1159b3361Sopenharmony_ci<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" 2159b3361Sopenharmony_ci "http://www.w3.org/TR/html4/loose.dtd"> 3159b3361Sopenharmony_ci<html> 4159b3361Sopenharmony_ci <head> 5159b3361Sopenharmony_ci <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1"> 6159b3361Sopenharmony_ci <meta name="GENERATOR" content="Mozilla/4.7 [en] (X11; I; Linux 2.2.12-20 i686) [Netscape]"> 7159b3361Sopenharmony_ci <meta http-equiv="content-type" content="text/html; charset=ISO-8859-1"> 8159b3361Sopenharmony_ci <title>LAME Changelog</title> 9159b3361Sopenharmony_ci </head> 10159b3361Sopenharmony_ci <body style="color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);" alink="#bbbbbb" link="#888888" vlink="#555555"> 11159b3361Sopenharmony_ci <center> 12159b3361Sopenharmony_ci <h1>History</h1> 13159b3361Sopenharmony_ci </center> 14159b3361Sopenharmony_ci Starting with LAME 3.0: <br> 15159b3361Sopenharmony_ci <font color="#ff0000">red = features and bug fixes which affect quality</font> <br> 16159b3361Sopenharmony_ci <font color="#3366ff">blue = features and bug fixes which affect speed</font> <br> 17159b3361Sopenharmony_ci black = usability, portability, other 18159b3361Sopenharmony_ci <hr> 19159b3361Sopenharmony_ci <h3>LAME 3.100.1 under construction</h3> 20159b3361Sopenharmony_ci <ul> 21159b3361Sopenharmony_ci <li> 22159b3361Sopenharmony_ci Robert Hegemann 23159b3361Sopenharmony_ci <ul> 24159b3361Sopenharmony_ci <li>Patch submitted by KO Myung-Hun, patch ticket <i>[ #80 ] OS/2 patches</i> 25159b3361Sopenharmony_ci </li> 26159b3361Sopenharmony_ci <li>Patch submitted by Elio Blanca, patch ticket <i>[ #82 ] Take advantage of terminal width on printing file names</i> 27159b3361Sopenharmony_ci </li> 28159b3361Sopenharmony_ci <li><font color="#ff0000"> 29159b3361Sopenharmony_ci Bug fix for item <i>[ #496 ] A critical bug in init_xrpow_core_sse</i> 30159b3361Sopenharmony_ci </font> 31159b3361Sopenharmony_ci </li> 32159b3361Sopenharmony_ci <li><font color="#ff0000"> 33159b3361Sopenharmony_ci Bug fix for item <i>[ #500 ] Buffer overflow in encoder</i> 34159b3361Sopenharmony_ci </font> 35159b3361Sopenharmony_ci </li> 36159b3361Sopenharmony_ci <li><font color="#ff0000"> 37159b3361Sopenharmony_ci Bug fix for item <i>[ #501 ] Encoder: Assertion 'eov->bitrate_index <= cfg->vbr_max_bitrate_index' failed during ABR encoding</i> 38159b3361Sopenharmony_ci </font> 39159b3361Sopenharmony_ci </li> 40159b3361Sopenharmony_ci <li>Bug fix for item <i>[ #444 ] msacmdrv.h (structure packing alignment)</i> 41159b3361Sopenharmony_ci </li> 42159b3361Sopenharmony_ci </ul> 43159b3361Sopenharmony_ci </li> 44159b3361Sopenharmony_ci <li> 45159b3361Sopenharmony_ci Alexander Leidinger 46159b3361Sopenharmony_ci <ul> 47159b3361Sopenharmony_ci <li><font color="#3366ff">Add faster CRC routine. The speed improvement is very small for typical use cases, but may be valuable if a lot of independent encodes are running and/or for a very long time. Patch by Robert Kausch of the fre:ac project. 48159b3361Sopenharmony_ci </font> 49159b3361Sopenharmony_ci </li> 50159b3361Sopenharmony_ci <li><font color="#3366ff">Fix configure script glob-ranges matching of compiler versions. This may result in faster code if no compiler optimization flags are specified during the configure step. This is part of patch ticket <i>[ #491 ] lame 3.100 slower than 3.99.5</i> 51159b3361Sopenharmony_ci </font> 52159b3361Sopenharmony_ci </li> 53159b3361Sopenharmony_ci <li><font color="#3366ff">Disable Takehiros IEEE753 hack by default. On modern CPUs (anything more recent from AMD than hammer/k8; Intel: Core2, i3/5/7/9 and similar -- no idea about recent Atom/Pentium) it is a speed pessimization. Add a configure option for it so that it can be enabled on old CPUs. 54159b3361Sopenharmony_ci </font> 55159b3361Sopenharmony_ci </li> 56159b3361Sopenharmony_ci <li>Update to more recent autotools based scripts, this may or may not fix issues during the configure stage for less popular or more recent OS or architectures. 57159b3361Sopenharmony_ci </li> 58159b3361Sopenharmony_ci </ul> 59159b3361Sopenharmony_ci </li> 60159b3361Sopenharmony_ci </ul> 61159b3361Sopenharmony_ci <br> 62159b3361Sopenharmony_ci <h3>LAME 3.100 October 13 2017</h3> 63159b3361Sopenharmony_ci <ul> 64159b3361Sopenharmony_ci <li> 65159b3361Sopenharmony_ci Rogério Brito 66159b3361Sopenharmony_ci <ul> 67159b3361Sopenharmony_ci <li> 68159b3361Sopenharmony_ci Don't include the debian directory as one that is needed during 69159b3361Sopenharmony_ci builds. Patch taken from Debian's packaging of lame. 70159b3361Sopenharmony_ci </li> 71159b3361Sopenharmony_ci <li> 72159b3361Sopenharmony_ci Resurrect Owen Taylor's code dated from 97-11-3 to properly deal 73159b3361Sopenharmony_ci with GTK1. This was transplanted back from aclocal.m4 with a patch 74159b3361Sopenharmony_ci provided by Andres Mejia. This change makes it easy to regenerate 75159b3361Sopenharmony_ci autotools' files with a simple invocation of <code>autoconf -vfi</code>. 76159b3361Sopenharmony_ci </li> 77159b3361Sopenharmony_ci <li> 78159b3361Sopenharmony_ci Fix possible race condition causing build failures in libmp3lame. 79159b3361Sopenharmony_ci Discovered in automated builds by the Debian project with patch 80159b3361Sopenharmony_ci provided by Andres Mejia. 81159b3361Sopenharmony_ci </li> 82159b3361Sopenharmony_ci </ul> 83159b3361Sopenharmony_ci </li> 84159b3361Sopenharmony_ci <li> 85159b3361Sopenharmony_ci Robert Hegemann 86159b3361Sopenharmony_ci <ul> 87159b3361Sopenharmony_ci <li> 88159b3361Sopenharmony_ci Improved detection of MPEG audio data in RIFF WAVE files. 89159b3361Sopenharmony_ci Tracker item <i>[ 3545112 ] Invalid sampling detection</i> 90159b3361Sopenharmony_ci </li> 91159b3361Sopenharmony_ci <li> 92159b3361Sopenharmony_ci New switch <b>--gain <decibel></b>, range -20.0 to 93159b3361Sopenharmony_ci +12.0, a more convenient way to apply Gain adjustment in 94159b3361Sopenharmony_ci decibels, than the use of <i>--scale <factor></i>. 95159b3361Sopenharmony_ci </li> 96159b3361Sopenharmony_ci <li> 97159b3361Sopenharmony_ci Fix for tracker item <i>[ 3558466 ] Bug in path handling</i> 98159b3361Sopenharmony_ci </li> 99159b3361Sopenharmony_ci <li> 100159b3361Sopenharmony_ci Fix for tracker item <i>[ 3567844 ] problem with Tag 101159b3361Sopenharmony_ci genre</i> 102159b3361Sopenharmony_ci </li> 103159b3361Sopenharmony_ci <li> 104159b3361Sopenharmony_ci Fix for tracker item <i>[ 3565659 ] no progress 105159b3361Sopenharmony_ci indication with pipe input</i> 106159b3361Sopenharmony_ci </li> 107159b3361Sopenharmony_ci <li> 108159b3361Sopenharmony_ci Fix for tracker item <i>[ 3544957 ] scale (empty) silent 109159b3361Sopenharmony_ci encode without warning</i> 110159b3361Sopenharmony_ci </li> 111159b3361Sopenharmony_ci <li> 112159b3361Sopenharmony_ci Fix for tracker item <i>[ 3580176 ] environment variable 113159b3361Sopenharmony_ci LAMEOPT doesn't work anymore</i> 114159b3361Sopenharmony_ci </li> 115159b3361Sopenharmony_ci <li> 116159b3361Sopenharmony_ci Fix for tracker item <i>[ 3608583 ] input file name 117159b3361Sopenharmony_ci displayed with wrong character encoding (on windows 