Blend Method Parameter

Language:

English • français • italiano • română • русский

The various compositing methods available for Layers in Synfig. (What compositing/blending is.)

If the following descriptions, 'A' refers to the color on the layer with the blend method setting, and 'B' refers to the color on the layers beneath it.

In the examples that follow a gradient ranging from black to white to transparent (A) is blended on the layer above an inline canvas of Tux on a transparent background (B).

A) & B)

The currently available blend methods are:

Composite

This blend method is the default option, it simply displays the content of the layer. This blend mode is similar the layer blend mode Normal Mode often found in 2D programs.

Straight

This blend method looks similar to the previous one, except that the objects under a straight-object will be invisible. So if there is a more or less transparent object on the Straight mode layer, the objects on the layers underneath won't show through it.

More precisely, the resulting color is "(A-B)*amount + B". So if amount is 1 the result is A and if amount is 0 the result is B. In particular, if amount is 1 and A is a very transparent color, the resulting color will also be A; despite the fact that A is very transparent, none of B's color is used.

Onto

If a layer is set to the Onto blend method, only the parts of the layer that are over a not transparent area will be visible.

Precisely: this is the same as the Composite blend method except that the transparency of the resulting color is set to be the same as the transparency level of layer B.

Straight Onto

This method is a combination of the two methods above. E.g. if an half-transparent object is set to Straight Onto, it will only be visible over a non-transparent area, and the non-transparent part under that object won't be visible.

Precisely: the resulting color is "(X-B)*amount + B" where X is A but with its transparency set to A's transparency times B's transparency.

So if amount is 1 the result is A, but with its transparency multiplied by that of B, and if amount is 0 the result is B. In particular, if amount is 1 and A is a very transparent color, the resulting color will be a possibly more transparent version of A; despite the fact that A is very transparent, none of B's color is used in the result.

(Yuck. Are these 'precisely' comments useful?)

Behind

This blend method makes the layer visible over transparent areas, and invisible over non-transparent areas, giving the impression that the layer is behind the other layers. It is often used for the "Shade" effect layer, to make a drop-shadow effect.

Precisely: this is the same as the composite blend method, but with A and B swapped. B is composited onto A instead of A being composited onto B.

Screen

This blend method is similar to the Screen Mode often found in 2D programs. It combines the colors of the screen mode layer and the ones behind it, and gives a lighter result in general.

Overlay

This is similar to PhotoShop - layer blend mode Overlay Mode

Hard Light

This is similar to PhotoShop - layer blend mode Hard Light Mode

Multiply

This is similar to PhotoShop - layer blend mode Multiply Mode

Precisely: the resulting colour is (((A*B)-B)*amount+B). The calculation is performed independently on red, green, and blue components. When amount is 0, this simplifies to B. When amount is 1 it simplifies to A*B.

Divide

Describe me

Precisely: the resulting color is (((B/A)-B)*amount)+B.

When amount is 0, this becomes simply B.

When amount is 1, this becomes B/A.

A very small quantity is added to A before dividing by it to avoid a divide-by-zero condition. This causes the divide blend method to bias toward positive values, but the effect is really neglegable.