rgb128

Rgb128 is a term used to describe a color representation that uses 128 bits per pixel. In typical interpretations, this is achieved with four channels—red, green, blue, and alpha—each stored with 32 bits. The per-channel data can be integers or floating point numbers, with common variants including RGBA32 (unsigned integers) and RGBA32F (32-bit floating point per channel). Using 32-bit floating point per channel provides very high precision and a wide dynamic range, supporting linear-light workflows, high-dynamic-range imaging, and wide-gamut color management.

Encoding and formats

In modern graphics pipelines, rgb128 often corresponds to a 128-bit per-pixel format such as R32G32B32A32_FLOAT. These formats are supported as render targets and textures in current graphics APIs and hardware, including DirectX, Vulkan, and OpenGL. The exact memory layout and channel interpretation can vary between implementations, but the common goal is to minimize color quantization and preserve detail through processing stages.

Usage and practicality

Rgb128 offers advantages in precision for professional rendering, compositing, scientific visualization, and HDR workflows. However, it has substantial memory and bandwidth costs compared with more common 8-bit or 16-bit-per-channel formats. Consumer displays and typical imaging pipelines rarely require rgb128, and many applications instead use lower-bit-depth formats or 32-bit per channel HDR approaches.

Relation to other formats

Rgb128 is a descriptive term rather than a single standardized format. It denotes high-precision color per pixel and can refer to several concrete implementations depending on the API and hardware, typically involving four channels at 32 bits each.