Bookmark and Share
Location : OVS > RED Camera Manual User Help Guide > RED One Theory of Operation > Redcode and Color
Implications of the REDCODE Codec on the color process and workflow compared to film.
Send a link

Redcode and Color

Now its up  to us, knowing that there is some filtering of the images, to decide if what comes out on the other end is true enough to the original to be useful for us as an approximation of film.  My opinion is that the RED One certainly approximates a super 16mm and comes pretty dang close to the quality of 35mm.  It is much better than other digital video cameras.  And because the images are stored RAW - we have far more real-world lattitude in post than we do with video and again nearly that which we get from film.

The RAW data recorded is independent of the RGB signal monitored from the monitoring path. ISO, white balance or other RGB color space adjustments made to the monitoring path are not burned into the recorded 12-bit RAW data.

OK what's good about this sentence it that it reaffirms some of the statements I've made so far about the usage of the camera.  And now you can read this sentence and understand every term as well as the film characteristic it translates:

  • RAW Data is recorded independent of the RGB signal to the monitors.
    • On film you'd have a negative that is the best quality you'll ever get.  But you use a video tap for on-set viewing.  That un-processed negative needs to be processed, then it becomes an inter-positive or inter-negative.  It still looks fishy but at least now the emulsion is set - it won't be ruined by light.  Then this inter print will be printed to positive for a Master Print.  From there there are a variety of workflows, but the important thing is that you still have 99.9% of the color information.  Usually you do a 1 lamp print for dailies and/or maybe some kind of telecine transfer.  Either way you end up with an uncompressed digital image at this point.  The color depth of the image is anywhere from 12 to 24 bits of color.  So the RED approximates the low end of the film world as far as color lattitude.
  • ISO, white balance or other RGB color space adjustments made to the monitors are not burned into the recorded 12-bit RAW data. 
    • Yipppeee!  We can do whatever we want on-set, the camera does all the work to save the information as meta data that accompanies the RAW data.  If we like the look - use it.  If we want to start over and adjust it all manually in post - the information is there and there are some great tools for doing this.  This is actually better than the film world because we have not had to have any processing or print from film, yet we still arrive at the same approximation - 12 bit RAW.
 The camera‚Äôs monitoring path converts 12-bit RAW sensor data white balanced 10-bit 1280 x
720 pixel RGB 4:4:4 video. This signal may be modified using ISO, white balance or other RGB color space adjustments and provides monitor feeds for the RED-EVF, RED-LCD, Preview HD-SDI and HDMI outputs.

Would you believe that there's another way to describe the color information?  This sentence describes the colorspace that goes to the monitors.  This includes the EVF, LCD, HD-SDI and HDMI.  The camera ,"converts 12-bit RAW sensor data  to white balanced 10-bit 1280 x 720 pixel RGB 4:4:4 video."  Things to note here:

  1. I think there's a typo in the manual - they forgot the "to". As in 12 bit RAW "to" 10 bit video. If we add the "to" the sentence makes sense.
  2. The characteristics of the video are lower resolution: 1280x720; 10 bit color.
    But what is that 4:4:4? - That is shorthand for uncompressed across Y:Cr:Cb.  Without a detailed explanation, take my word for it that those values can be basically translated as Y = Luminance:Cr = Chrominance or color:Cb = Saturation.  What can be a little confusing is that uncompressed in the film world means 4:4:4 (or potentially even 8:4:4) but somehow in the real world broadcast tv color that was in fact compressed to 4:2:2 came to be known as uncompressed.  Then there were the "consumer" and "pro-sumer" codecs like HDV and mini-DV - these were compressed to 4:2:0 and 4:1:1 respectively. 

    What it means here is the monitors are getting the best 10 bit color possible at 4:4:4 uncompressed.
  3. This signal may be modified using ISO, white balance or other RGB color space adjustments.   Freedom!  Experiment a little in camera and know that you can change things in post.
  4. The monitor path includes all the monitors.
Previous Up a Level Next
<< RED One Image Processing RED One Theory of Operation RED One Audio Recording >>

Page last modified on Friday 27 of March, 2009 20:44:02 EDT

Style: ovsdarkroom.css [toggle]

Online users [toggle]

16 online users