118159b3361Sopenharmony_ci console with CP_UTF8)</i> 119159b3361Sopenharmony_ci </li> 120159b3361Sopenharmony_ci <li> 121159b3361Sopenharmony_ci Fix for bug ticket <i>[ #447 ] Fix dereference NULL and 122159b3361Sopenharmony_ci Buffer not NULL terminated issues. Thanks to Surabhi 123159b3361Sopenharmony_ci Mishra</i> 124159b3361Sopenharmony_ci </li> 125159b3361Sopenharmony_ci <li> 126159b3361Sopenharmony_ci Fix for bug ticket <i>[ #445 ] dereference of a null 127159b3361Sopenharmony_ci pointer possible in loop. Thanks to Renu Tyagi</i> 128159b3361Sopenharmony_ci </li> 129159b3361Sopenharmony_ci <li> 130159b3361Sopenharmony_ci Fix for bug ticket <i>[ #449 ] Make sure functions with 131159b3361Sopenharmony_ci SSE instructions maintain their own properly aligned 132159b3361Sopenharmony_ci stack. Thanks to Fabian Greffrath</i> 133159b3361Sopenharmony_ci </li> 134159b3361Sopenharmony_ci <li> 135159b3361Sopenharmony_ci Fix for bug ticket <i>[ #458 ] Multiple Stack and Heap 136159b3361Sopenharmony_ci Corruptions from Malicious File. Thanks to Gareth Evans 137159b3361Sopenharmony_ci and Elio Blanca</i> 138159b3361Sopenharmony_ci </li> 139159b3361Sopenharmony_ci <li> 140159b3361Sopenharmony_ci Fix for bug ticket <i>[ #460 ] A division by zero 141159b3361Sopenharmony_ci vulnerability. Thanks to Wang Shiyang, Liu Bingchang</i> 142159b3361Sopenharmony_ci </li> 143159b3361Sopenharmony_ci <li> 144159b3361Sopenharmony_ci Fix for bug ticket <i>[ #461 ] CVE-2017-9410 145159b3361Sopenharmony_ci fill_buffer_resample function in libmp3lame/util.c 146159b3361Sopenharmony_ci heap-based buffer over-read and ap</i> 147159b3361Sopenharmony_ci </li> 148159b3361Sopenharmony_ci <li> 149159b3361Sopenharmony_ci Fix for bug ticket <i>[ #462 ] CVE-2017-9411 150159b3361Sopenharmony_ci fill_buffer_resample function in libmp3lame/util.c 151159b3361Sopenharmony_ci invalid memory read and application crash</i> 152159b3361Sopenharmony_ci </li> 153159b3361Sopenharmony_ci <li> 154159b3361Sopenharmony_ci Fix for bug ticket <i>[ #463 ] CVE-2017-9412 155159b3361Sopenharmony_ci unpack_read_samples function in frontend/get_audio.c 156159b3361Sopenharmony_ci invalid memory read and application crash</i> 157159b3361Sopenharmony_ci </li> 158159b3361Sopenharmony_ci <li> 159159b3361Sopenharmony_ci Fix for bug ticket <i>[ #434 ] clip detect scale 160159b3361Sopenharmony_ci suggestion unaware of scale input value</i> 161159b3361Sopenharmony_ci </li> 162159b3361Sopenharmony_ci <li> 163159b3361Sopenharmony_ci HIP decoder bug fixed: decoding mixed blocks of lower 164159b3361Sopenharmony_ci sample frequency Layer3 data resulted in internal buffer 165159b3361Sopenharmony_ci overflow (write). Thanks to Henri Salo</i> 166159b3361Sopenharmony_ci </li> 167159b3361Sopenharmony_ci </ul> 168159b3361Sopenharmony_ci </li> 169159b3361Sopenharmony_ci <li> 170159b3361Sopenharmony_ci Alexander Leidinger 171159b3361Sopenharmony_ci <ul> 172159b3361Sopenharmony_ci <li> 173159b3361Sopenharmony_ci Feature request, patch ticket <i>[ #27 ] Add 174159b3361Sopenharmony_ci lame_encode_buffer_interleaved_int() by Michael Fink</i> 175159b3361Sopenharmony_ci </li> 176159b3361Sopenharmony_ci </ul> 177159b3361Sopenharmony_ci </li> 178159b3361Sopenharmony_ci </ul> 179159b3361Sopenharmony_ci <br> 180159b3361Sopenharmony_ci <hr> 181159b3361Sopenharmony_ci <h3>LAME 3.99.5 February 28 2012</h3> 182159b3361Sopenharmony_ci <ul> 183159b3361Sopenharmony_ci <li> 184159b3361Sopenharmony_ci Rogério Brito 185159b3361Sopenharmony_ci <ul> 186159b3361Sopenharmony_ci <li> 187159b3361Sopenharmony_ci autotools: Fix compilation on alpha using proper ifdef 188159b3361Sopenharmony_ci guards. Thanks to Andres Mejia. 189159b3361Sopenharmony_ci </li> 190159b3361Sopenharmony_ci <li>Small correction of the documentation.</li> 191159b3361Sopenharmony_ci </ul> 192159b3361Sopenharmony_ci </li> 193159b3361Sopenharmony_ci <li> 194159b3361Sopenharmony_ci Robert Hegemann 195159b3361Sopenharmony_ci <ul> 196159b3361Sopenharmony_ci <li> 197159b3361Sopenharmony_ci Workaround for cygwin build problem. Tracker item 198159b3361Sopenharmony_ci <i>[ 3480692 ] mingw-w64 compilation error</i> 199159b3361Sopenharmony_ci </li> 200159b3361Sopenharmony_ci <li> 201159b3361Sopenharmony_ci Fix for a problem when LAME is compiled with <i>fast 202159b3361Sopenharmony_ci floating point math</i>. It could result in <i>bitrate 203159b3361Sopenharmony_ci bloat</i> with silent input data. 204159b3361Sopenharmony_ci </li> 205159b3361Sopenharmony_ci <li> 206159b3361Sopenharmony_ci <font color="#ff0000">Bug fix for tracker item <i>[ 207159b3361Sopenharmony_ci 3486753 ] Artifacts at the beginning of <b>decoded</b> 208159b3361Sopenharmony_ci file</i></font> 209159b3361Sopenharmony_ci </li> 210159b3361Sopenharmony_ci </ul> 211159b3361Sopenharmony_ci </li> 212159b3361Sopenharmony_ci </ul> 213159b3361Sopenharmony_ci <br> 214159b3361Sopenharmony_ci <h3>LAME 3.99.4 January 25 2012</h3> 215159b3361Sopenharmony_ci <ul> 216159b3361Sopenharmony_ci <li> 217159b3361Sopenharmony_ci Robert Hegemann 218159b3361Sopenharmony_ci <ul> 219159b3361Sopenharmony_ci <li> 220159b3361Sopenharmony_ci Fix for tracker item <i>[ 3475581 ] lame crashes at .w64 221159b3361Sopenharmony_ci input file</i> 222159b3361Sopenharmony_ci </li> 223159b3361Sopenharmony_ci <li> 224159b3361Sopenharmony_ci Addressing things brought to attention by tracker item 225159b3361Sopenharmony_ci <i>[ 3463197 ] 3.99.x problem WFED and PCST frames </i> 226159b3361Sopenharmony_ci <ul> 227159b3361Sopenharmony_ci <li> 228159b3361Sopenharmony_ci WFED and PCST frames can now be added, to tag 229159b3361Sopenharmony_ci podcasts iTunes recognizes</li> 230159b3361Sopenharmony_ci <li>USER frames are now supported</li> 231159b3361Sopenharmony_ci <li> 232159b3361Sopenharmony_ci COMM frames can now have a description, when 233159b3361Sopenharmony_ci passed via --tv "COMM=description=full text" 234159b3361Sopenharmony_ci </li> 235159b3361Sopenharmony_ci <li> 236159b3361Sopenharmony_ci possible divide-by-zero exception should be fixed 237159b3361Sopenharmony_ci </li> 238159b3361Sopenharmony_ci <li> 239159b3361Sopenharmony_ci adding malformed user-defined-frames could result 240159b3361Sopenharmony_ci in abnormal program termination, fixed 241159b3361Sopenharmony_ci </li> 242159b3361Sopenharmony_ci </ul> 243159b3361Sopenharmony_ci </li> 244159b3361Sopenharmony_ci </ul> 245159b3361Sopenharmony_ci </li> 246159b3361Sopenharmony_ci </ul> 247159b3361Sopenharmony_ci <br> 248159b3361Sopenharmony_ci <h3>LAME 3.99.3 November 26 2011</h3> 249159b3361Sopenharmony_ci <ul> 250159b3361Sopenharmony_ci <li> 251159b3361Sopenharmony_ci Robert Hegemann 252159b3361Sopenharmony_ci <ul> 253159b3361Sopenharmony_ci <li> 254159b3361Sopenharmony_ci Fix for tracker item <i>[ 3441349 ] --tg does not handle 255159b3361Sopenharmony_ci genre number when adding unicode tag</i> 256159b3361Sopenharmony_ci </li> 257159b3361Sopenharmony_ci </ul> 258159b3361Sopenharmony_ci </li> 259159b3361Sopenharmony_ci </ul> 260159b3361Sopenharmony_ci <br> 261159b3361Sopenharmony_ci <h3>LAME 3.99.2 November 18 2011</h3> 262159b3361Sopenharmony_ci <ul> 263159b3361Sopenharmony_ci <li> 264159b3361Sopenharmony_ci Robert Hegemann 265159b3361Sopenharmony_ci <ul> 266159b3361Sopenharmony_ci <li> 267159b3361Sopenharmony_ci Due to some bugs in 3rd party HW/SW decoders, those were not 268159b3361Sopenharmony_ci be able to make use of LAME's extended info tag anymore, 269159b3361Sopenharmony_ci resulting in problems with things like gapless playback. 