShowMessage.asp?MessageID=768368&Replies=12.
• Firstlight
  • Does more than I imagined!
    • Has translation/projection adjustments (e.g. frame, zoom, pan)
    • It has “looks” (via LUT’s) e.g. “Bleach Bypass”.  Can scroll through them (e.g. via up/down arrow keys)
    • Can be keyframed hence dynamically changing (e.g. a pan to follow a moving object or adjust level/hue as sun sinks).
    • Can copy & paste attributes between different clips (select-all>Paste to make all clips have the same style)
    • Can “branch” projects to provide a dropdown-selectable menu of alternative grades/looks.
    • Great suggestions for incorporation in workflow:
      • While one person is editing, another can be grading.
      • Tip: store FirstLight projectsin DropBox folders, then everyonbe sees same instant-updates (e.g. extra alternative styles).
  • http://community.avid.com/forums/t/98416.aspx
    • Possible there exists a problem when used with Avid Media Composer 5.5.2 (at least).
    • Conjecture: (if) First Light … is OpenGL accelerated when dealing with Cineform media (then) when it is handed to to other applications, the playback and processing of the LUT (metadata) is all CPU, thus causing a performance discrepency.
• Vimeo: Cineform Examples & Tutorials & Groups:
  • http://vimeo.com/canetoad/videos
    • David Newman’s Videos
  • http://vimeo.com/groups/cineform
    • Created and moderated by Jake Segraves od Cineform.
      • http://vimeo.com/13021420 = “CineForm Tutorial for San Diego’s 48 Hour Film Project”
      • http://vimeo.com/11886542 = “Introduction to FirstLight — Part 1″
      • http://vimeo.com/11887177 = “Introduction to FirstLight — Part 2″
      • http://vimeo.com/10024749 = “CineForm FirstLight color correction through Dropbox”
  • http://vimeo.com/channels/cineform
  • http://vimeo.com/27220198
    • Cineform Workflow in Post Production - Part 1

DaVinci Resolve Lite - free:

• http://www.sonycreativesoftware.com/forums/ShowMessage.asp?MessageID=758711
• http://www.blackmagic-design.com/press/pressdetails/?releaseID=11851
• http://www.blackmagic-design.com/products/davinciresolve
• http://www.blackmagic-design.com/support/detail/?sid=3948&pid=11735&os=mac&leg=0

The full version has 3D tracking “Power Windows”, so you don’t have to manually generate and track masks.  It is said by some to be runnable on linux.

Although Avid Media Composer (MC) itself has no native Secondary Color Correction, that functionality can be achieved via plug-ins such as Boris (e.g. BCC Colors & Blurs) or Spectra Mate.

• http://community.avid.com/forums/p/68172/381466.aspx
  • Douglas’s video shows how to do it (but use a browser other than Safari).

To obtain/maintain/increase/verify an intuitive feel and hence greater confidence for the appropriate use of gamma in camera settings, I did some thinking & research.  The basic idea is as follows:

• A typical consumer camcorder produces crisp images
• “Film Look” use of a professional camera may employ non-standard gamma settings in the camera settings.
• The straight results of this are images of “washed-out” appearance.  To obtain a pleasing result requires grading (levels & gamma, saturation, color curves etc.).  Example: http://reduser.net/forum/showpost.php?p=687243

Regarding the second, more professional approach:

• The immediate result is “scary” because it looks washed-out
• The goal is not to produce an immediately-pleasing image but to capture “as much information as possible” (an often-quoted phrase) from a scene, with the intention and indeed requirement for grading.  One has to see it “through the eyes of a grader”.  A naive person (e.g. a newbie or a client) will of course not immediately see it that way.
  • Example references to this:
    • http://www.dvxuser.com/V6/showthread.php?232389-The-Red-quot-LOOK-quot
      • <<The RED RAW look, the washed out, flat, low contrast, incredibly versatile form in which the footage originates … screams possibility in our faces. Low contrast can, to the DP, imply power … being precious with the RED footage, and trying hard to save every bit of detail we started with.>>
      • <<I can see how the washed out look can become something in and of itself, and have people like it, and others not.>>
• What does this mean?  In general, possibly:
  • The complete levels and color space of that scene, un-clipped (clipping destroys information).
  • Any subtle light/shade within shadows of the scene.
• Questionable aspects:
  • Grading takes time (bad for quick-turnaround jobs) and if written to intermediate files (e.g. prior to editing) then it can also eat disk space.
  • There is a trade-off between generality and specificness.
    • Capturing maximum information provides the grader with greatest freedom.
    • On the other hand if it is known in advance that crushed shadows are required, e.g. to obtain silhouettes / film noir effects, then it is a waste of effort / bits if not counter-productive to boost them in the camera.
  • The degree to which grading can be applied in practice depends on the levels and color space resolution of the camera.
    • Prosumer cameras such as Z1 or XDCAM-EX record to 8-bit levels resolution.   And then only a sub-part of that levels-space (typically 16..255 or 16..235, depending on camera and settings).
      • For cameras whose sensors work at greater resolution (and can output this information) there is the option to record to external devices at that greater resolution (e.g. 10 bits 4:2:2).
    • While it is possible to apply effects like levels, gamma or color-curves (S-curves) to “professional” washed-out imagery, beyond a certain degree, the image will appear ragged or flesh-tones will appear sunburn etc., as the gaps between successive values of the bit-space get stretched too big.  One can actually see the gaps (between striations) in a Waveform Monitor (applied to the result of grading).
      • In that case we have in fact lost information, defeating the original goal…
    • If the results of grading are pretty-much identical (or, from the previous point, possibly inferior) to what would have been obtained in-camera using a more standard setting, then what was the point?
• Reassessment:
  • Due to the trade-off issues, the real goal should be to record the maximum relevant information.   In other words, to be a little bit specialised.
    • This is the logic behind employing physical filters on a camera, such as grad filters (”sunglasses” e.g. for the upper - sky - part of the image).
    • Even on feature movie sets I have come across formal instructions for film cameras to be deliberately “pushed a stop or two”.  Committing at record-time to something that could, presumably, have been achieved equally-well in post, which itself can be done almost immediately based on HD footage recorded simultaneously from HD cameras attached to the main camera.  I have seen directors receive rushes and quick cuts from such cameras within seconds…
  • The degree of commitment/specialization may depend on the type or uncertainty of the scene and on the consequence of making a mistake.  Feature films are very planned and their shooting is very iterative.  On the other hand there can be one-offs such as special-effects or VIP moments.  At the other extreme may be live events where anything can  happen - subjects, lighting, over-bright/over-dark etc.
  • The missing factors in the “maximum information” principle are then:
    • Relevance - what kinds of information are relevant?
    • Resolution limitations.
      • If we only have 8 bits, then what is the practical limit of grading?
      • Conversely, if we need to maintain maximum latitude etc., when do we need more than 8 bits (in practice mostly 10 bits)?

Which modes of Cineform are appropriate under which circumstances:

• Paraphrased from [http://www.sonycreativesoftware.com/forums/ShowMessage.asp?MessageID=742985&Replies=3, as of 2010-12-23]
  • NeoScene and NeoHD uses can select 4:2:2 in any quality.
  • High quality is for finished material (e.g. to be output to DVD/BluRay) but if further grading is a possibility then use Filmscan 1
    • Filmscan 2 is overkill.
  • Neo4K and Neo3D users also get 4:4:4 and 4:4:4:4 support.
    • Requesting these modes with (other variants of the software?) will result in a watermark.
    • For extensive post, filmscan and 4:4:4 is probably a benefit.
    • 4:4:4:4 requires lots of power and raid arrays.
  • The uncompressed mode should generally be avoided - it is only intended for camera acquisition to save battery power.

Cineform is a codec for digital intermediates.  When used in appropriate ways, it offers visually lossless compression/decompression.  Some quality reduction does occur but only to a degree that is not important to (or even noticeable by) most people’s eyes, even after several rounds of compression/recompression.  In contrast, delivery formats such as DV, DVD, XDCAM-EX are suitable only for a single round of compression/decompression, and even then are lossy (lose quality) to a degree that many people can notice, especially when playback is paused at a single frame.

I worried about and noticed in practice an effect where if I was using CINE gammas on the XDCAM-EX and exposed for faces at 70% (by zebras) then the gamma rolloff would result in “pasty-face” appearance.  It does …and did…  The solution for good looking faces is one of the following:

• Under-expose in shoot, raise in post.
• proper-expose in shoot, use standard gamma (not cine gamma), be careful not to let the face hit the knee (?) e.g. set knee to 90% or 95%.
•  take a given gamma curve (or even a flat standard one) and tweak it using gamma level & black-stretch adjustments etc. until it fits the scene.

