history.html

音频编码

<h3> LAME 3.26beta&nbsp; September 10 1999</h3>
<ul>
  <li> The "-m m" option (mono .mp3 file) will automatically mix left and right 
    channels if the input file is stereo.&nbsp; (Alfred Weyers)</li>
  <li> <font color="#ff0000">New quant_compare algorithm (method for deciding 
    which of two quantizations is better) enabled with -X4 (Greg Maxwell)</font></li>
  <li> <font color="#ff0000">New mid/side VBR bit allocation formula.&nbsp; 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).&nbsp; This might not be optimal, but it should be pretty good and no 
    one knows what the optimal solution should be.&nbsp; (Greg Maxwell)</font></li>
  <li> <font color="#ff0000">New VBR (#define RH) tunings based on detailed listening 
    tests by Macik and Greg Maxwell.</font></li>
  <li> Sigbj&oslash;rn Skj&aelig;ret fixed several compiler warnings (which turned 
    out to be potential bugs)</li>
  <li> Takehiro Tominaga fixed a low bitrate bug in reduce_side()</li>
  <li> Alfred Weyers fixed some buffer overflows.</li>
  <li> <font color="#ff0000">New ATH (absolute threshold of hearing) formula replaces 
    buggy ISO code, and adds&nbsp; analog silence treatment&nbsp; (removal of 
    coefficients below below ATH).&nbsp;&nbsp; These are turned on by default 
    but have not been fully tested.&nbsp; (Robert Hegemann)</font></li>
  <li> <font color="#ff0000">Bug in short block spreading function fixed.&nbsp; 
    (Robert Hegemann)</font></li>
</ul>
<h3> LAME 3.25beta&nbsp; August 22 1999</h3>
<ul>
  <li> Sigbj&oslash;rn Skj&aelig;ret fixed a zero byte malloc call.&nbsp; This 
    bug&nbsp; was introduced in 3.24 and causes problems on non Linux systems.</li>
  <li> Bit allocation routines would sometimes allocate more than 4095 bits to 
    one channel of one granule.&nbsp; A couple of people reported problems that 
    might be caused by this, especially at higher bitrates.</li>
  <li> Nils Faerber updated the man page and fixed many of the compiler warnings.</li>
  <br>
  &nbsp; 
</ul>
<h3> LAME 3.24beta&nbsp;&nbsp; August 15 1999</h3>
<ul>
  <li> This release contains the following new code (for developers) which is 
    disabled by default:</li>
  <li> Robert Hegemann:&nbsp; Completely overhauled VBR code.&nbsp; Now computes 
    exact number of bits required for the given qualty and then quantized with 
    the appropriate bitrate.</li>
  <li> Several new quantization quality measures.</li>
</ul>
<h3> LAME 3.23beta&nbsp; August 8 1999</h3>
<ul>
  <li> Very nice continuously updated VBR histogram display from Iwasa Kazmi.&nbsp; 
    (disabled with --nohist).</li>
  <li> More modulerization work.&nbsp; The encoding engine can now be compiled 
    into libmp3lame, but the interface is awkward.</li>
  <li> <font color="#ff0000">Bug fixed in FFT Hann window formula (Leonid A. Kulakov).</font></li>
  <li> New LAME logo on the download page.&nbsp; Created by Chris Michalisles.</li>
  <li> <font color="#ff0000">Several VBR algorithm improvements from Robert Hegemann.&nbsp; 
    New quantization noise metrics and VBR quality measure takes into account 
    mid/side encoding.&nbsp; Should produce smaller files with the same quality, 
    especially when using jstereo.</font></li>
</ul>
<h3> LAME 3.22beta&nbsp; July 27 1999</h3>
<ul>
  <li> Downsampling (stereo to mono) bug with MPEG2 fixed.&nbsp; (Mike Oliphant)</li>
  <li> Downsampling now merges L &amp; R channels - before it only took the L 
    channel.</li>
  <li> More modularization and code cleanup from Albert Faber and myself.</li>
  <li> Input filesize limit removed for raw pcm input files.&nbsp; For other file 
    types, LAME will still only read the first 2^32 samples, (27 hours of playing 
    time at 44.1 kHz).</li>
  <br>
  &nbsp; 
</ul>
<h3> LAME 3.21beta&nbsp; July 26 1999</h3>
<ul>
  <li> <font color="#ff0000">Correct Mid/Side masking thresholds for JSTEREO mode!&nbsp; 
    This is enabled with -h.&nbsp; It&nbsp; makes LAME about 20% slower since 