270159b3361Sopenharmony_ci </li> 271159b3361Sopenharmony_ci </ul> 272159b3361Sopenharmony_ci </li> 273159b3361Sopenharmony_ci </ul> 274159b3361Sopenharmony_ci <br> 275159b3361Sopenharmony_ci <h3>LAME 3.99.1 November 5 2011</h3> 276159b3361Sopenharmony_ci <ul> 277159b3361Sopenharmony_ci <li> 278159b3361Sopenharmony_ci Robert Hegemann 279159b3361Sopenharmony_ci <ul> 280159b3361Sopenharmony_ci <li> 281159b3361Sopenharmony_ci Fixes for several issues with ID3v2 unicode tags, using 282159b3361Sopenharmony_ci Big-Endian text encodings. Because of several other 283159b3361Sopenharmony_ci software (like Windows Media Player), 284159b3361Sopenharmony_ci LAME writes Little-Endian unicode tags only.<br> 285159b3361Sopenharmony_ci Thanks to Taihei Monma, for reporting these issues. 286159b3361Sopenharmony_ci Tracker items: <i>[ 3431203, 3431222, 3431241 ]</i> 287159b3361Sopenharmony_ci </li> 288159b3361Sopenharmony_ci </ul> 289159b3361Sopenharmony_ci </li> 290159b3361Sopenharmony_ci </ul> 291159b3361Sopenharmony_ci <br> 292159b3361Sopenharmony_ci <h3>LAME 3.99 October 15 2011</h3> 293159b3361Sopenharmony_ci <ul> 294159b3361Sopenharmony_ci <li>3.99 beta 1 becomes 3.99</li> 295159b3361Sopenharmony_ci </ul> 296159b3361Sopenharmony_ci <br> 297159b3361Sopenharmony_ci <h3>LAME 3.99 beta 1 not yet released</h3> 298159b3361Sopenharmony_ci <ul> 299159b3361Sopenharmony_ci <li> 300159b3361Sopenharmony_ci Robert Hegemann 301159b3361Sopenharmony_ci <ul> 302159b3361Sopenharmony_ci <li> 303159b3361Sopenharmony_ci Fix for Bugtracker item <i>[ 3395813 ] 304159b3361Sopenharmony_ci id3tag_set_fieldvalue for URL link frames Broken</i> 305159b3361Sopenharmony_ci </li> 306159b3361Sopenharmony_ci <li> 307159b3361Sopenharmony_ci Fix for Bugtracker item <i>[ 3368977 ] Wave Parser 308159b3361Sopenharmony_ci doesn't pad chunks</i> 309159b3361Sopenharmony_ci </li> 310159b3361Sopenharmony_ci <li> 311159b3361Sopenharmony_ci Fix for Bugtracker item <i>[ 3034259 ] confusing fatal 312159b3361Sopenharmony_ci error: can't update LAME-tag frame!</i> 313159b3361Sopenharmony_ci </li> 314159b3361Sopenharmony_ci </ul> 315159b3361Sopenharmony_ci </li> 316159b3361Sopenharmony_ci </ul> 317159b3361Sopenharmony_ci <h3>LAME 3.99 beta 0 not officially released</h3> 318159b3361Sopenharmony_ci <ul> 319159b3361Sopenharmony_ci <li> 320159b3361Sopenharmony_ci Josep Maria Antolín Segura 321159b3361Sopenharmony_ci <ul> 322159b3361Sopenharmony_ci <li> 323159b3361Sopenharmony_ci Documentation work 324159b3361Sopenharmony_ci </li> 325159b3361Sopenharmony_ci </ul> 326159b3361Sopenharmony_ci </li> 327159b3361Sopenharmony_ci <li> 328159b3361Sopenharmony_ci Rogério Brito 329159b3361Sopenharmony_ci <ul> 330159b3361Sopenharmony_ci <li> 331159b3361Sopenharmony_ci Improve Debian packaging. 332159b3361Sopenharmony_ci <ul> 333159b3361Sopenharmony_ci <li>Include suggestions made by Fabian Greffrath</li> 334159b3361Sopenharmony_ci <li> 335159b3361Sopenharmony_ci Update version of the Debian package to reflect 336159b3361Sopenharmony_ci that it is taken from CVS. 337159b3361Sopenharmony_ci </li> 338159b3361Sopenharmony_ci <li>Get rid of a transitional package</li> 339159b3361Sopenharmony_ci <li>Much more to come...</li> 340159b3361Sopenharmony_ci </ul> 341159b3361Sopenharmony_ci <li>Minor fix to validate the history file correctly</li> 342159b3361Sopenharmony_ci </ul> 343159b3361Sopenharmony_ci </li> 344159b3361Sopenharmony_ci <li> 345159b3361Sopenharmony_ci Robert Hegemann 346159b3361Sopenharmony_ci <ul> 347159b3361Sopenharmony_ci <li> 348159b3361Sopenharmony_ci Porting of LAME <b>mp3rtp</b> program to Windows platform 349159b3361Sopenharmony_ci </li> 350159b3361Sopenharmony_ci <li> 351159b3361Sopenharmony_ci Fix: <font color="#ff0000">for free format mp3 above 320 352159b3361Sopenharmony_ci kbps, the maximum allowed bits was limited too far.</font> 353159b3361Sopenharmony_ci </li> 354159b3361Sopenharmony_ci <li> 355159b3361Sopenharmony_ci Changed behaviour: LAME replaces common suffixes by 356159b3361Sopenharmony_ci MP3/WAV when no output filename was given. For example, 357159b3361Sopenharmony_ci a simple <tt>lame x.wav</tt> will create output file 358159b3361Sopenharmony_ci <tt>x.mp3</tt> 359159b3361Sopenharmony_ci </li> 360159b3361Sopenharmony_ci <li> 361159b3361Sopenharmony_ci UNICODE support on Windows OS, file names and ID3v2 tags; 362159b3361Sopenharmony_ci related: Bugtracker item <i>[ 3052230 ] Lame.exe can't 363159b3361Sopenharmony_ci find files with non-ASCII names</i> 364159b3361Sopenharmony_ci </li> 365159b3361Sopenharmony_ci <li> 366159b3361Sopenharmony_ci Added support for WAVE FORMAT IEEE FLOAT input files 367159b3361Sopenharmony_ci </li> 368159b3361Sopenharmony_ci <li> 369159b3361Sopenharmony_ci New switch <b>--swap-channel</b> added, see tracker 370159b3361Sopenharmony_ci item <i>[ 1118412 ] RFE: option to swap L and R channels</i> 371159b3361Sopenharmony_ci </li> 372159b3361Sopenharmony_ci <li> 373159b3361Sopenharmony_ci New switches <b>-m l</b> and <b>-m r</b> to select either 374159b3361Sopenharmony_ci left or right channel for mono encoding. 375159b3361Sopenharmony_ci </li> 376159b3361Sopenharmony_ci <li> 377159b3361Sopenharmony_ci Feature request item <i>[ 3134258 ] Allow larger images 378159b3361Sopenharmony_ci with LAME (>128KB)</i> 379159b3361Sopenharmony_ci </li> 380159b3361Sopenharmony_ci <li> 381159b3361Sopenharmony_ci Fix for Bugtracker item <i>[ 2962223 ] Encoder delay 382159b3361Sopenharmony_ci and padding values ignored when re-encoding</i> 383159b3361Sopenharmony_ci </li> 384159b3361Sopenharmony_ci <li> 385159b3361Sopenharmony_ci Fix for Bugtracker item <i>[ 2986823 ] Error when 386159b3361Sopenharmony_ci building Lame with NASM support -still exists</i> 387159b3361Sopenharmony_ci </li> 388159b3361Sopenharmony_ci <li> 389159b3361Sopenharmony_ci Fix for Bugtracker item <i>[ 3125235 ] make -f 390159b3361Sopenharmony_ci Makefile.mingw32 fails</i> 391159b3361Sopenharmony_ci </li> 392159b3361Sopenharmony_ci <li> 393159b3361Sopenharmony_ci Fix for Bugtracker item <i>[ 3021935 ] ID3 tag TXXX 394159b3361Sopenharmony_ci cannot be used more than once</i> 395159b3361Sopenharmony_ci </li> 396159b3361Sopenharmony_ci <li> 397159b3361Sopenharmony_ci Fix for Bugtracker item <i>[ 3025801 ] gain_analysis.h 398159b3361Sopenharmony_ci needs int constants</i> 399159b3361Sopenharmony_ci </li> 400159b3361Sopenharmony_ci <li> 401159b3361Sopenharmony_ci Fix for Bugtracker item <i>[ 3277412 ] Segmentation 402159b3361Sopenharmony_ci fault at layer2.c:105 with --decode</i> 403159b3361Sopenharmony_ci </li> 404159b3361Sopenharmony_ci <li> 405159b3361Sopenharmony_ci Changes in new VBR code: 406159b3361Sopenharmony_ci <ul> 407159b3361Sopenharmony_ci <li><font color="#ff0000">tuning on PSY model</font></li> 408159b3361Sopenharmony_ci <li> 409159b3361Sopenharmony_ci <font color="#770000">tuning on VBR scale