RAYZ is a procedural, node-based compositor in which the completed shot consists of a network of nodes, where the nodes represent image operations and the connections among them represent the flow of the image data through the network.
A RAYZ project file starts with one or more source nodes, such as the Image In node. The Image In node points to the location (on a local or networked disk drive or other storage device) of a sequence of image files you will be modifying, such as the background plates for the scene or a foreground element rendered from a 3D modeling program. The source image data is loaded into memory for manipulation in RAYZ.
The source imagery is modified and combined using various nodes, which are categorized by general function: color correction, matte generation, composite operations, and so forth.
The nodes are connected in a flowchart. As you build this flowchart, you are creating a description of the shot. You can think of the nodes as the building blocks, which you can add, delete, and rearrange as needed to produce the appropriate result. The source images begin streams of data that flow through this network of nodes, being modified by each node in turn.
The RAYZ interface is designed to let you focus on the images instead of the flowchart, however, if you choose. To a great extent, the flowchart of nodes builds itself as you work on the images, and it is always available to be reviewed and edited in the Worksheet.
Whatever operations you perform on this imagery, finally you will want to render new image files of the completed shot to disk. You can render any node image by connecting it to an Image Out node, where you specify the name, location, and format of the files and then render them whenever you choose.
The dataflow model also has the advantage of being "self-documenting," which simply means that the structure of the network itself shows you how the shot has been constructed, making it easier to pinpoint areas to be changed or fine-tuned--by the shot's creator or by another artist.
The Worksheet is used not only to build the flowchart of nodes, but also to control which node data appears in other parts of the interface. To see how an image looks in any particular node, and to modify the node parameters, you select the node in the Worksheet. The node image is automatically displayed in the Image Viewer and the node parameters show up in the Node Panel.
This concept is called dynamic focus and it is basic to controlling what image data is displayed where and when in the interface. For a detailed description of how to use it, and how to override it when you wish, refer to Dynamic Focus in chapter 4.
RAYZ will use non-adjacent concatenation of operations wherever feasible to render imagery, and employ the fastest possible filtering algorithms that do not compromise image quality.
The resolution of the output is only determined at render time. Whatever output frame size you specify, RAYZ is storing the image data in an infinite x,y coordinate space so that data is not lost when it is transformed out of frame. When you render an image, all pixels that fall within the area you specify at that time are rendered.
Once you import imagery into RAYZ, you can modify it by selecting operations directly in the Image Viewer. (Hold down the Space bar for a menu of all available node operations.) Each time you select a filter, transform, or other node, not only is the image updated but also the Worksheet area, where the corresponding node is inserted into the network.
The parameters that can be set for the various node operations appear automatically in the Node Panel for the currently selected node image. The individual parameters are organized to show the most common, convenient options by default and to hide the rest, but every option can be displayed with a mouse click, as when you need control of individual image channels, advanced filtering options, etc.
Node parameter values can be displayed in pixel units, or in floating point units, as you choose. However, RAYZ always stores the values at floating point precision. This means that you can switch back and forth between proxy and full size images, or clone a node and connect the clones to different size images, without having to make any adjustments.
You can view and modify an image or any individual channel in an image. Whenever appropriate, parameters offer a master control that operates on all channels, with the option of accessing the individual channel controls to adjust them separately.
You can carry the opacity data (the matte) with the RGB image as an integrated alpha channel and still control which channel a node affects; you do not have to send a separate alpha channel image through the network unless you choose.
In fact, most node parameters enable you to control five image channels: the RGB channels, the alpha channel, and a fifth channel (labeled "O" for Other), such as that used to carry z-depth or other image information.
RAYZ lets you combine images with different resolutions and bit depths without having to explicitly specify which image is scaled, translated, promoted or demoted, and so forth.
For images of different sizes, pixels are aligned according to their x,y coordinates in relation to a 0,0 point, which means that images are aligned to the lower left corner unless you specify an offset or apply some other type of spatial transform.
For images of different bit depths, the lower is linearly remapped to the higher. RAYZ operates in 8-bit, 16-bit, or floating point (32-bit) color depth per channel, as you choose.
By default, any Cineon 10-bit log images you import are automatically remapped to 16-bit linear, using Kodak specifications. But you always have the option to modify the default conversion values or to use the unconverted log data.
RAYZ will automatically choose an appropriate LUT or other method (such as Cineview emulation) for monitor display conversion as you view different types of imagery. You can also set the display conversion method manually, and you can add custom LUTs to the list of options.
By design the RAYZ interface is dynamic and flexible. You can add, duplicate, and delete the main components--the Image Viewer, the Worksheet, etc.--as well as resize and reposition them. The components can all be docked together in one window, or individual components can float in separate windows.
The interface can be adjusted on the fly, but you can also save and reuse any configuration, which is called a layout. The individual parameters and tools are just as flexible, and you can set preferences to control colors, default values, and hotkey assignments.
Be sure to read Chapter 4: Overview of the Interface, for a thorough introduction to the RAYZ interface and how to customize it for your work habits and projects.