Jump to content

Talk:lzop

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia


Move?

[edit]

Should this article be moved to "lzop" (well, "Lzop") ? While the compression format is named "LZO", the compression tool is named "lzop" by parallel with "gzip" and "bzip2". It is always referred to in lowercase on the official website, even at the beginnings of sentences.

Small Benchmark

[edit]

I did a fast comparison "lzop 1.01" vs "gzip -1" on a 173 MB text file:

  • gzip -1 .. -9, resulting file sizes are 53.5 .. 42.5 MB;
  • lzop -1 .. -6, resulting file sizes are 65.5 .. 66.0 MB;
  • (lzop -7 .. -9, resulting file sizes are ~48..47.0 MB)

More interesting, compression+decompression times are (Athlon 2300+):

  • lzop -1..-6: compress time 6.7-7.4 sec (25.9 MB/sec) ; decompression time ~1.3 sec (133.5 MB/sec);
  • gzip -1: compress time 12.9 sec (13.4 MB/sec) ; decompression time ~3.1 sec ( 56.0 MB/sec)

So, lzop is about 1.9 x faster on compress (vs gzip -1), and 2.38 x faster on decompress (vs gzip -1), for the faster modes (-1 .. -6, all about the same), at the cost of a 23% larger compressed image (vs gzip -1) in this case (53.5 MB (gzip), 66 MB(lzop)).

Note that lzop -7 .. -9 modes are very slow on compressing (even 2-3x slower than gzip -9 !!), but the filesize is still slightly worse than e.g. gzip -5 (average mode); decompressing speed is as good as for other levels (~ 7 sec). [so, if reasonable compress times are important for you, not only decompress times, don't use levels -7 .. -9 (consider gzip); if compression time does not matter, but fast decompression is important: use level -7 .. -9]

Note2: lzop modes -1..-6 and -7..-9 are similar (in resulting size); [-2..-6 are even exactly the same mode!] —Preceding unsigned comment added by 78.43.95.214 (talk) 08:53, 24 December 2008 (UTC)[reply]

Notability (or lack thereof)

[edit]

I fail to see how lzop itself is notable. LZO, the algorithm, yes, but lzop the utility? Not really. 76.119.157.207 (talk) 05:57, 1 August 2011 (UTC)[reply]