and 410159b3361Sopenharmony_ci resulting bitrates</font> 411159b3361Sopenharmony_ci </li> 412159b3361Sopenharmony_ci <li> 413159b3361Sopenharmony_ci <font color="#ff0000">Minor bug fix for sfb21 414159b3361Sopenharmony_ci encoding</font> 415159b3361Sopenharmony_ci </li> 416159b3361Sopenharmony_ci </ul> 417159b3361Sopenharmony_ci </li> 418159b3361Sopenharmony_ci <li> 419159b3361Sopenharmony_ci <font color="#ff0000">All encoding modes use the PSY 420159b3361Sopenharmony_ci model from new VBR code, addresses Bugtracker item <i>[ 421159b3361Sopenharmony_ci 3187397 ] Strange compression behavior</i></font> 422159b3361Sopenharmony_ci </li> 423159b3361Sopenharmony_ci <li> 424159b3361Sopenharmony_ci Fix for a possible out of bounds array access in HIP 425159b3361Sopenharmony_ci decoder, addresses Bugtracker item <i>[ 3091776 ] 426159b3361Sopenharmony_ci crash at III_dequantize_sample</i> 427159b3361Sopenharmony_ci </li> 428159b3361Sopenharmony_ci </ul> 429159b3361Sopenharmony_ci </li> 430159b3361Sopenharmony_ci </ul> 431159b3361Sopenharmony_ci <hr> 432159b3361Sopenharmony_ci <h3>LAME 3.98.4 March 22 2010</h3> 433159b3361Sopenharmony_ci <ul> 434159b3361Sopenharmony_ci <li> 435159b3361Sopenharmony_ci Joseph Flynn 436159b3361Sopenharmony_ci <ul> 437159b3361Sopenharmony_ci <li> 438159b3361Sopenharmony_ci Improvements for LAME DirectShow filter: 439159b3361Sopenharmony_ci <ul> 440159b3361Sopenharmony_ci <li>Added support for the DirectShow IAMStreamConfig Interface to the LAME encoder filter output pin. 441159b3361Sopenharmony_ci </li> 442159b3361Sopenharmony_ci <li>Modified the DirectShow filter registration section so that the LAME Encoder filter is correctly registered in the Audio Compressors filter category. This will allow third-party encoding applications using the DirectShow System Device Enumerator Interface to correctly detect the LAME encoder when querying the Audio Compressors filter group. 443159b3361Sopenharmony_ci </li> 444159b3361Sopenharmony_ci <li>Modified the filter registration information so that the MP3 audio subtype is correctly reported as being supported on the encoder output pin. This will allow third-party encoding applications using the DirectShow IFilterMapper2 Interface to recognize that the LAME encoder supports MP3 output. 445159b3361Sopenharmony_ci </li> 446159b3361Sopenharmony_ci <li>Altered the Filter Merit Value that was being used when the filter was registered so that it is now using the standard DirectShow compressor filter merit value of MERIT_DO_NOT_USE (0x200000). Previously, the filter was being registered using a value of MERIT_SW_COMPRESSOR (0x100000), which was at a lower priority (i.e. worse priority) than MERIT_DO_NOT_USE. This prevented the LAME Encoder filter from being selected for use by some third-party encoding applications. 447159b3361Sopenharmony_ci </li> 448159b3361Sopenharmony_ci <li>Added code to calculate the frame length of the audio frames used for the nBlockSize element of the WAVEFORMATEX output structure. Previously this value was simply hard-coded to 1. 449159b3361Sopenharmony_ci </li> 450159b3361Sopenharmony_ci </ul> 451159b3361Sopenharmony_ci </li> 452159b3361Sopenharmony_ci </ul> 453159b3361Sopenharmony_ci </li> 454159b3361Sopenharmony_ci <li> 455159b3361Sopenharmony_ci Robert Hegemann 456159b3361Sopenharmony_ci <ul> 457159b3361Sopenharmony_ci <li><font color="#ff0000">Fix for Bugtracker item <i>[ 2973877 ] A problem regarding the new drain code</i></font> 458159b3361Sopenharmony_ci </li> 459159b3361Sopenharmony_ci </ul> 460159b3361Sopenharmony_ci </li> 461159b3361Sopenharmony_ci </ul> 462159b3361Sopenharmony_ci <h3>LAME 3.98.3 February 27 2010</h3> 463159b3361Sopenharmony_ci <ul> 464159b3361Sopenharmony_ci <li> 465159b3361Sopenharmony_ci Rogério Brito: 466159b3361Sopenharmony_ci <ul> 467159b3361Sopenharmony_ci <li>Update the debian packaging for the new release. 468159b3361Sopenharmony_ci </li> 469159b3361Sopenharmony_ci </ul> 470159b3361Sopenharmony_ci </li> 471159b3361Sopenharmony_ci <li> 472159b3361Sopenharmony_ci Robert Hegemann 473159b3361Sopenharmony_ci <ul> 474159b3361Sopenharmony_ci <li>The <b>ignore-tag-errors</b> switch had no effect when embedding album art, fixed.</li> 475159b3361Sopenharmony_ci <li>Library API change: lame_decode functions are now obsolete but still present, please use hip_decode instead. 476159b3361Sopenharmony_ci The reason for this change is: lame_decode functions use a single global variable within the library 477159b3361Sopenharmony_ci to store decoder setup, hip_decode functions don't. 478159b3361Sopenharmony_ci The encoder now uses hip_decode internally and it is now possible to use <b>clipdetect</b> 479159b3361Sopenharmony_ci feature while reencoding mp3 to mp3. 480159b3361Sopenharmony_ci </li> 481159b3361Sopenharmony_ci <li>Workaround for FFMPEG bug, which uses to call lame_encode_flush more than once in a loop. 482159b3361Sopenharmony_ci </li> 483159b3361Sopenharmony_ci <li>Windows: program icon and version info added (when building with VC9) 484159b3361Sopenharmony_ci </li> 485159b3361Sopenharmony_ci <li>Fix for Bugtracker item <i>[ 2688413 ] lib name problem in Microsoft Visual Studio 6</i> 486159b3361Sopenharmony_ci </li> 487159b3361Sopenharmony_ci <li>Fix for Bugtracker items <i>[ 2051870, 2423650, 2928684 ] several small documentation issues</i> 488159b3361Sopenharmony_ci </li> 489159b3361Sopenharmony_ci <li>Fix for Bugtracker item <i>[ 2723518 ] resampling in 3.98 and 3.99alpha</i> 490159b3361Sopenharmony_ci </li> 491159b3361Sopenharmony_ci <li>Fix for Bugtracker item <i>[ 2891879 ] Because of Windows API change, there was a problem with the <b>prority</b> switch.</i> 492159b3361Sopenharmony_ci </li> 493159b3361Sopenharmony_ci <li>Fix for Bugtracker item <i>[ 2893101 ] Access Violation in BladeMP3EncDLL if UNICODE was defined.</i> 494159b3361Sopenharmony_ci </li> 495159b3361Sopenharmony_ci <li>Fix for Bugtracker item <i>[ 2887359 ] Wrong length in ID3v2 tag when num_samples isn't set</i> 496159b3361Sopenharmony_ci </li> 497159b3361Sopenharmony_ci <li>Fix for Bugtracker item <i>[ 2872590 ] LameTAG: "Music length" missmatch in LAME 3.98</i> 498159b3361Sopenharmony_ci </li> 499159b3361Sopenharmony_ci <li>Fix for Bugtracker item <i>[ 2824296 ] wrong enc_padding value in LAME 3.99a and 3.98.3 (from CVS)</i> 500159b3361Sopenharmony_ci </li> 501159b3361Sopenharmony_ci <li><font color="#ff0000">Revisiting the FhG decoder problem (FhG V1.5 build 50, ships with MS Windows): 502159b3361Sopenharmony_ci enabling the new-drain-code seems to solve that issue better, than restricting the buffer size (see below: 3.98 beta 1, May 16 2007).