The 7-Zip Plug-In supports some advanced compression parameters, which are explained below.
Parameter |
Meaning |
Fast Bytes |
Sets the number of fast bytes. It can be in the range from 3 to 255. Bigger numbers provide slightly better compression strength at the expense of longer compression time. A bigger value of the fast bytes parameter can significantly increase compression ratio when files being compressed contain long identical sequences of bytes. |
Number of Passes |
Sets the number of passes. It can be in the range from 1 to 4. Bigger
numbers provide slightly better compression strength at the expense of
longer compression time. |
Deflate is the most popular
compression algorithm, originally introduced by PK-Zip in 1992. Deflate was a very good compression
algorithm among its contemporaries, however advances in compression technology
have rendered its compression ratios feeble.
Parameter |
Meaning | ||||||||||||||||||||||||
Dictionary Size |
Sets the size of the dictionary. The maximum value for the dictionary size is 256MB = 2^28 bytes. A larger dictionary provides for significantly enhanced compression ratios, at the expense of compression speed. For decompressing a file compressed using the LZMA algorithm with dictionary size N you need about N*2 memory available. | ||||||||||||||||||||||||
Match Finder Method
|
Sets the match finder method. Methods from the Binary Tree group require less memory than methods from the Patricia Tree group. Typically, Binary Tree with 2-3-4 bytes hashing works faster than any Patricia Tree group member, but for some types of files Patricia Tree group members may work faster. Algorithms from the Hash Chain group do not provide good compression ratios. Memory requirements for compression depend on the dictionary size.
| ||||||||||||||||||||||||
Fast Bytes |
Sets the number of fast bytes. It can be in the range from 3 to 255. Bigger numbers provide slightly better compression strength at the expense of longer compression time. A bigger value of the fast bytes parameter can significantly increase compression ratio when files being compressed contain long identical sequences of bytes. |
The LZMA algorithm is a derivative
of the Lempel-Ziv algorithm. It provides very fast decompression (about
10-20 times faster than compression), and the strongest compression available
today!
Parameter |
Meaning |
Memory Size |
Sets the size of memory usable by PPMd. More memory results in better compression. PPMd uses the same amount of memory for compression and decompression. The maximum value for the memory size is 2GB = 2^31 bytes. |
Model Order |
Sets the model order for PPMd. It can be in the range from 2 to 32. |
PPMd is a PPM based algorithm. This algorithm is mostly based on Dmitry Shkarin's PPMdH sources. PPMd provides very good compression ratios for plain text files.
Finally, you may also use a converter algorithm for 7Z archives. Converter algorithms pre-process files to be compressed, increasing their compressibility. Two converters are available:
BCJ2 (32-bit x86 executable CALL, JUMP, JCC converter [version 2])
BCJ (32-bit x86 executable CALL, JUMP converter)