Package icyllis.arc3d.core

Class ColorInfo

java.lang.Object
icyllis.arc3d.core.ColorInfo
public final class ColorInfo extends Object
Describes pixel dimensions and encoding.

ColorInfo is used to interpret a color: color type + alpha type + color space.

  • Field Details

    • COMPRESSION_NONE

      public static final int COMPRESSION_NONE
      Compression types.
      Core GL_COMPRESSED_* VK_FORMAT_*_BLOCK
      ETC2_RGB8_UNORM RGB8_ETC2 ETC2_R8G8B8_UNORM
      BC1_RGB8_UNORM RGB_S3TC_DXT1_EXT BC1_RGB_UNORM
      BC1_RGBA8_UNORM RGBA_S3TC_DXT1_EXT BC1_RGBA_UNORM
      See Also:

    • COMPRESSION_ETC2_RGB8_UNORM

      public static final int COMPRESSION_ETC2_RGB8_UNORM
      Compression types.
      Core GL_COMPRESSED_* VK_FORMAT_*_BLOCK
      ETC2_RGB8_UNORM RGB8_ETC2 ETC2_R8G8B8_UNORM
      BC1_RGB8_UNORM RGB_S3TC_DXT1_EXT BC1_RGB_UNORM
      BC1_RGBA8_UNORM RGBA_S3TC_DXT1_EXT BC1_RGBA_UNORM
      See Also:

    • COMPRESSION_BC1_RGB8_UNORM

      public static final int COMPRESSION_BC1_RGB8_UNORM
      Compression types.
      Core GL_COMPRESSED_* VK_FORMAT_*_BLOCK
      ETC2_RGB8_UNORM RGB8_ETC2 ETC2_R8G8B8_UNORM
      BC1_RGB8_UNORM RGB_S3TC_DXT1_EXT BC1_RGB_UNORM
      BC1_RGBA8_UNORM RGBA_S3TC_DXT1_EXT BC1_RGBA_UNORM
      See Also:

    • COMPRESSION_BC1_RGBA8_UNORM

      public static final int COMPRESSION_BC1_RGBA8_UNORM
      Compression types.
      Core GL_COMPRESSED_* VK_FORMAT_*_BLOCK
      ETC2_RGB8_UNORM RGB8_ETC2 ETC2_R8G8B8_UNORM
      BC1_RGB8_UNORM RGB_S3TC_DXT1_EXT BC1_RGB_UNORM
      BC1_RGBA8_UNORM RGBA_S3TC_DXT1_EXT BC1_RGBA_UNORM
      See Also:

    • COMPRESSION_COUNT

      public static final int COMPRESSION_COUNT
      Compression types.
      Core GL_COMPRESSED_* VK_FORMAT_*_BLOCK
      ETC2_RGB8_UNORM RGB8_ETC2 ETC2_R8G8B8_UNORM
      BC1_RGB8_UNORM RGB_S3TC_DXT1_EXT BC1_RGB_UNORM
      BC1_RGBA8_UNORM RGBA_S3TC_DXT1_EXT BC1_RGBA_UNORM
      See Also:

    • AT_UNKNOWN

      public static final int AT_UNKNOWN
      Alpha types.

      Describes how to interpret the alpha component of a pixel. A pixel may be opaque, or alpha, describing multiple levels of transparency.

      In simple blending, alpha weights the source color and the destination color to create a new color. If alpha describes a weight from zero to one:

      result color = source color * alpha + destination color * (1 - alpha)

      In practice alpha is encoded in two or more bits, where 1.0 equals all bits set.

      RGB may have alpha included in each component value; the stored value is the original RGB multiplied by alpha. Premultiplied color components improve performance, but it will reduce the image quality. The usual practice is to premultiply alpha in the GPU, since they were converted into floating-point values.

      See Also:

    • AT_OPAQUE

      public static final int AT_OPAQUE
      Alpha types.

      Describes how to interpret the alpha component of a pixel. A pixel may be opaque, or alpha, describing multiple levels of transparency.

      In simple blending, alpha weights the source color and the destination color to create a new color. If alpha describes a weight from zero to one:

      result color = source color * alpha + destination color * (1 - alpha)

      In practice alpha is encoded in two or more bits, where 1.0 equals all bits set.

      RGB may have alpha included in each component value; the stored value is the original RGB multiplied by alpha. Premultiplied color components improve performance, but it will reduce the image quality. The usual practice is to premultiply alpha in the GPU, since they were converted into floating-point values.

      See Also:

    • AT_PREMUL

      public static final int AT_PREMUL
      Alpha types.

      Describes how to interpret the alpha component of a pixel. A pixel may be opaque, or alpha, describing multiple levels of transparency.

      In simple blending, alpha weights the source color and the destination color to create a new color. If alpha describes a weight from zero to one:

      result color = source color * alpha + destination color * (1 - alpha)

      In practice alpha is encoded in two or more bits, where 1.0 equals all bits set.

