NeXT 2-bit RLE
From Just Solve the File Format Problem
(Difference between revisions)
(Updated link) |
|||
| Line 25: | Line 25: | ||
== Software == | == Software == | ||
| − | * | + | * {{Libtiff}} (read-only) |
[[Category:Graphics]] | [[Category:Graphics]] | ||
[[Category:TIFF]] | [[Category:TIFF]] | ||
Latest revision as of 12:25, 18 September 2016
NeXT 2-bit RLE is an image compression scheme that can be used in TIFF files, using TIFF compression code 32766. It uses a pixel-oriented run-length encoding scheme to compress 4-level grayscale images.
It is extremely rare at best, and we have not located much information about it. Everything we know about it comes from libtiff.
[edit] Format
The information here is derived from libtiff's decoder, and may be wrong or incomplete.
Each row of the image uses one of three formats, as indicated by the first byte of encoded data corresponding to that row:
| Initial byte | Encoding scheme | Description |
|---|---|---|
| 0x00 | Literal row | The row is uncompressed. The uncompressed pixel data starts after the initial 0x00 byte. Four pixels are packed into each byte. |
| 0x40 | Literal span | The initial byte is followed by two bytes representing a byte offset (as a big-endian integer), and two bytes representing a byte count. Following that are byte count bytes of uncompressed pixel data, which are to be interpreted as beginning at byte offset (assuming four pixels per byte). Any pixels in the row that are not in this range are assigned a gray value of 3 (white). |
| Any other | RLE | For each byte (including the initial byte), the two highest bits represent the gray value (from 0 to 3), and the six lowest bits represent the run length (from 0 to 63). |
[edit] Software
- LibTIFF (read-only)