</font> 503159b3361Sopenharmony_ci </li> 504159b3361Sopenharmony_ci <li>Patch submitted by Bernhard Doebler, tracker item <i>[ 2807676 ] Error when building Lame with NASM support</i> 505159b3361Sopenharmony_ci </li> 506159b3361Sopenharmony_ci <li>Patch submitted by Mancuso Raffaele, tracker item <i>[ 2406420 ] compile lame_enc.dll under cygwin</i> 507159b3361Sopenharmony_ci </li> 508159b3361Sopenharmony_ci </ul> 509159b3361Sopenharmony_ci </li> 510159b3361Sopenharmony_ci </ul> 511159b3361Sopenharmony_ci <h3>LAME 3.98.2 September 22 2008</h3> 512159b3361Sopenharmony_ci <ul> 513159b3361Sopenharmony_ci <li> 514159b3361Sopenharmony_ci Robert Hegemann 515159b3361Sopenharmony_ci <ul> 516159b3361Sopenharmony_ci <li>Fix for Bugtracker item <i>[ 2123206 ] lame 3.98.1 segfaults with -h</i> 517159b3361Sopenharmony_ci </li> 518159b3361Sopenharmony_ci </ul> 519159b3361Sopenharmony_ci </li> 520159b3361Sopenharmony_ci </ul> 521159b3361Sopenharmony_ci <h3>LAME 3.98.1 September 21 2008</h3> 522159b3361Sopenharmony_ci <ul> 523159b3361Sopenharmony_ci <li> 524159b3361Sopenharmony_ci Rogério Brito: 525159b3361Sopenharmony_ci <ul> 526159b3361Sopenharmony_ci <li> 527159b3361Sopenharmony_ci More fixes for the abx tool for Unix systems: 528159b3361Sopenharmony_ci <ul> 529159b3361Sopenharmony_ci <li>Plugged a memory leak.</li> 530159b3361Sopenharmony_ci <li>Fixed an endianness problem: users of big-endian machines 531159b3361Sopenharmony_ci can now do abx tests. 532159b3361Sopenharmony_ci </li> 533159b3361Sopenharmony_ci </ul> 534159b3361Sopenharmony_ci </li> 535159b3361Sopenharmony_ci <li>Fixed history's HTML doctype</li> 536159b3361Sopenharmony_ci <li> 537159b3361Sopenharmony_ci Fixed history so that it <em>finally</em> validates 538159b3361Sopenharmony_ci at <a href="http://validator.w3.org/">W3's validator</a> 539159b3361Sopenharmony_ci </li> 540159b3361Sopenharmony_ci <li> 541159b3361Sopenharmony_ci Fixed compilation of frontend <code>mp3rtp.c</code>. Thanks to Kris Karas. 542159b3361Sopenharmony_ci Bugtracker item <i>[ 2015432 ] mp3rtp missing uint16_t in lame 3.98</i> 543159b3361Sopenharmony_ci </li> 544159b3361Sopenharmony_ci </ul> 545159b3361Sopenharmony_ci </li> 546159b3361Sopenharmony_ci <li> 547159b3361Sopenharmony_ci Robert Hegemann: 548159b3361Sopenharmony_ci <ul> 549159b3361Sopenharmony_ci <li>Fix for Bugtracker item <i>[ 2031704 ] --id3v1-only didnt work in 3.98-final</i></li> 550159b3361Sopenharmony_ci <li>Fix for Bugtracker item <i>[ 2022035 ] encoder_padding value and resampling</i></li> 551159b3361Sopenharmony_ci <li>Fix for Bugtracker item <i>[ 2029282 ] Frequency filtering API broken in 3.98</i></li> 552159b3361Sopenharmony_ci <li>Fix for Bugtracker item <i>[ 2039648 ] potential memory leak in parse_args() function in parse.c</i></li> 553159b3361Sopenharmony_ci <li> 554159b3361Sopenharmony_ci Fix for some tagging issues: 555159b3361Sopenharmony_ci <ul> 556159b3361Sopenharmony_ci <li> 557159b3361Sopenharmony_ci Made search for ID3v1 genres more sloppy, abbrevations may match more often as some simple typos. 558159b3361Sopenharmony_ci Examples: 559159b3361Sopenharmony_ci <ul> 560159b3361Sopenharmony_ci <li>--tg "Alt. Rock" matches genre "Alternate Rock"</li> 561159b3361Sopenharmony_ci <li>--tg "acapela" matches genre "A Cappella"</li> 562159b3361Sopenharmony_ci </ul> 563159b3361Sopenharmony_ci </li> 564159b3361Sopenharmony_ci <li>New switch --pad-id3v2-size "n": adds ID3v2 tag with n padding bytes.</li> 565159b3361Sopenharmony_ci </ul> 566159b3361Sopenharmony_ci </li> 567159b3361Sopenharmony_ci </ul> 568159b3361Sopenharmony_ci </li> 569159b3361Sopenharmony_ci </ul> 570159b3361Sopenharmony_ci <hr> 571159b3361Sopenharmony_ci <h3>LAME 3.98 July 4 2008</h3> 572159b3361Sopenharmony_ci <ul> 573159b3361Sopenharmony_ci <li> 574159b3361Sopenharmony_ci Anton Sergunov: 575159b3361Sopenharmony_ci <ul> 576159b3361Sopenharmony_ci <li>Frontend DirectShow: enabling LAME dshow filter to connect to "File Writer Filter". 577159b3361Sopenharmony_ci </li> 578159b3361Sopenharmony_ci </ul> 579159b3361Sopenharmony_ci </li> 580159b3361Sopenharmony_ci <li> 581159b3361Sopenharmony_ci Rogério Brito: 582159b3361Sopenharmony_ci <ul> 583159b3361Sopenharmony_ci <li>Updates to the Debian Packaging</li> 584159b3361Sopenharmony_ci <li>Fixes to the abx tool for Unix systems (so that more people 585159b3361Sopenharmony_ci can evaluate LAME's compression against the original files) 586159b3361Sopenharmony_ci </li> 587159b3361Sopenharmony_ci </ul> 588159b3361Sopenharmony_ci </li> 589159b3361Sopenharmony_ci <li> 590159b3361Sopenharmony_ci Alexander Leidinger: 591159b3361Sopenharmony_ci <ul> 592159b3361Sopenharmony_ci <li>explicitely link the math lib to the lame lib</li> 593159b3361Sopenharmony_ci <li>add switch to disable the use of the compaq optimized math lib</li> 594159b3361Sopenharmony_ci </ul> 595159b3361Sopenharmony_ci </li> 596159b3361Sopenharmony_ci </ul> 597159b3361Sopenharmony_ci <h3>LAME 3.98 beta 8 April 13 2008</h3> 598159b3361Sopenharmony_ci <ul> 599159b3361Sopenharmony_ci <li> 600159b3361Sopenharmony_ci Robert Hegemann: 601159b3361Sopenharmony_ci <ul> 602159b3361Sopenharmony_ci <li>LAME now accepts a floating point value in the range [0,...,10[ as VBR quality setting, like <b>-V5.678</b> 603159b3361Sopenharmony_ci </li> 604159b3361Sopenharmony_ci <li>Found and fixed some suspicious code in additive masking calculation for VBR-NEW 605159b3361Sopenharmony_ci </li> 606159b3361Sopenharmony_ci <li>bug-fix:<font color="#ff0000">experimental code was defaulted by accident for VBR-NEW</font> 607159b3361Sopenharmony_ci </li> 608159b3361Sopenharmony_ci <li>fix for some endianess problem on big-endian machines 609159b3361Sopenharmony_ci </li> 610159b3361Sopenharmony_ci </ul> 611159b3361Sopenharmony_ci </li> 612159b3361Sopenharmony_ci </ul> 613159b3361Sopenharmony_ci <h3>LAME 3.98 beta 7 April 6 2008</h3> 614159b3361Sopenharmony_ci <ul> 615159b3361Sopenharmony_ci <li> 616159b3361Sopenharmony_ci Robert Hegemann: 617159b3361Sopenharmony_ci <ul> 618159b3361Sopenharmony_ci <li>libmp3lame API: allow frontends to separately retrieve LAME/Xing and ID3 data, because the old library automatism 619159b3361Sopenharmony_ci makes it impossible to make fully buffered encodes. 620159b3361Sopenharmony_ci </li> 621159b3361Sopenharmony_ci <li>libmp3lame API: added some experimental unicode ID3 tagging code. 622159b3361Sopenharmony_ci </li> 623159b3361Sopenharmony_ci <li>frontends: write itself final ID3 tags and LAME/Xing header frame 624159b3361Sopenharmony_ci </li> 625159b3361Sopenharmony_ci <li>lame_enc.dll: writes itself final LAME/Xing header frame 626159b3361Sopenharmony_ci </li> 627159b3361Sopenharmony_ci <li> 628159b3361Sopenharmony_ci Latest changes to the new VBR psymodel: 629159b3361Sopenharmony_ci <ul> 630159b3361Sopenharmony_ci <li>uses a different spreading function 631159b3361Sopenharmony_ci </li> 632159b3361Sopenharmony_ci <li><font color="#ff0000">bug-fix for out-of-bounds array access (program stack corruption possible)</font> 633159b3361Sopenharmony_ci </li> 634159b3361Sopenharmony_ci </ul> 635159b3361Sopenharmony_ci </li> 636159b3361Sopenharmony_ci </ul> 637159b3361Sopenharmony_ci </li> 638159b3361Sopenharmony_ci </ul> 639159b3361Sopenharmony_ci <h3>LAME 3.98 beta 6 December 16 2007</h3> 640159b3361Sopenharmony_ci <ul> 641159b3361Sopenharmony_ci <li> 642159b3361Sopenharmony_ci Robert Hegemann: 643159b3361Sopenharmony_ci <ul> 644159b3361Sopenharmony_ci <li>Feature request <i>[ 1811483 ] WAVE_FORMAT_EXTENSIBLE support (PCM)</i> 645159b3361Sopenharmony_ci </li> 646159b3361Sopenharmony_ci <li><font color="#ff0000">Fix for some rare scalefactor selection issue the newer vbr code had at low compression levels 647159b3361Sopenharmony_ci </font> 648159b3361Sopenharmony_ci </li> 649159b3361Sopenharmony_ci <li>Fix for Bugtracker item<i>[ 1813496 ] AIFF parsing bug</i> 650159b3361Sopenharmony_ci </li> 651159b3361Sopenharmony_ci <li> 652159b3361Sopenharmony_ci Latest changes to the new VBR code: 653159b3361Sopenharmony_ci <ul> 654159b3361Sopenharmony_ci <li><font color="#3366ff">it now has its own psy model, a derivation from NSPSY.