      RGB may have alpha included in each component value; the stored value is the original RGB multiplied by alpha. Premultiplied color components improve performance, but it will reduce the image quality. The usual practice is to premultiply alpha in the GPU, since they were converted into floating-point values.

      See Also:

    • AT_UNPREMUL

      public static final int AT_UNPREMUL
      Alpha types.

      Describes how to interpret the alpha component of a pixel. A pixel may be opaque, or alpha, describing multiple levels of transparency.

      In simple blending, alpha weights the source color and the destination color to create a new color. If alpha describes a weight from zero to one:

      result color = source color * alpha + destination color * (1 - alpha)

      In practice alpha is encoded in two or more bits, where 1.0 equals all bits set.

      RGB may have alpha included in each component value; the stored value is the original RGB multiplied by alpha. Premultiplied color components improve performance, but it will reduce the image quality. The usual practice is to premultiply alpha in the GPU, since they were converted into floating-point values.

      See Also:

    • CT_UNKNOWN

      public static final int CT_UNKNOWN
      Color types.

      Describes a layout of pixel data in CPU or GPU memory. A pixel may be an alpha mask, a grayscale, RGB, or RGBA. It specifies the channels, their type, and width. It does not refer to a texture format and the mapping to texture formats may be many-to-many. It does not specify the sRGB encoding of the stored values.

      Color types are divided into two classes: array and packed.
      For array types, the components are listed in order of where they appear in memory. For example, CT_RGBA_8888 means that the pixel memory should be interpreted as an array of uint8 values, and the R channel appears at the first uint8 value. This is the same naming convention as Vulkan.
      For packed types, the first component appear in the least-significant bits. For example, CT_BGR_565 means that each pixel is packed as (b << 0) | (g << 5) | (r << 11), an uint16 value. This is in the reverse order of Vulkan's naming convention.

      Note that if bytes-per-pixel of a color type is 1, 2, 4, or 8, then Arc3D requires pixel memory to be aligned to bytes-per-pixel, otherwise it should be aligned to the size of data type as normal.

      See Also:

    • CT_BGR_565

      public static final int CT_BGR_565
      Color types.

      Describes a layout of pixel data in CPU or GPU memory. A pixel may be an alpha mask, a grayscale, RGB, or RGBA. It specifies the channels, their type, and width. It does not refer to a texture format and the mapping to texture formats may be many-to-many. It does not specify the sRGB encoding of the stored values.

      Color types are divided into two classes: array and packed.
      For array types, the components are listed in order of where they appear in memory. For example, CT_RGBA_8888 means that the pixel memory should be interpreted as an array of uint8 values, and the R channel appears at the first uint8 value. This is the same naming convention as Vulkan.
      For packed types, the first component appear in the least-significant bits. For example, CT_BGR_565 means that each pixel is packed as (b << 0) | (g << 5) | (r << 11), an uint16 value. This is in the reverse order of Vulkan's naming convention.

      Note that if bytes-per-pixel of a color type is 1, 2, 4, or 8, then Arc3D requires pixel memory to be aligned to bytes-per-pixel, otherwise it should be aligned to the size of data type as normal.

      See Also:

    • CT_R_8

      public static final int CT_R_8
      Color types.

      Describes a layout of pixel data in CPU or GPU memory. A pixel may be an alpha mask, a grayscale, RGB, or RGBA. It specifies the channels, their type, and width. It does not refer to a texture format and the mapping to texture formats may be many-to-many. It does not specify the sRGB encoding of the stored values.

      Color types are divided into two classes: array and packed.
      For array types, the components are listed in order of where they appear in memory. For example, CT_RGBA_8888 means that the pixel memory should be interpreted as an array of uint8 values, and the R channel appears at the first uint8 value. This is the same naming convention as Vulkan.
      For packed types, the first component appear in the least-significant bits. For example, CT_BGR_565 means that each pixel is packed as (b << 0) | (g << 5) | (r << 11), an uint16 value. This is in the reverse order of Vulkan's naming convention.

      Note that if bytes-per-pixel of a color type is 1, 2, 4, or 8, then Arc3D requires pixel memory to be aligned to bytes-per-pixel, otherwise it should be aligned to the size of data type as normal.

      See Also:

    • CT_RG_88

      public static final int CT_RG_88
      Color types.

      Describes a layout of pixel data in CPU or GPU memory. A pixel may be an alpha mask, a grayscale, RGB, or RGBA. It specifies the channels, their type, and width. It does not refer to a texture format and the mapping to texture formats may be many-to-many. It does not specify the sRGB encoding of the stored values.

      Color types are divided into two classes: array and packed.
      For array types, the components are listed in order of where they appear in memory. For example, CT_RGBA_8888 means that the pixel memory should be interpreted as an array of uint8 values, and the R channel appears at the first uint8 value. This is the same naming convention as Vulkan.
      For packed types, the first component appear in the least-significant bits. For example, CT_BGR_565 means that each pixel is packed as (b << 0) | (g << 5) | (r << 11), an uint16 value. This is in the reverse order of Vulkan's naming convention.

