PIPELINE

This chapter looks at how assets and their pixels, or rather colors, are transformed inside WATCHOUT. Although related to the Asset Manager chapter, this information places more emphasis on the color transformation taking place inside WATCHOUT.

Below is an overview of the different stages an asset and its pixels go through before being shown to an audience:

Alt text

  • Original

    • This is the first stage where you record/create/receive the original asset file that you want to visualize using WATCHOUT.
    • WATCHOUT supports a number of file formats and codecs which you can read more about here.
    • The quality of the original will have a big impact on the visual quality you can expect in the final stage.
    • Factors that affect the quality are:
      • Color space.
      • Resolution (the number of pixels in the image or video).
      • Frames per second (if it is a video).
      • The type of compression, if any, that was used when the asset was created.
        • Compression may reduce the accuracy of the colors in an image or video.

  • Asset Manager

    • The asset reaches this state when it has been added to the Asset Manager.
    • At this stage the asset can be transcoded to a different codec depending on the settings you have chosen in the Media Settings, which can be accessed from the Asset Manager file menu.
    • If the asset is transcoded from a codec using a color space with a wide color gamut to a more narrow color gamut you may lose color accuracy.
      • This depends on which colors are within the color space being used.
    • You may also lose quality due to compression.
    • It is important to understand that loss of quality might still be acceptable and a good trade-off to improve the playback performance.
      • Knowledge of the system capabilities where the WATCHOUT show will be played is key to making a good decision on this.
    • The settings inside Producer are deemed to be good defaults and therefore recommended. Only change these if you know what you are doing.
    • The end result of this stage can be referred to as the WO asset.

  • Visual Renderer

    • When the original asset has been processed and a WO asset has been created, it is ready to be sent to Visual Renderer.
    • The main goal of Visual Renderer is to output color accurate pixels at the right time and as fast as possible. This process includes the following steps:
      • Read the frame(s) of the asset as fast as possible.
        • Multiple frames are read in parallel to assure the data is available when it needs to be presented.
          • Disk and CPU clock speed, along with the number of assets that need be loaded at the same time, will affect how well this task can be performed.
      • Apply effects that have been assigned to the asset.
        • WATCHOUT strives to maintain color accuracy. The internal engine supports 16-bit color depth which should be enough for most use cases.
        • The effects are applied in linear Rec. 2020 color space as this allows more colors to be represented. Additionally, some effects may alter the color in a way that it could not be accurately represented with sRGB for instance.
        • In the Output section of the Display Properties you can select the output color depth and color space you want to use.
        • It is important to understand the capabilities of your output display/projector/device.
          • There is no point in trying to output colors in a configuration that is not supported by the output device.
      • Composite the pixels from an asset against assets already rendered.
      • Send the pixels to the next and final stage.

  • Display

    • This is the final stage of the pipeline and where the pixels are finally shown to the audience.
    • As outlined in the previous stage, the output can be configured in the Output section of the Display Properties.

WHAT COLOR SPACE SHOULD I USE IN WATCHOUT?

Imagine a scenario in which you want to show a commercial in a mall on a big screen. This "big" screen consists of 8 x 8 smaller screens that must play in sync. What color space should be deployed?

The question is a bit too generic. You may have one color space for input (camera) and a completely different color space for output (display). Bear in mind that if you record using a narrow color space you generally have little to gain by using a wider color range for output, at least if the aim is to replay the content as close to reality as possible.

So to answer this question, you need to first address other key questions:

  • What content do you want to show in the commercial?
  • From what distance will the audience view the content?
  • What color spaces are supported by the:
    • Camera used to shoot the commercial?
    • The screens that will be used?
    • Tool/software used to process the playback?
  • What frame rate and resolution do you want to use during playback and are the screens capable of reading, processing and displaying the pixels fast enough?
  • What effects do you want to use and how do these effects impact playback performance?

The list goes on. As you can see there are many things that can affect this decision and it often boils down to trade-offs. For instance:

  • Are you willing to sacrifice frame rate for more vivid colors?
  • Are you willing to sacrifice color accuracy to be able to get smooth 60 FPS playback?

So to conclude there is no clear-cut answer to this question. Experience and experiments will guide your way.