You can use a Write node to both render an image and read the rendered image back in. Because reading the output of a Write node from a file is faster than calculating its output by processing the node tree upstream, this can be particularly useful on large comps. When you have finished working on a branch of the node tree, you can insert a Write node after it, render the output, and use the same Write node to read the rendered image in. If you later need to edit the nodes upstream, simply make your changes and render the Write node again to update the image being read in.
1. | Render an image as described in To Render a Single Write Node. We recommend rendering the image as an .exr. This way, Nuke writes the hash value of the incoming node tree into the rendered file. If the node tree changes and the rendered file gets out of date, the hashes won’t match and Nuke notifies you of the problem. |
2. | In the Write node properties, check read file. When this is on, Nuke ignores the upstream node tree and uses the rendered image as the output of the Write node. |
3. | To check whether the input file is up to date with the input tree connected to the Write node, check check file matches input. This only works with .exr files written by Nuke and when the proxy mode and down-rez are disabled. If the input file cannot be checked, Nuke displays the word unchecked on the Write node in the Node Graph. |
4. | If there is an error when loading the rendered file, select what to do from the missing frames dropdown menu: |
• error - display an error message on any missing frames.
• black - replace any missing frames with black.
• checkerboard - replace any missing frames with a checkerboard image.
• read input - display the result of the input tree rather than the rendered file on any missing frames.
TIP: You can also use the Precomp node (Other > Precomp) to reduce portions of the node tree to pre-rendered image inputs.
The hash value is a unique number (for example, b1c9c0aff2012a8) calculated from a node and the entire tree of nodes connected to its input. The class of the node and all the current control settings contribute to the hash value.
You can display the hash value at any point in the node tree by selecting a node in the Node Graph and pressing I. The hash is different at different points in the tree.