      Note that if bytes-per-pixel of a color type is 1, 2, 4, or 8, then Arc3D requires pixel memory to be aligned to bytes-per-pixel, otherwise it should be aligned to the size of data type as normal.

      See Also:

    • CT_RGB_888

      @Internal public static final int CT_RGB_888
      See Also:

    • CT_RGBX_8888

      public static final int CT_RGBX_8888
      See Also:

    • CT_RGBA_8888

      public static final int CT_RGBA_8888
      See Also:

    • CT_BGRA_8888

      public static final int CT_BGRA_8888
      See Also:

    • CT_RGBA_8888_SRGB

      @Internal public static final int CT_RGBA_8888_SRGB
      See Also:

    • CT_RGBA_1010102

      public static final int CT_RGBA_1010102
      See Also:

    • CT_BGRA_1010102

      public static final int CT_BGRA_1010102
      See Also:

    • CT_R_16

      @Internal public static final int CT_R_16
      See Also:

    • CT_R_F16

      @Internal public static final int CT_R_F16
      See Also:

    • CT_RG_1616

      @Internal public static final int CT_RG_1616
      See Also:

    • CT_RG_F16

      @Internal public static final int CT_RG_F16
      See Also:

    • CT_RGBA_16161616

      public static final int CT_RGBA_16161616
      See Also:

    • CT_RGBA_F16

      public static final int CT_RGBA_F16
      See Also:

    • CT_RGBA_F16_CLAMPED

      public static final int CT_RGBA_F16_CLAMPED
      See Also:

    • CT_RGBA_F32

      public static final int CT_RGBA_F32
      See Also:

    • CT_ALPHA_8

      public static final int CT_ALPHA_8
      See Also:

    • CT_ALPHA_16

      public static final int CT_ALPHA_16
      See Also:

    • CT_ALPHA_F16

      public static final int CT_ALPHA_F16
      See Also:

    • CT_GRAY_8

      public static final int CT_GRAY_8
      See Also:

    • CT_GRAY_ALPHA_88

      @Internal public static final int CT_GRAY_ALPHA_88
      See Also:

    • CT_ABGR_8888

      @Internal public static final int CT_ABGR_8888
      Special format for big-endian CPU; GPU does not support this.
      See Also:

    • CT_ARGB_8888

      @Internal public static final int CT_ARGB_8888
      Special format for big-endian CPU; GPU does not support this.
      See Also:

    • CT_RGBA_8888_NATIVE

      public static final int CT_RGBA_8888_NATIVE
      A runtime alias based on host endianness, packed as (r << 0) | (g << 8) | (b << 16) | (a << 24) an uint32 value.

      This is not a standalone packed format, it just depends on CPU: on big-endian machine this is CT_ABGR_8888; on little-endian machine this is CT_RGBA_8888.

    • CT_BGRA_8888_NATIVE

      public static final int CT_BGRA_8888_NATIVE
      A runtime alias based on host endianness, packed as (b << 0) | (g << 8) | (r << 16) | (a << 24) an uint32 value.

      This is not a standalone packed format, it just depends on CPU: on big-endian machine this is CT_ARGB_8888; on little-endian machine this is CT_BGRA_8888.

    • CT_R5G6B5_UNORM

      public static final int CT_R5G6B5_UNORM
      Aliases.
      See Also:

    • CT_R8G8_UNORM

      public static final int CT_R8G8_UNORM
      Aliases.
      See Also:

    • CT_A16_UNORM

      public static final int CT_A16_UNORM
      Aliases.
      See Also:

    • CT_A16_FLOAT

      public static final int CT_A16_FLOAT
      Aliases.
      See Also:

    • CT_A16G16_UNORM

      public static final int CT_A16G16_UNORM
      Aliases.
      See Also:

    • CT_R16G16_FLOAT

      public static final int CT_R16G16_FLOAT
      Aliases.
      See Also:

    • CT_R16G16B16A16_UNORM

      public static final int CT_R16G16B16A16_UNORM
      Aliases.
      See Also:

    • CT_COUNT

      @Internal public static final int CT_COUNT
      See Also:

  • Method Details

    • colorTypeCount

      public static int colorTypeCount()
      Returns the number of color types but avoids inlining at compile-time.

    • bytesPerPixel

      public static int bytesPerPixel(int ct)
      Returns the number of bytes required to store a pixel.
      Returns:
      bytes per pixel

    • maxBitsPerChannel

      public static int maxBitsPerChannel(int ct)

    • validateAlphaType

      public static int validateAlphaType(int ct, int at)
      Returns a valid AlphaType for ct. If there is more than one valid AlphaType, returns at, if valid.
      Returns:
      a valid AlphaType
      Throws:
      IllegalArgumentException - at is unknown, ct is not unknown, and ct has alpha channel.

    • colorTypeChannelFlags

      @Internal public static int colorTypeChannelFlags(int ct)

    • colorTypeIsAlphaOnly

      public static boolean colorTypeIsAlphaOnly(int ct)

    • colorTypeToString

      public static String colorTypeToString(int ct)