Results 1 to 10 of 10

Thread: PAQ8L

  1. #1
    Expert
    Matt Mahoney's Avatar
    Join Date
    May 2008
    Location
    Melbourne, Florida, USA
    Posts
    3,255
    Thanks
    306
    Thanked 779 Times in 486 Posts
    I have a minor PAQ update, PAQ8L
    http://cs.fit.edu/~mmahoney/compression/
    (source code and Win32 exe)

    It should improve slightly on paq8jd and paq8k but without the slow speed of paq8k. I added a DMC model to paq8jd and removed some redundant sparse models. It uses a little less memory than paq8jd with about the same speed.

  2. #2
    Programmer
    Join Date
    Feb 2007
    Location
    Germany
    Posts
    420
    Thanks
    28
    Thanked 153 Times in 18 Posts
    Hi Matt!
    Quote Originally Posted by Matt Mahoney
    I added a DMC model to paq8jd and removed some redundant sparse models. It uses a little less memory than paq8jd with about the same speed.
    This is great!
    Besides text compression (HPx) Im wondering what changes will be next because PAQ already has so many powerful models.

  3. #3
    Tester
    Nania Francesco's Avatar
    Join Date
    May 2008
    Location
    Italy
    Posts
    1,565
    Thanks
    220
    Thanked 146 Times in 83 Posts
    Great Matt! Optimal job!

  4. #4
    Member
    Join Date
    Dec 2006
    Posts
    611
    Thanks
    0
    Thanked 1 Time in 1 Post
    Thanks Matt! It runs at 8kB/s, which is great, but it's worse than PAQ8JD by about 0,06%

  5. #5
    Member
    Join Date
    Mar 2007
    Location
    Ghent
    Posts
    56
    Thanks
    0
    Thanked 0 Times in 0 Posts
    Tested PAQ 8l on my benchmark site:
    http://uclc.info/

    Back up to speed of 8jd and improved results, only loss on audio it seems compared to 8jd.

  6. #6
    Moderator

    Join Date
    May 2008
    Location
    Tristan da Cunha
    Posts
    2,034
    Thanks
    0
    Thanked 4 Times in 4 Posts
    Thanks Matt!

  7. #7
    Member
    Join Date
    Dec 2006
    Posts
    611
    Thanks
    0
    Thanked 1 Time in 1 Post
    Quote Originally Posted by Black_Fox
    but its worse than PAQ8JD by about 0,06%
    Well, it can be actually littlier. -7 switch unexpectedly took only 620MB

  8. #8
    Member
    Join Date
    Mar 2007
    Location
    Ghent
    Posts
    56
    Thanks
    0
    Thanked 0 Times in 0 Posts
    It has 2 new records compared to 8jd (8k is too slow to really compare to):
    http://uclc.info/multilingual_text_compression_test.htm
    1st non-dictionary emtry

    http://uclc.info/ut2004_map_compression_test.htm

    Top spot!

    I surely wouldn't say it's worse.
    It only is noticeably worse than 8jd on audio data:
    http://uclc.info/lossless_audio_compression_test.h tm

  9. #9
    Member
    Join Date
    Dec 2006
    Posts
    611
    Thanks
    0
    Thanked 1 Time in 1 Post
    Quote Originally Posted by JohanDeBock
    I surely wouldnt say its worse.
    I only mean in my benchmark (about 3kB behind 8jd). Also in my test it holds 1st place for 1 file.

  10. #10
    Programmer giorgiotani's Avatar
    Join Date
    May 2008
    Location
    Italy
    Posts
    166
    Thanks
    3
    Thanked 2 Times in 2 Posts
    This update is interesting, and as for I can see the syntax is compatible with PAQ8F so it's possible to use it in place of 8F or 8JD from current PeaZip frontend.
    I think I'll add full support to it in upcoming PeaZip 1.5!

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •