Overview
The basic node for compositing two images. It is equivalent to a layer plus blend mode in painting software, covering everything from placing an asset over a base to Multiply shadows and Add glows.
Usage tips
- Choose the mode by intent: place an asset unchanged →
Normal; shadows or grime →Multiply; light or highlights →ScreenorAdd(Add gives a stronger glow); emphasize color →Overlay. - If the effect is too strong, reduce
opacitybefore changing modes. It multiplies the top image's alpha:0is an exact base-image identity, including hidden transparent RGB, and1uses the top image's original alpha. A practical range is often 0.3–0.6. - When compositing images of different sizes, decide
output_sizeandanchorfirst. For placing a sprite on a canvas, the go-to setup isoutput_size = Base+anchor = Center. - For three or more images, chain Blend nodes. If there are many layers,
image_layer_composeis easier to organize. - Use
blend_by_maskwhen compositing through a shape.
Hover over a mode in the drop-down menu to see its tooltip description.
Common pitfalls
- The top image completely hides the bottom one:
Normaloverwrites the base. Check the top image's transparent areas or reduceopacityif the base should show through. - Overflow is clipped:
output_size = Base(the default) outputs only the base image bounds. ChooseLargerto fit both images after alignment and offsets. - The position is offset incorrectly: placement is resolved from
anchor, thenoffset. The offset slider range follows the canvas size; use numeric input (up to ±4096) for larger movements.
Placement parameters
output_size (output size)
| Value | Label | Description |
|---|---|---|
| Base | Base | Output the base image size (default). The base is placed at (0,0) |
| Blend | Blend | Output the blend image size |
| Larger | Larger | Fit both images after alignment and offsets; negative positions shift the output origin |
| Smaller | Smaller | Use the smaller dimension on each axis: min(base.w, blend.w) × min(base.h, blend.h) |
When using anything other than
Base, the base image is centered on the initial canvas.Largerthen expands to both image bounds, so negative offsets are retained.
alignment
Choose one of nine positions for the blend image on the output canvas:
Top-Left Top-Center Top-Right
Center-Left Center Center-Right
Bottom-Left Bottom-Center Bottom-Right
offset_x / offset_y (offset)
Fine-tune the position in pixels after alignment. Positive values move right/down.
Dynamic slider range: the slider range follows the project canvas size (offset_x → ±canvas width, offset_y → ±canvas height). Numeric fields accept direct values up to ±4096.
Examples
Basic: compositing same-size images
Composite a 32×32 overlay over a 32×32 base with Normal, opacity: 0.5.
Different sizes: centering a small image
Place a 16×16 icon on a 32×32 base with alignment: Center → the icon occupies
(8,8)–(23,23).
Placement with an offset
Place a 32×32 sprite on a 64×32 base with alignment: Top-Left, offset_x: 4 → the
sprite starts at (4,0).
Related nodes
image_layer_compose— composite many layers at once (best for three or more)blend_by_mask— control the composited area with a mask imageerase_by_mask— erase with a mask instead of compositingsolid_color— create a base for compositing