lame 3.26beta will soon be on the web page.

Robert's new ATH (absolute threshold of hearing) and an analog silence
removal are on by default, with the baseline ATH at 3.3kHz coming from
that tuning I described in a previous message.
I haven't been able to test any of this because i'm having some
technical problems on my home pc.  I've only done a quick check at
work to make sure the new code doesn't sound horrible.

I also added Greg's -X4 code, but only to quant_compare().
quant_compare() is the routine which compares two quantizations
with the same number of bits and decides which is the best.  
With VBR, there is one other step:  Setting the tolerances 
used by VBR_compare(), and this still needs work. 

To explain this better: the #define RH VBR routine calls outer_loop()
with different numbers of bits until it finds the minimum number of
bits which produce an acceptable quantization, where acceptable means
all the noise measures are less than the prescribed tolerances.  Each
call to outer_loop() computes many different quantizations with a
fixed number of bits, and takes the best one using the quant_compare()
routine (which has the new -X4 code).






Here's a list of all the new features:



      The "-m m" option (mono .mp3 file) will automatically mix left and
      right channels if the input file is stereo.  (Alfred Weyers)

      New quant_compare algorithm (method for deciding which of two
      quantizations is better) enabled with -X4 (Greg Maxwell)

      New mid/side VBR bit allocation formula.  Mid channel bits are
      set by the quality requirements, and then the side channel uses
      a reduced number of bits (in a proportion coming from the fixed
      bitrate code).  This might not be optimal, but it should be
      pretty good and no one knows what the optimal solution should
      be.  (Greg Maxwell)

      New VBR (#define RH) tunings based on detailed listening tests
      by Macik and Greg Maxwell.

      Sigbj�rn Skj�ret fixed several compiler warnings (which turned
      out to be potential bugs)

      Takehiro Tominaga fixed a low bitrate bug in reduce_side()

      Alfred Weyers fixed some buffer overflows.

      New ATH (absolute threshold of hearing) formula replaces buggy
      ISO code, and adds analog silence treatment (removal of
      coefficients below below ATH).  These are turned on by default
      but have not been fully tested.  (Robert Hegemann)

      Bug in short block spreading function fixed.  (Robert Hegemann) 

--
MP3 ENCODER mailing list ( http://geek.rcc.se/mp3encoder/ )

Reply via email to