The “Continuously Rasterize” switch is overloaded. Although it often works as the “keep it sharp” button, this switch with the sun icon has two very different functions, depending on layer type.
For vector layers, this switch tells Ae, “Transform this layer first, then redraw the vector at its new size, position and rotation after the transformation.” This gives you sharp lines, no matter how big you scale your layer. It also changes the render order, rendering effects after transformations instead of before. Without Continuously Rasterize checked, the layer is drawn (“rasterized”) only once at its original 100% size, then effects are rendered, then that raster image is transformed.
For composition layers, the switch is actually called “Collapse Transformations.” This tells Ae, “Treat the layers inside this precomp as if they were a bunch of layers that were directly in the containing comp.” While this sometimes has the same effect as Continuous Rasterization (like when you have vector layers inside a precomp with no effects on the precomp layer), it works very differently.
Because Collapse Transformations essentially pulls the layers from the precomp into the main comp for rendering, it lets you do cool things like use blend modes, combine 3D spaces, and, like its name suggests, concatenate transformations. This means that the two separate transform operations (position, scale, rotation) from inside the comp and outside the comp are combined into a single operation. In other words, a 25% scale on a layer inside the precomp, and a 400% scale on the precomp layer inside the main comp will cancel each other out and equal 100% scale, instead of creating a blurry pixelated mess.
The downside is that effects applied to the precomp layer will no longer work quite the same. Remember, we’re now treating the contents of the precomp as if they were a set of individual layers inside the containing comp; we are no longer rendering the precomp in memory and treating it as a single image in the containing comp before applying effects. Ae will apply any effects on the precomp layer to each and every layer within the precomp after transformations have been collapsed, and it will apply them in their own native layer spaces. This can yield really different results than applying the same effects to the one composite of all those layers.
This is a confusing topic (and a confusing UX in Ae). If you have any questions, please ask away and I’ll do my best to clarify.
Walter Soyka
Designer & Mad Scientist at Keen Live [link]
Motion Graphics, Widescreen Events, Presentation Design, and Consulting
@keenlive | RenderBreak [blog] | Profile [LinkedIn]