Painting By Objects

Let's take a closer look at this scene:-

There are actually more levels of nesting under Scene than Scene < (House Wood Background). The full structure looks like this:-

Now lets take a closer look at House < (Window) and Wood < (Tree).

House < (Window)

All the windows are in fact the same window transformed in some way e.g. by using the scale transformation on the layers that Window was loaded to in House.

Wood < (Tree)

Similarly here there is only really one conifer, the one in Tree. It has been loaded to five layers in Wood and then each layer has been altered in some way: e.g. stretched or shrunk and colour-corrected. Notice also that in Scene I have loaded Wood twice so there are a total of ten conifers, all slightly different, produced from one conifer.


Why do it this way? Well suppose you, or your client, change their minds and ask for red window frames on the house and oak trees instead of conifers, all you have to change is one window frame in Window and one tree in Tree.

You can achieve the same results by copying layers or saving parts of the image as rir files and loading them to layers. However if you copy layers you have no link between the original and the copy; changing the original does nothing to the copy. The same is true if you save part of an image as an rir file; you must to remember to rerender the rir file when you make a change.


I have not found any drawbacks yet though there are a couple of points to bear in mind.

  1. Using nested canvases does make Satori do a lot of work when rendering the final scene. However, if the scene is complex then whichever method you use Satori will have to do that work at some point in the process of designing and rendering the canvas. I prefer to have a quick response while I am working on a canvas and let the batch renderer take the strain producing the final version.
  2. Saving canvases with "low compression" does increase the size of the canvas files. I consider this to be an acceptable trade off for fast response while working on a canvas. If you want to reduce the size of a canvas for any reason, you can save it with high compression without any loss of information because all the objects in the canvas are still stored. 

<--- Back to Part 1 (Introduction)           Forward to Part 3 (The Third Dimension) --->


 Nigel Sutton