PackBits
m (→Specifications) |
|||
| Line 7: | Line 7: | ||
It is byte-oriented (as opposed to pixel-oriented), so depending on how the image data is organized, it may be ineffective for images that use more than one byte per pixel. | It is byte-oriented (as opposed to pixel-oriented), so depending on how the image data is organized, it may be ineffective for images that use more than one byte per pixel. | ||
| − | == Format == | + | == Format details == |
The compressed data stream consists of a code byte, followed by zero or more data bytes, followed by another code byte, and so on. | The compressed data stream consists of a code byte, followed by zero or more data bytes, followed by another code byte, and so on. | ||
| Line 32: | Line 32: | ||
The reserved code byte (128 or −128) should be ignored, unless a format specification says otherwise. A few formats use it to mark the end of the compressed data. | The reserved code byte (128 or −128) should be ignored, unless a format specification says otherwise. A few formats use it to mark the end of the compressed data. | ||
| + | |||
| + | == Variants == | ||
| + | PackBits can easily be modified such that a unit of decompressed data is some fixed number of bytes greater than one. Formats that may use such variants include [[PICT]], [[Palm bitmap]], and [[IFF-DEEP]]. | ||
== Related formats == | == Related formats == | ||
Latest revision as of 15:10, 9 May 2024
PackBits (or Apple Macintosh PackBits, or Macintosh RLE) is a form of run-length encoding often used to compress image data.
It is byte-oriented (as opposed to pixel-oriented), so depending on how the image data is organized, it may be ineffective for images that use more than one byte per pixel.
Contents |
[edit] Format details
The compressed data stream consists of a code byte, followed by zero or more data bytes, followed by another code byte, and so on.
| Code byte (N) | Instructions |
|---|---|
| 0 ≤ N ≤ 127 | Interpret the next N+1 bytes literally. |
| N = 128 | Reserved |
| 129 ≤ N ≤ 255 | Repeat the next byte 257−N times. |
For whatever reason, PackBits is usually described in terms of signed bytes, which leads to the equivalent definition:
| N = −128 | Reserved |
| −127 ≤ N ≤ −1 | Repeat the next byte 1−N times. |
| 0 ≤ N ≤ 127 | Interpret the next N+1 bytes literally. |
The reserved code byte (128 or −128) should be ignored, unless a format specification says otherwise. A few formats use it to mark the end of the compressed data.
[edit] Variants
PackBits can easily be modified such that a unit of decompressed data is some fixed number of bytes greater than one. Formats that may use such variants include PICT, Palm bitmap, and IFF-DEEP.
[edit] Related formats
Formats that may use PackBits compression include MacPaint, PICT, TIFF, ILBM, and DEGAS.
[edit] Specifications
- TIFF 6.0 Specification, Section 9: PackBits Compression
- Technical Note TN1023: Understanding PackBits
- AtariForumWiki: PackBits Compression Algorithm
