Results 1 to 6 of 6

Thread: MM v1.1

  1. #1
    Member Skymmer's Avatar
    Join Date
    Mar 2009
    Location
    Russia
    Posts
    681
    Thanks
    37
    Thanked 168 Times in 84 Posts

    MM v1.1

    Since this is my first post here I would like to say Hello ! to everybody. Realy nice forum
    But back to deal. I recently played with MM preprocessor by Bulat Ziganshin. More exactly speaking with v1.1 (2007-09-11) and found it ... well ... just great !
    You can find more info at this blog and download mentioned version here.
    Here is a brief test conducted on Nuyorican Soul - I Am The Black Gold of The Sun (4 Hero Remix) converted from APE (Lossless audio compressor) to WAV. Original WAV size is 90 107 636 bytes. Then I processed this file with MM with no options and with MM -r1\MM -r2. After that resulting files have been compressed with 7z v4.65 Ultra (non-solid).

    Code:
    Test.wav        79 946 884
    Test.mm         65 915 812
    Test_r1.mm      62 937 513
    Test_r2.mm      65 915 812
    -r1 gives 21.3% better compression over original WAV file and 4.5% better compression over default mode. -r2 size is similar to default just because only one first byte differ in resulting files. But as I see both -r1 and -r2 modes are unfinished (decompression just don't work) so I would like to ask the question to Bulat. Are there any chances to see new completed version of MM and include it to FreeARC ?
    Anyway, respect for the work you're doing
    Last edited by Skymmer; 31st March 2009 at 02:36.

  2. #2
    Tester
    Nania Francesco's Avatar
    Join Date
    May 2008
    Location
    Italy
    Posts
    1,565
    Thanks
    220
    Thanked 146 Times in 83 Posts

    Hi Skymmer!

    Good Work! Thanks!

  3. #3
    Programmer Bulat Ziganshin's Avatar
    Join Date
    Mar 2007
    Location
    Uzbekistan
    Posts
    4,497
    Thanks
    733
    Thanked 659 Times in 354 Posts
    1. we can't rely on one test to decide whether one mode (or program) is better than other. one can do a lot of tests with variyng data before making conclusion

    2. for wav compression, freearc includes TTA algorithm which is much stronger

    3. i don't have plans to finish -r1/r2 implementation now, but this should be rather easy

  4. #4
    Member Skymmer's Avatar
    Join Date
    Mar 2009
    Location
    Russia
    Posts
    681
    Thanks
    37
    Thanked 168 Times in 84 Posts
    Bulat, thanx 4 answering.

    Quote Originally Posted by Bulat Ziganshin View Post
    1. we can't rely on one test to decide whether one mode (or program) is better than other. one can do a lot of tests with variyng data before making conclusion

    2. for wav compression, freearc includes TTA algorithm which is much stronger

    3. i don't have plans to finish -r1/r2 implementation now, but this should be rather easy
    1. You're completely right. But excuse me, I didn't make any conclusion. I just showed up how usefull MM can be. On terrabyte set it surely can be better or worse.

    2. I know about TTA algo in FreeARC but I'm talking about MM behaviour in general, not only on pure PCM WAV files. Here is another test.
    PC_Eng.str file from Hitman: Blood Money game. It's natural container which contains different formats but mainly RAW PCM data in uncompressed form. TTA will just refuse it. Original size is 1 877 041 133
    I used MM 4378+2*16 and then FreeARC with
    arc a -r -ep1 -lc- -di+$ -ma- -s- -mlzma:a2:d64m:h64m:lc3:fb64:mfbt4
    and got
    Code:
    PC_Eng.str            1.438.882.106
    PC_Eng_Default.mm     1.251.700.867
    PC_Eng_r1.mm          1.211.539.907
    3. It's pity. Especially considering the fact it's easy

  5. #5
    Member Skymmer's Avatar
    Join Date
    Mar 2009
    Location
    Russia
    Posts
    681
    Thanks
    37
    Thanked 168 Times in 84 Posts
    Quote Originally Posted by Skymmer View Post
    TTA will just refuse it.
    I was terribly wrong writing it
    FA successfully done it with tta:m3: o4378:2*16 and the resulting size of 1 175 649 039 bytes. Cool ! But I wonder where it comes from. I mean if the original size is 1 877 041 133, offset is 4378, we have 1 877 036 755 bytes of data to work on. Considering the 2*16 scheme, we have 4 bytes blocks so we have 3 byte tail. TTA can't deal with RAW data, at least originally and the only lossless tool which made it for me (among others with RAW support) was OptimFROG, so I think its FA's TTA implementation, isn't it?
    Damn, I start to like FA more and more
    Last edited by Skymmer; 28th April 2009 at 05:44.

  6. #6
    Programmer Bulat Ziganshin's Avatar
    Join Date
    Mar 2007
    Location
    Uzbekistan
    Posts
    4,497
    Thanks
    733
    Thanked 659 Times in 354 Posts
    original tta deals only wioth proper WAV files. fa compression method should work with any input data, for example when you added a few bytes to WAV manually. so it may copy some bytes at the end intact

Posting Permissions

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