</font> 655159b3361Sopenharmony_ci </li> 656159b3361Sopenharmony_ci <li><font color="#ff0000">some more tuning has been done for this new psy model.</font> 657159b3361Sopenharmony_ci Many thanks to Horst Albrecht and Myles Thaiss. 658159b3361Sopenharmony_ci </li> 659159b3361Sopenharmony_ci <li><font color="#ff0000">the "out-of-bits" strategy is reworked</font> 660159b3361Sopenharmony_ci </li> 661159b3361Sopenharmony_ci </ul> 662159b3361Sopenharmony_ci <li><font color="#ff0000">It was possible, that the "interchannel masking effects feature" 663159b3361Sopenharmony_ci was used by the dual-channel-mode for bi-lingual encodings too. It was meant to work 664159b3361Sopenharmony_ci on stereo L/R channels only.</font> 665159b3361Sopenharmony_ci </li> 666159b3361Sopenharmony_ci </ul> 667159b3361Sopenharmony_ci </li> 668159b3361Sopenharmony_ci </ul> 669159b3361Sopenharmony_ci <h3>LAME 3.98 beta 5 August 12 2007</h3> 670159b3361Sopenharmony_ci <ul> 671159b3361Sopenharmony_ci <li> 672159b3361Sopenharmony_ci Jonathan Stott: 673159b3361Sopenharmony_ci <ul> 674159b3361Sopenharmony_ci <li>Bug tracker items: <i>[ 1590693 ] ID3v2 tag not writing, [ 1636267 ] ID3v2 tags overwritten</i><br> 675159b3361Sopenharmony_ci If the output file is opened 'write-only', then LAME can't update the LAME tag. 676159b3361Sopenharmony_ci In this case LAME silently overwrote the first bytes of the file and an 677159b3361Sopenharmony_ci optional ID3v2 tag disappeared. Now an error message will be printed and 678159b3361Sopenharmony_ci no data is written in this case. 679159b3361Sopenharmony_ci </li> 680159b3361Sopenharmony_ci </ul> 681159b3361Sopenharmony_ci </li> 682159b3361Sopenharmony_ci <li> 683159b3361Sopenharmony_ci Robert Hegemann: 684159b3361Sopenharmony_ci <ul> 685159b3361Sopenharmony_ci <li>Fix for Bugtracker item <i>[ 1719593 ] Track numbers > 255 not allowed even with --id3v2-only</i> 686159b3361Sopenharmony_ci <li>Fix for Bugtracker item <i>[ 1742623 ] fail(lame --mp3input -m m -b 128 --resample 8 *.mp3 **.mp3)</i><br> 687159b3361Sopenharmony_ci The problem here was, the input files are MPEG-1 Layer2 files named as MP3s. Even if you leave out 688159b3361Sopenharmony_ci the --mp3input switch LAME tried to decode the input files as Layer3 files because of the file name 689159b3361Sopenharmony_ci extension and because it found some valid looking Layer3 synchronization header. 690159b3361Sopenharmony_ci The fixed LAME version does not assume the file name extension is always correct and treats the files 691159b3361Sopenharmony_ci depending on the first found MPEG sync word. The files in question are now correctly detected as 692159b3361Sopenharmony_ci Layer2 files and transcoding does succeed. 693159b3361Sopenharmony_ci </li> 694159b3361Sopenharmony_ci <li>Fix for Bugtracker item <i>[ 1445175 ] Input being stdin fails in Windows on WAV files</i><br> 695159b3361Sopenharmony_ci The problem here was, seeking on pipes shows some different behaviour depending on C-Library 696159b3361Sopenharmony_ci implementations. The workaround tries to detect it's working on a pipe and doing some 697159b3361Sopenharmony_ci reading instead of seeking in that case. 698159b3361Sopenharmony_ci </li> 699159b3361Sopenharmony_ci <li>Fixing some memory leak in the 'lame_enc.dll'. 700159b3361Sopenharmony_ci </li> 701159b3361Sopenharmony_ci <li>Fix for Bugtracker items <i>[ 1160757, 1160741 ] --little-endian / --big-endian not working</i><br> 702159b3361Sopenharmony_ci These switches where originally intended to be used together with Libsndfile only. 703159b3361Sopenharmony_ci </li> 704159b3361Sopenharmony_ci <li>Fix for Bugtracker item <i>[ 1746336 ] Incorrect Bitrate with ABR und --resample, LAME 3.98b4</i><br> 705159b3361Sopenharmony_ci Some earlier bug-fix had some typo. As a result, when adding a '--resample 123' switch, 706159b3361Sopenharmony_ci the average bitrate rised upto maximum bitrate. 707159b3361Sopenharmony_ci </li> 708159b3361Sopenharmony_ci </ul> 709159b3361Sopenharmony_ci </li> 710159b3361Sopenharmony_ci </ul> 711159b3361Sopenharmony_ci <h3>LAME 3.98 beta 4 June 23 2007</h3> 712159b3361Sopenharmony_ci <ul> 713159b3361Sopenharmony_ci <li>Dennis Lambe Jr: Added support for total track count (id3v2) in the frontend 714159b3361Sopenharmony_ci </li> 715159b3361Sopenharmony_ci <li> 716159b3361Sopenharmony_ci Nyaochi: 717159b3361Sopenharmony_ci <ul> 718159b3361Sopenharmony_ci <li>Ability to set user-defined ID3v2.3 frame</li> 719159b3361Sopenharmony_ci <li>Ability to include albumArt in ID3v2.3 tag</li> 720159b3361Sopenharmony_ci </ul> 721159b3361Sopenharmony_ci </li> 722159b3361Sopenharmony_ci <li> 723159b3361Sopenharmony_ci Robert Hegemann: 724159b3361Sopenharmony_ci <ul> 725159b3361Sopenharmony_ci <li>Bugfix: the "play length in ms", which is stored in the ID3v2 tag TLEN, was not correctly computed. 726159b3361Sopenharmony_ci Some hardware and software players were confused by this garbage data. 727159b3361Sopenharmony_ci </li> 728159b3361Sopenharmony_ci <li><font color="#ff0000">Out of bits strategy for the newer VBR code overhauled</font></li> 729159b3361Sopenharmony_ci <li>LAME API: the ID3 tag functions do not store the pointers passed anymore, 730159b3361Sopenharmony_ci they do make deep copies of strings passed as parameters. 731159b3361Sopenharmony_ci </li> 732159b3361Sopenharmony_ci <li>Changes in LAME frontend switches regarding ID3 tags: 733159b3361Sopenharmony_ci <br><tt>--tg "MyGenre"</tt> will route unknown ID3v1 genres to "Other" for ID3v1 tags 734159b3361Sopenharmony_ci and will be stored as plain text "MyGenre" for ID3v2 tags. Genres given by known 735159b3361Sopenharmony_ci ID3v1 numbers will be stored as its corresponding text in ID3v2 tags. 736159b3361Sopenharmony_ci <br><tt>--tn "02/02"</tt> will store the track number specified as plain text as-is 737159b3361Sopenharmony_ci for ID3v2 tags. 738159b3361Sopenharmony_ci </li> 739159b3361Sopenharmony_ci </ul> 740159b3361Sopenharmony_ci </li> 741159b3361Sopenharmony_ci </ul> 742159b3361Sopenharmony_ci <br> 743159b3361Sopenharmony_ci <h3>LAME 3.98 beta 3 May 22 2007</h3> 744159b3361Sopenharmony_ci <ul> 745159b3361Sopenharmony_ci <li> 746159b3361Sopenharmony_ci Robert Hegemann: 747159b3361Sopenharmony_ci <ul> 748159b3361Sopenharmony_ci <li>Fixes regarding max number of bits limitation</li> 749159b3361Sopenharmony_ci </ul> 750159b3361Sopenharmony_ci </li> 751159b3361Sopenharmony_ci </ul> 752159b3361Sopenharmony_ci <br> 753159b3361Sopenharmony_ci <h3>LAME 3.98 beta 2 May 20 2007</h3> 754159b3361Sopenharmony_ci <ul> 755159b3361Sopenharmony_ci <li> 756159b3361Sopenharmony_ci Robert Hegemann: 757159b3361Sopenharmony_ci <ul> 758159b3361Sopenharmony_ci <li>Bug tracker item: <i>[ 1693461 ];</i> 759159b3361Sopenharmony_ci Fixed memory leaks in ACM codec 760159b3361Sopenharmony_ci </li> 761159b3361Sopenharmony_ci <li>Fixed encoding of non-standard sampling rates in CBR</li> 762159b3361Sopenharmony_ci <li><font color="#ff0000">Improved VBR strategy when running out of bits</font></li> 763159b3361Sopenharmony_ci </ul> 764159b3361Sopenharmony_ci </li> 765159b3361Sopenharmony_ci </ul> 766159b3361Sopenharmony_ci <br> 767159b3361Sopenharmony_ci <h3>LAME 3.98 beta 1 May 16 2007</h3> 768159b3361Sopenharmony_ci <ul> 769159b3361Sopenharmony_ci <li> 770159b3361Sopenharmony_ci Alexander Leidinger: 771159b3361Sopenharmony_ci <ul> 772159b3361Sopenharmony_ci <li>Add TLEN (ID3v2) support (Submitted by: Linus Walleij).