(more…)

Someone noticed that XDCAM-EX footage imported to FCP appeared different as compared to Avid (AMA import).  Addressed in an Avid forum thread started May 2010, referring to FCP 6.06 and Avid 4.02: http://community.avid.com/forums/p/84153/473798.aspx

• What AMA gives is, is _exactly_ what the camera has captured. What FCP shows you, is a remapped image, most often with a gamma shift. 
  • (For Avid AMA imports, Avid settings for RGB or 601 etc. make no difference - it’s always as-recorded).
  • … imports (to FCP) will look different (to expected), because FCP/QT “corrects” the gamma when bringing in footage (even if you would not want that).
• Most people seem to agree that FCP works in 0-235, not 0-255, not 16-235. And without the option to leave things untouched. So if you import something into FCP, there’s no getting it back to the original levels anymore.

I guess I’d better do some experiments with ramps & scopes etc…

The real image manipulation happens with the desaturation and level filters. Desaturation affects the richness of colors and level (via a un-intutive set of controls) will adjust the black point, white point and distribution of brightness across the whole image. The settings of these filters will depend on the scene and the camera you have.Set Desaturate to -50. This will boost the colors just a tad. Adjust to taste. Set the level controls as follows: input = 0 input tolerance = 100 gamma = 1 output = 50 ouput tolerance = 80 Ken Stone agrees the Levels filter is non intuitive:[http://www.kenstone.net/fcp_homepage/basic_image_control.html ] <<< The Levels filter is supposed to be a more advanced version of the Brightness/Contrast filter, offering separate controls over highlight, midtone and shadow areas. In addition this filter offers the choice of working the image in RGB mode or any one color channel independently, Red, Green or Blue.However this filter is poorly implemented and is clumsy at best. It has five slider controls; input, Input Tolerance are used to lighten the image. The Gamma slider controls the midtone areas and Output and Output Tolerance used to darken the image. What makes this filter so difficult to understand and use is the fact that the filter opens with the default settings of the Input Tolerance and Output Tolerances sliders set at 100. With these two settings at 100 neither the Input Slider nor Output sliders work. It is necessary to lower either the Input or Output Tolerance sliders then start adjusting the Input or Output sliders. The real problem is that the highlight and shadow areas have two sliders each for control. Photoshop has a Levels filter but it is a different animal. Levels in Photoshop also has a ‘Histogram’ which gives a graphic display of all the pixels in the image based on their brightness values. The ‘Histogram’ display is essential to setting levels but the Levels filter in FCP has no ‘Histogram’. To be honest the Levels filter in FCP does not work for me at all - I just don’t get it. The Proc Amp filter does basically the same thing and works well. If you want to play with this filter I suggest that you set either Input or Output to a setting of 20 then start lowering the Tolerance filter down from 100. As you lower the Tolerance settings more effect will be applied. If anyone can offer any insight into this filter I would love to hear about it. >>> 

Things To Remember [http://www.kenstone.net/fcp_homepage/video_levels_nattress.html]

• Still images are RGB in a 0 to 255 range
• Levels above 100% FCP, 100 IRE or code 235 are called “super white”
• Final Cut Pro automatically converts images to fit into the 0% FCP to 100% FCP range when it is set to “White”.
• Final Cut Pro automatically converts images to fit into the 0% FCP to 100% FCP range when it is set to “Super White”.
• Final Cut Pro only looks at the 0% FCP to 100% FCP range upon still image export, and maps this range to 0 to 255 RGB.
• Picture information can be lost on still image export if you have picture information in the 100% FCP to 110% FCP range.
• Final Cut Pro always applies a gamma correction of about 0.8 to imported still images. If necessary, this can be corrected by applying a gamma correction of 1.2 in Final Cut Pro or 0.8 in Photoshop.
• Don’t preempt Final Cut Pro’s conversion of still images by setting your blacks in Photoshop to 16 and whites to 235. Final Cut Pro expects you to use the full 0 to 255 RGB range.
• “White” and “super white” settings only effect the import of still images.
• “White” and “super white” settings do not effect video or video rendering.