Array to texture
Array to texture turns a float value array into a texture. Use it for generated scalar fields, lookup tables or color data, not encoded PNG or JPEG file bytes.
Ports and controls
| Port or control | Meaning |
|---|---|
| Input | Flat float array containing consecutive pixel components. |
| Size | Image width and height in pixels. |
| Format | Pixel storage format and channel count. |
| Output | Texture containing the uploaded pixel data. |
Supply width × height × components values. For example, Size 2 × 1 with Format RGBA32F expects eight values: [1, 0, 0, 1, 0, 1, 0, 1] describes a red pixel followed by a green pixel. Connect Output to a texture consumer in the Render Pipeline.
Formats and numeric ranges
The converter implements these storage formats:
- 8-bit components: RGBA8, BGRA8, R8, RG8, RED_OR_ALPHA8 and R8UI.
- 16-bit unsigned components: R16 and RG16.
- 32-bit unsigned components: R32UI, RG32UI and RGBA32UI.
- Floating-point components: R16F, RGBA16F, R32F and RGBA32F.
Values are converted directly to the destination component type: there is no automatic 0–1 to 0–255 scaling for 8-bit formats. Supply values in the format’s representable range; for example, an RGBA8 red pixel is [255, 0, 0, 255], not [1, 0, 0, 1]. Channel order follows the selected format, including BGRA rather than RGBA where selected.
Extra values are ignored after the image is full. A short array does not define all pixels, so do not rely on missing components being filled for you. Changes to Input, Size or Format recreate and upload the texture.
Texture to buffer produces GPU buffer bytes, not a float array, and is therefore not the direct inverse of this process. See Geometry and buffer utilities for CPU-array and GPU-buffer conversions.