    it computes psycho-acoustics for L,R Mid and Side channels.</font></li>
  <li> <font color="#ff0000">"Analog silence" threshold added.&nbsp; Keeps VBR 
    from upping the bitrate during very quite passages.&nbsp; (Robert.Hegemann)</font></li>
  <li> <font color="#ff0000">New VBR quality setting from Robert Hegemann.&nbsp; 
    It is based on the idea that distortion at lower bit rates sounds worse than 
    at higher bitrates, and so the allowed distortion (VBR quality setting) is 
    proportional to the bitrate.&nbsp; Because of this, default minimum bitrate 
    is now 32kbs.</font></li>
  <li> <font color="#ff0000">Expermental subblock gain code enabled with -Z.</font></li>
  <li> New "-r" option for raw pcm input files.&nbsp; With -r, LAME will not do 
    any fseek()'s or look for wav and aiff headers on the input file.</li>
  <li> Bug fixes in mp3x (frame analyzer) for viewing frames near end of the file.</li>
  <li> Bug fixed to allow setting the sampling rate of raw pcm input files.</li>
</ul>
<h3> LAME 3.20beta&nbsp; July 19 1999</h3>
<ul>
  <li> Bug in get_audio.c fixed.&nbsp; Libsndfile wrappers would not compile (Miguel 
    Revilla&nbsp; Rodriguez)</li>
  <li> Nils Faerber found some unitialized variables and some wierd extranous 
    computations in filter_subband, now fixed.&nbsp; This was causing seg faults 
    on some machines.</li>
  <br>
  &nbsp; 
</ul>
<h3> LAME 3.19beta&nbsp; July 18 1999</h3>
<ul>
  <li> <font color="#ff0000">Oops!&nbsp; Robert Hegemann immediatly found a bug 
    in the new&nbsp; (old -Z option) quantization code.&nbsp; calc_noise1 was 
    not returning tot_noise, so non ms-stereo frames were buggy.</font></li>
  <br>
  &nbsp; 
</ul>
<h3> LAME 3.18beta&nbsp; July 17 1999</h3>
<ul>
  <li> <font color="#ff0000">Many psycho-acoustic bug fixes.&nbsp; Dan Nelson 
    discovered a bug in MPEG2: For short blocks, the code assumes 42 partition 
    bands.&nbsp; MPEG1 sometimes has less, MPEG2 can have more.&nbsp; In MPEG1, 
    this bug would not have effected the output if your compiler initializes static 
    variables to 0 on creation.&nbsp; In MPEG2 it leads to array out-of-bounds 
    access errors. Finally, there was a related bug in MPEG1/MPEG2, short &amp; 
    long blocks where the energy above 16 kHz was all added to partition band 
    0. (the lowest frequeny partition band!)</font></li>
  <li> <font color="#ff0000">The -Z option (Gabriel Bouvigne's idea of using total 
    quantization noise to choose between two quantizations with the same value 
    of "over") is now the default.&nbsp; I believe this helps remove the trilling 
    sound in Jan's testsignal4.wav.&nbsp; The quality of testsignal2.wav and testsignal4.wav 
    are now better than Xing and getting closer to FhG.</font></li>
  <li> Bug fixes in frame &amp; sample count for downsampling mode. (ben "jacobs")</li>
  <li> Patches to improve modulization.&nbsp; (ben "jacobs")</li>
</ul>
<h3> LAME 3.17beta&nbsp; July 11 1999</h3>
<ul>
  <li> substantial code cleanup towards goal of making LAME more modular.</li>
</ul>
<h3> LAME 3.16beta&nbsp; July 11 1999</h3>
<ul>
  <li> <font color="#ff0000">New tunings of window switching, and better bit allocation 
    based on pe.&nbsp; (Jan Rafaj.&nbsp; improves both testsignal2.wav and testsignal4.wav).</font></li>
  <li> <font color="#ff0000">Bug in mid/side quantization when side channel was 
    zero fixed.&nbsp; (Albert Faber)</font></li>
  <li> Removed some extranous computations in l3psy.c (Robert Hegemann)</li>
  <li> More detailed timing status info, including hours display. (Sakari Ailus) 
    and percentage indicator (Conrad Sanderson).</li>
  <li> <font color="#3366ff">Window_subband and calc_noise1,calc_noise2 speedups.&nbsp; 
    Quantize_xrpow speedup should be significant on non GNU/intel systems. (Mike 
    Cheng)</font></li>
  <li> <font color="#3366ff">Better initial guess for VBR bitrate.&nbsp; Should 
