pyktx.ktx_astc_params module

class pyktx.ktx_astc_params.KtxAstcParams(verbose: bool = False, thread_count: int = 1, block_dimension: KtxPackAstcBlockDimension = KtxPackAstcBlockDimension.D4x4, mode: KtxPackAstcEncoderMode = KtxPackAstcEncoderMode.DEFAULT, quality_level: int = KtxPackAstcQualityLevels.FASTEST, normal_map: bool = False, perceptual: bool = False, input_swizzle: bytes = b'\x00\x00\x00\x00')[source]

Bases: object

Data for passing extended parameters to KtxTexture2.compress_astc().

block_dimension: KtxPackAstcBlockDimension = 0: Combinations of block dimensions that astcenc supports i.e. 6x6, 8x8, 6x5 etc

input_swizzle: bytes = b'\x00\x00\x00\x00'

A swizzle to provide as input to astcenc.

It must match the regular expression /^[rgba01]{4}$/.

mode: KtxPackAstcEncoderMode = 0: Can be {ldr/hdr} from astcenc

normal_map: bool = False

Tunes codec parameters for better quality on normal maps.

In this mode normals are compressed to X,Y components Discarding Z component, reader will need to generate Z component in shaders.

parse_swizzle(swizzle: str) → None[source]

perceptual: bool = False

The codec should optimize for perceptual error, instead of direct RMS error.

This aims to improves perceived image quality, but typically lowers the measured PSNR score. Perceptual methods are currently only available for normal maps and RGB color data.

quality_level: int = 0: astcenc supports -fastest, -fast, -medium, -thorough, -exhaustive

thread_count: int = 1: Number of threads used for compression. Default is 1.

verbose: bool = False: If true, prints Astc encoder operation details to stdout. Not recommended for GUI apps.