</li> 773159b3361Sopenharmony_ci <li>Add number of total tracks per album (ID3v2) support 774159b3361Sopenharmony_ci (Submitted by: Kyle VanderBeek). 775159b3361Sopenharmony_ci </li> 776159b3361Sopenharmony_ci <li>Some seatbelts for overflowing arrays in the ID3v2 777159b3361Sopenharmony_ci support. 778159b3361Sopenharmony_ci </li> 779159b3361Sopenharmony_ci <li>Update the RPM spec (Submitted by: Kyle VanderBeek).</li> 780159b3361Sopenharmony_ci <li>Fix some mem-leaks in the error case.</li> 781159b3361Sopenharmony_ci <li>Update to newer autotools versions.</li> 782159b3361Sopenharmony_ci <li>Update to use a recent libsndfile (submitted by 783159b3361Sopenharmony_ci libsndfile author). 784159b3361Sopenharmony_ci </li> 785159b3361Sopenharmony_ci <li><font color="#3366ff">Intrinsics support 786159b3361Sopenharmony_ci enabled for gcc</font> 787159b3361Sopenharmony_ci </li> 788159b3361Sopenharmony_ci </ul> 789159b3361Sopenharmony_ci </li> 790159b3361Sopenharmony_ci <li> 791159b3361Sopenharmony_ci Robert Hegemann: 792159b3361Sopenharmony_ci <ul> 793159b3361Sopenharmony_ci <li><font color="#3366ff">The newer VBR code is now LAME's default 794159b3361Sopenharmony_ci VBR routine</font> 795159b3361Sopenharmony_ci </li> 796159b3361Sopenharmony_ci <li><font color="#ff0000">Fixed: in case of 797159b3361Sopenharmony_ci not enough bits the new vbr code incorrectly used old vbr routine</font> 798159b3361Sopenharmony_ci </li> 799159b3361Sopenharmony_ci <li><font color="#ff0000">Improved ATH 800159b3361Sopenharmony_ci adjustment in low volume cases</font> 801159b3361Sopenharmony_ci </li> 802159b3361Sopenharmony_ci <li><font color="#ff0000">Fixed (PSY model): mapping from convolution bands 803159b3361Sopenharmony_ci to partition bands was broken since we replaced tables by own calculation 804159b3361Sopenharmony_ci several years ago</font> 805159b3361Sopenharmony_ci </li> 806159b3361Sopenharmony_ci <li><font color="#ff0000">Fixed (PSY model): loss of fraction in equal loudness weighting</font></li> 807159b3361Sopenharmony_ci <li><font color="#ff0000">Fixed (PSY model): in NSPSY highpass filter, out of bounds access in fircoef</font></li> 808159b3361Sopenharmony_ci <li><font color="#ff0000">Known problem samples for the new VBR code: 809159b3361Sopenharmony_ci many of them are at an acceptable quality level now;</font> 810159b3361Sopenharmony_ci with a big <b>'Thank You'</b> to Francis Niechcial 811159b3361Sopenharmony_ci </li> 812159b3361Sopenharmony_ci <li><font color="#ff0000">Modified VBR strategy to handle out of bits cases</font></li> 813159b3361Sopenharmony_ci <li>Restricted bitreservoir size for 320 kbps frames to 814159b3361Sopenharmony_ci the size used for sideinfo, because of decoding problems 815159b3361Sopenharmony_ci with FhG decoders installed on almost every Windows system 816159b3361Sopenharmony_ci </li> 817159b3361Sopenharmony_ci <li>LAME aborts on unsupported input files 818159b3361Sopenharmony_ci or unrecognized parameter options passed more often now 819159b3361Sopenharmony_ci </li> 820159b3361Sopenharmony_ci <li>Bug tracker item: <i>[ 1596306 ] "fatal error during initialization";</i> 821159b3361Sopenharmony_ci an invalid MPEG samplerate was returned by optimum_samplefreq function 822159b3361Sopenharmony_ci </li> 823159b3361Sopenharmony_ci <li>Bug tracker item: <i>[ 1585942 ] lame not --silent when TERM not set;</i> 824159b3361Sopenharmony_ci in case LAME was build with TERMCAP defined and no TERM 825159b3361Sopenharmony_ci environment is defined, now we do not issue an error message and 826159b3361Sopenharmony_ci silently fallback to the default behaviour as if LAME was 827159b3361Sopenharmony_ci compiled without TERMCAP defined. 828159b3361Sopenharmony_ci </li> 829159b3361Sopenharmony_ci <li>Bug tracker item: <i>[ 1711980 ] LAME writes invalid Xing header when ID3 tags exist;</i> 830159b3361Sopenharmony_ci LAME was sometimes writing an invalid Xing/Info header 831159b3361Sopenharmony_ci </li> 832159b3361Sopenharmony_ci <li>Feature request: <i>[ 1588283 ] Flushing output stream in lame.exe;</i> 833159b3361Sopenharmony_ci 'flush' option added 834159b3361Sopenharmony_ci </li> 835159b3361Sopenharmony_ci <li><font color="#3366ff">Added FFTSSE and FFT3DNOW assembler code from 836159b3361Sopenharmony_ci Lame4 branch</font> 837159b3361Sopenharmony_ci </li> 838159b3361Sopenharmony_ci <li>Changes in lame frontend switches: 839159b3361Sopenharmony_ci -k removed, add lowpass and highpass switches if you need to change 840159b3361Sopenharmony_ci them; --short/noshort/allshort - degraded into DEVELOPER ONLY switches 841159b3361Sopenharmony_ci normal users shouldn't use them; -X -Z degraded to 842159b3361Sopenharmony_ci DEVELOPER ONLY switches, -X is too tough to communicate to 843159b3361Sopenharmony_ci end users and -Z isn't used actualy 844159b3361Sopenharmony_ci </li> 845159b3361Sopenharmony_ci <li>Fixed some console printing problems</li> 846159b3361Sopenharmony_ci <li>Windows: ACM code now uses LAME library API only, all 847159b3361Sopenharmony_ci references to private include files are removed 848159b3361Sopenharmony_ci </li> 849159b3361Sopenharmony_ci <li>Windows: DirectShow code now uses LAME library API only, all 850159b3361Sopenharmony_ci references to private include files are removed 851159b3361Sopenharmony_ci </li> 852159b3361Sopenharmony_ci <li>Windows: disabled code that resets processor affinity, 853159b3361Sopenharmony_ci because this doesn't belong to LAME, but seems to work around 854159b3361Sopenharmony_ci some problems the parent process has (in most cases EAC) 855159b3361Sopenharmony_ci </li> 856159b3361Sopenharmony_ci </ul> 857159b3361Sopenharmony_ci </li> 858159b3361Sopenharmony_ci <li> 859159b3361Sopenharmony_ci John33: 860159b3361Sopenharmony_ci <ul> 861159b3361Sopenharmony_ci <li> 862159b3361Sopenharmony_ci Fixed mp2 and mp3 decoding: For mp3 and mp2 decoding, 863159b3361Sopenharmony_ci this now yields the same output as foobar2000 but the 864159b3361Sopenharmony_ci error checking remains unchanged 865159b3361Sopenharmony_ci </li> 866159b3361Sopenharmony_ci </ul> 867159b3361Sopenharmony_ci </li> 868159b3361Sopenharmony_ci <li> 869159b3361Sopenharmony_ci Gabriel Bouvigne: 870159b3361Sopenharmony_ci <ul> 871159b3361Sopenharmony_ci <li>VC8 project files</li> 872159b3361Sopenharmony_ci <li>Added support for x64 under VC8</li> 873159b3361Sopenharmony_ci <li>Restricted MPEG 2.5 to 64kbps frames</li> 874159b3361Sopenharmony_ci </ul> 875159b3361Sopenharmony_ci </li> 