    speed up VBR encoding.&nbsp; (Gabriel Bouvigne)</font></li>
  <li> More advanced .wav header parsing.&nbsp; fixes bugs involving click in 
    first frame. (Robert.Hegemann)</li>
  <li> Correct filesize and total frame computation when using LIBSNDFILE (ben 
    "jacobs")</li>
  <li> Click in last frame (buffering problem) when using libsndfile fixed.</li>
  <li> Audio I/O code overhauled.&nbsp; There is now a uniform audio i/o interface 
    to libsndfile or the LAME built in wav/aiff routines.&nbsp; All audio i/o 
    code localized to get_audio.c.</li>
  <br>
  &nbsp; 
</ul>
<h3> LAME 3.15beta</h3>
<ul>
  <li> times()/clock() problem fixed for non-unix OS.&nbsp; (Ben "Jacobs")</li>
  <li> Fixed uninitialized pe[] when using fast mode.&nbsp; (Ben "Jacobs")</li>
</ul>
<h3> LAME 3.13&nbsp; June 24 1999</h3>
<ul>
  <li> Patches for BeOS from Gertjan van Ratingen.</li>
  <li> Makefile info for OS/2 Warp 4.0&nbsp; (from dink.org).</li>
  <li> Status display now based on wall clock time, not cpu time.</li>
  <li> mem_alloc no longer allocates twice as much memory as needed (Jan Peman).</li>
</ul>
<h3> 3.12pre9</h3>
<ul>
  <li> Updated BLADEDLL code to handle recent changes (Albert Faber).</li>
  <li> Bug fixed in parsing options when not using GTK (Albert Faber).</li>
  <li> <font color="#ff0000">MPEG2 Layer III psycho acoustics now working.</font></li>
  <li> <font color="#3366ff">Improved huffman encoding Chris Matrakidis. (10% 
    faster).&nbsp; I dont know how he finds these improvements!&nbsp; LAME with 
    full quality now encodes faster than real time on my PII 266.</font></li>
  <li> Fixed time display when encoding takes more than 60 minutes.</li>
</ul>
<h3> 3.12pre8</h3>
<ul>
  <li> <font color="#ff0000">New <a href="gpsycho/ms_stereo.html">mid/side stereo</a> 
    criterion.&nbsp; LAME will use mid/side stereo only when the difference between 
    L &amp; R masking thresholds (averaged over all scalefactors) is less then 
    5db.&nbsp; In several test samples it does a very good job mimicking the FhG 
    encoder.</font></li>
  <li> <font color="#ff0000">Bug in mid/side stereo fixed:&nbsp; independent variation 
    of mid &amp; side channel scalefactors disabled.&nbsp; Because of the way 
    outer_loop is currently coded, when encoding mid/side coefficietns using left/right 
    thresholds, you have to vary the scalefactors simultaneously.</font></li>
  <li> <font color="#ff0000">Bug in side/mid energy ratio calculation fixed. (Thanks 
    to Robert Hegemann)</font></li>
  <li> Default mode is stereo (not jstereo) if bitrate is chosen as 192kbs or 
    higher.&nbsp; Tero Auvinen first pointed out that FhG seems to think at 160kbs, 
    their encoder is so good it doesn't need jstereo tricks. Since LAME is not 
    as good as FhG, I am going to claim that 192kbs LAME is so good it doens't 
    need jstereo tricks, and thus it is disabled by default.</li>
  <li> WAV header parsing for big-endian machines, and automatic detection of 
    big-endian machines.&nbsp; (Thanks to&nbsp; Sigbj&oslash;rn Skj&aelig;ret).</li>
  <li> added 56 sample delay to sync LAME with FhG.</li>
  <li> MP3x (frame analyzer) can now handle MPEG2 streams.</li>
</ul>
<h3> 3.12pre7</h3>
<ul>
  <li> MPEG2 layer III now working!&nbsp; lower bit rates (down to 8kbs) and 3 
    more sampling frequencies:&nbsp; 16000, 22050, 24000Hz. Quality is poor - 
    the psy-model does not yet work with these sampling frequencies.</li>
  <li> Fixed "ERROR: outer_loop(): huff_bits &lt; 0." bug when using VBR.</li>
  <li> bash and sh scripts to run LAME on multiple files now included.&nbsp; (from 
    Robert Hegemann and Gerhard Wesp respectively)</li>
  <li> bug fix in encoding times for longer files from&nbsp; (Alvaro Martinez 
    Echevarria)</li>
  <li> yet another segfault in the frame analyzer fixed.</li>
  <li> ISO psy-model/bit allocation routines removed.&nbsp; This allowed makeframe() 
    to be made much simpler, and most of the complicated buffering is now gone. 
