This structure describes format of a floating point bitmap.
The difference between chunky and planar layout lies in the way
how color channels are interleaved. For a chunky format, all
channel data for a single pixel lies consecutively in memory. For a
planar layout, the first channel of all pixel is stored
consecutive, followed by the second channel, and so forth.
The color components are stored in the half-float format,
i.e. in a 16 bit value, with 5 bit exponent, 10 bit mantissa
and a sign bit. The bits of each value are stored consecutive
in memory.
The color components are stored in the half-float format,
i.e. in a 16 bit value, with 5 bit exponent, 10 bit mantissa
and a sign bit. For multi-channel Bitmaps, the pixel of the
first color channel are stored consecutively, follow by the
second channel and so forth.
The color components are stored in the IEEE single-precision
floating point format, i.e. in a 32 bit value, with 8 bit
exponent, 23 bit mantissa and a sign bit. The bits of each
value are stored consecutive in memory.
The color components are stored in the IEEE single-precision
floating point format, i.e. in a 32 bit value, with 8 bit
exponent, 23 bit mantissa and a sign bit. For multi-channel
Bitmaps, the pixel of the first color channel are stored
consecutively, follow by the second channel and so forth.
The color components are stored in the IEEE double-precision
floating point format, i.e. in a 64 bit value, with 16 bit
exponent, 47 bit mantissa and a sign bit. The bits of each
value are stored consecutive in memory.
The color components are stored in the IEEE double-precision
floating point format, i.e. in a 64 bit value, with 16 bit
exponent, 47 bit mantissa and a sign bit. For multi-channel
Bitmaps, the pixel of the first color channel are stored
consecutively, follow by the second channel and so forth.