Home > Archive > Compression > August 2007 > LZPM - new ROLZ-based compressor
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
LZPM - new ROLZ-based compressor
|
|
| encode 2007-08-06, 6:56 pm |
| Let me introduce a new version of the LZPM file compressor.
LZPM uses a ROLZ-driven compression and represents some sort of an
improved QUAD.
The engine is highly asymmetric in both terms of speed and memory
usage:
~272 MB for compression (17*N)
~20 MB for decompression
The decompression is pretty fast - much like LZMA.
The difference from a baseline LZ77 is that ROLZ (Reduced Offset LZ)-
based compressors "reduce" active offsets by current N-order context.
For string searching LZPM uses a hash chaining, much like Deflate. It
also uses an optimized parsing - a variation of the Flexible Parsing
scheme.
In addition, LZPM has a huge files support (64-bit file sizes).
Link:
http://www.encode.ru/downloads/lzpm007.zip (47 KB)
http://www.encode.ru/lzpm/lzpm.htm
-- Ilia Muraviev
| |
| Jim Leonard 2007-08-07, 6:56 pm |
| On Aug 6, 12:52 pm, encode <enc...@narod.ru> wrote:
> The difference from a baseline LZ77 is that ROLZ (Reduced Offset LZ)-
> based compressors "reduce" active offsets by current N-order context.
How does this differ from LZP? (more info on LZP here:
http://www.cbloom.com/papers/lzp.html)
|
|
|
|
|