    Eventually I would like the encoding engine to be a stand alone library.</li>
</ul>
<h3> 3.12pre6</h3>
<ul>
  <li> <font color="#ff0000">Better VBR tuning.&nbsp; Find minimum bitrate with 
    distortion less than the allows maximum.&nbsp; A minimum bit rate is imposed 
    on frames with short blocks (where the measured distortion can not be trusted).&nbsp;&nbsp; 
    A minimum frame bitrate can be specified with -b, default=64kbs.</font></li>
  <li> <a href="http://www.zip.com.au/%7Eerikd/libsndfile">LIBSNDFILE</a> support.&nbsp; 
    With libsndfile, LAME can encode almost all sound formats.&nbsp; Albert Faber 
    did the work for this, including getting libsndfile running under win32.</li>
  <li> CRC checksum now working!&nbsp;&nbsp; (Thanks to Johannes Overmann )</li>
  <li> frame analyzer will now work with mono .mp3 files</li>
  <li> <font color="#3366ff">more code tweeks from Jan Peman.</font></li>
  <li> <font color="#3366ff">Compaq-Alpha(Linux) fixes and speedups from Nils 
    Faerber.</font></li>
  <li> <font color="#3366ff">Faster bin_search_StepSize from Juha Laukala.</font></li>
  <li> <font color="#3366ff">Faster quantize() from Mike Cheng</font></li>
  <li> <font color="#3366ff">Faster quantize_xrpow() from Chris Matrakidis.&nbsp; 
    xrpow_flag removed since this option is now on by default.</font></li>
  <li> Fixed .wav header parsing from Nils Faerber.</li>
  <li> Xing VBR frame info header code from Albert Faber.&nbsp;&nbsp; "Xing" and 
    "LAME 3.12" embedded in first frame.</li>
  <li> <font color="#ff0000">Bug in VBR bit allocation based on "over" value fixed.</font></li>
</ul>
<h3> LAME 3.11&nbsp; June 3 1999</h3>
<ul>
  <li> Almost all warnings (-Wall) now fixed!&nbsp; (Thanks to Jan Peman)</li>
  <li> More coding improvements from Gabriel Bouvigne and Warren Toomey.</li>
  <li> <font color="#ff0000">VBR&nbsp; (variable bit rate).&nbsp;&nbsp; Increases 
    bit rate for short blocks and for frames where the number of bands containing 
    audible distortion is greater than a given value.&nbsp; Much tuning needs 
    to be done.</font></li>
  <li> Patch to remove all atan() calls from James Droppo.</li>

</ul>
<h3> LAME 3.10 May 30 1999</h3>
<ul>
  <li> <font color="#3366ff">Fast mode (-f)&nbsp; disables psycho-acoustic model 
    for real time encoding on older machines.&nbsp; Thanks to Lauri Ahonen who 
    first sent a patch for this.</font></li>
  <li> <font color="#ff0000">New bit reservoir usage scheme to accommodate the 
    new pr