876159b3361Sopenharmony_ci <li> 877159b3361Sopenharmony_ci Takehiro Tominaga: 878159b3361Sopenharmony_ci <ul> 879159b3361Sopenharmony_ci <li><font color="#3366ff">SSE version of FFT</font></li> 880159b3361Sopenharmony_ci </ul> 881159b3361Sopenharmony_ci </li> 882159b3361Sopenharmony_ci </ul> 883159b3361Sopenharmony_ci <br> 884159b3361Sopenharmony_ci <h3>LAME 3.97 September 24 2006</h3> 885159b3361Sopenharmony_ci <ul> 886159b3361Sopenharmony_ci <li>3.97 beta 3 becomes 3.97 </li> 887159b3361Sopenharmony_ci </ul> 888159b3361Sopenharmony_ci <br> 889159b3361Sopenharmony_ci <h3>LAME 3.97 beta 3 August 19 2006</h3> 890159b3361Sopenharmony_ci <ul> 891159b3361Sopenharmony_ci <li> 892159b3361Sopenharmony_ci Gabriel Bouvigne: 893159b3361Sopenharmony_ci <ul> 894159b3361Sopenharmony_ci <li><font color="#ff0000">Workaround against a 895159b3361Sopenharmony_ci short blocks detection issue</font> 896159b3361Sopenharmony_ci </li> 897159b3361Sopenharmony_ci </ul> 898159b3361Sopenharmony_ci </li> 899159b3361Sopenharmony_ci </ul> 900159b3361Sopenharmony_ci <br> 901159b3361Sopenharmony_ci <h3>LAME 3.97 beta 2 November 26 2005</h3> 902159b3361Sopenharmony_ci <ul> 903159b3361Sopenharmony_ci <li> 904159b3361Sopenharmony_ci Gabriel Bouvigne: 905159b3361Sopenharmony_ci <ul> 906159b3361Sopenharmony_ci <li>Fixed an initialization error when input is not using a 907159b3361Sopenharmony_ci standard sampling frequency 908159b3361Sopenharmony_ci </li> 909159b3361Sopenharmony_ci <li>Fixed a possible assertion failure in very low bitrate 910159b3361Sopenharmony_ci encoding 911159b3361Sopenharmony_ci </li> 912159b3361Sopenharmony_ci <li><font color="#ff0000">Slight change 913159b3361Sopenharmony_ci regarding ATH adjustment with V5</font> 914159b3361Sopenharmony_ci </li> 915159b3361Sopenharmony_ci <li><font color="#ff0000">Reinstated bit 916159b3361Sopenharmony_ci reservoir for 320kbps CBR</font> 917159b3361Sopenharmony_ci </li> 918159b3361Sopenharmony_ci <li><font color="#3366ff">ReplayGain analysis 919159b3361Sopenharmony_ci should now be faster when encountering silent parts</font> 920159b3361Sopenharmony_ci </li> 921159b3361Sopenharmony_ci </ul> 922159b3361Sopenharmony_ci </li> 923159b3361Sopenharmony_ci <li> 924159b3361Sopenharmony_ci Takehiro Tominaga: 925159b3361Sopenharmony_ci <ul> 926159b3361Sopenharmony_ci <li>Fixed a possible link problem of assembly code</li> 927159b3361Sopenharmony_ci </ul> 928159b3361Sopenharmony_ci </li> 929159b3361Sopenharmony_ci </ul> 930159b3361Sopenharmony_ci <br> 931159b3361Sopenharmony_ci <h3>LAME 3.97 beta 1 September 12 2005</h3> 932159b3361Sopenharmony_ci <ul> 933159b3361Sopenharmony_ci <li> 934159b3361Sopenharmony_ci Takehiro Tominaga: 935159b3361Sopenharmony_ci <ul> 936159b3361Sopenharmony_ci <li>Fixed an out of array access in mp3rtp </li> 937159b3361Sopenharmony_ci <li><font color="#ff0000">Fixed a quality 938159b3361Sopenharmony_ci setting in DLL</font> 939159b3361Sopenharmony_ci </li> 940159b3361Sopenharmony_ci <li>Fixed display when using --silent</li> 941159b3361Sopenharmony_ci </ul> 942159b3361Sopenharmony_ci </li> 943159b3361Sopenharmony_ci <li> 944159b3361Sopenharmony_ci Vitaly Ivanov: 945159b3361Sopenharmony_ci <ul> 946159b3361Sopenharmony_ci <li>Updated DirectShow interface</li> 947159b3361Sopenharmony_ci </ul> 948159b3361Sopenharmony_ci </li> 949159b3361Sopenharmony_ci <li> 950159b3361Sopenharmony_ci Robert Hegemann: 951159b3361Sopenharmony_ci <ul> 952159b3361Sopenharmony_ci <li><font color="#ff0000">Fixed an out of 953159b3361Sopenharmony_ci array access</font> 954159b3361Sopenharmony_ci </li> 955159b3361Sopenharmony_ci <li><font color="#ff0000">Fixed some small 956159b3361Sopenharmony_ci rounding problem in vbr-new quantization routines</font> 957159b3361Sopenharmony_ci </li> 958159b3361Sopenharmony_ci <li><font color="#ff0000">Fixed a bug in 959159b3361Sopenharmony_ci vbr-new regarding high frequencies (sfb21) when using -Y</font> 960159b3361Sopenharmony_ci </li> 961159b3361Sopenharmony_ci <li><font color="#ff0000">Fixed a few bugs in 962159b3361Sopenharmony_ci vbr-new when using -Y</font> 963159b3361Sopenharmony_ci </li> 964159b3361Sopenharmony_ci <li><font color="#ff0000">Updated scalefactors 965159b3361Sopenharmony_ci allocation scheme in vbr-new</font> 966159b3361Sopenharmony_ci </li> 967159b3361Sopenharmony_ci <li>Fixed mingw32 configure problems</li> 968159b3361Sopenharmony_ci <li>Resolved some compiler warnings</li> 969159b3361Sopenharmony_ci <li>Updated command-line visualisation</li> 970159b3361Sopenharmony_ci </ul> 971159b3361Sopenharmony_ci </li> 972159b3361Sopenharmony_ci <li> 973159b3361Sopenharmony_ci Gabriel Bouvigne: 974159b3361Sopenharmony_ci <ul> 975159b3361Sopenharmony_ci <li>Changed some FLOAT8 to FLOAT</li> 976159b3361Sopenharmony_ci <li>Added project files for VC7</li> 977159b3361Sopenharmony_ci <li><font color="#ff0000">Reworked -q1 and -q0</font></li> 978159b3361Sopenharmony_ci <li><font color="#ff0000">Updated presets</font></li> 979159b3361Sopenharmony_ci <li><font color="#ff0000">Fixed an error in 980159b3361Sopenharmony_ci ISO quantization on systems not using the IEEE754 hack</font> 981159b3361Sopenharmony_ci </li> 982159b3361Sopenharmony_ci <li><font color="#3366ff">Faster quantization</font></li> 983159b3361Sopenharmony_ci <li><font color="#3366ff">SSE version of 984159b3361Sopenharmony_ci init_xrpow</font> 985159b3361Sopenharmony_ci </li> 986159b3361Sopenharmony_ci </ul> 987159b3361Sopenharmony_ci </li> 988159b3361Sopenharmony_ci <li> 989159b3361Sopenharmony_ci Rogério Brito: 990159b3361Sopenharmony_ci <ul> 991159b3361Sopenharmony_ci <li>Updated Debian packaging</li> 992159b3361Sopenharmony_ci <li>Documentation work</li> 993159b3361Sopenharmony_ci </ul> 994159b3361Sopenharmony_ci </li> 995159b3361Sopenharmony_ci <li> 996159b3361Sopenharmony_ci Chris Miller: 997159b3361Sopenharmony_ci <ul> 998159b3361Sopenharmony_ci <li>Support for x64 platform SDK in makefile.msvc</li> 999159b3361Sopenharmony_ci </ul> 1000159b3361Sopenharmony_ci </li> 1001159b3361Sopenharmony_ci </ul> 1002159b3361Sopenharmony_ci <br> 1003159b3361Sopenharmony_ci <h3>LAME 3.96.1 July 25 2004</h3> 1004159b3361Sopenharmony_ci <ul> 1005159b3361Sopenharmony_ci <li> 1006159b3361Sopenharmony_ci Robert Hegemann: 1007159b3361Sopenharmony_ci <ul> 1008159b3361Sopenharmony_ci <li><font color="#ff0000">Fixed a rare bug in 1009159b3361Sopenharmony_ci vbr-new (could lead to crashes or data corruption)</font> 1010159b3361Sopenharmony_ci </li> 1011159b3361Sopenharmony_ci </ul> 1012159b3361Sopenharmony_ci </li> 1013159b3361Sopenharmony_ci <li> 1014159b3361Sopenharmony_ci Gabriel Bouvigne: 1015159b3361Sopenharmony_ci <ul> 1016159b3361Sopenharmony_ci <li>some fixes in ACM codec</li> 1017159b3361Sopenharmony_ci <li>fixed padding when encoding to 320kbps</li> 1018159b3361Sopenharmony_ci <li><font color="#ff0000">fixed block size 1019159b3361Sopenharmony_ci selection for mid and side channels</font> 1020159b3361Sopenharmony_ci </li> 1021159b3361Sopenharmony_ci </ul> 1022159b3361Sopenharmony_ci </li> 1023159b3361Sopenharmony_ci </ul> 1024159b3361Sopenharmony_ci <br> 1025159b3361Sopenharmony_ci <h3>LAME 3.96 April 11 2004</h3> 1026159b3361Sopenharmony_ci <ul> 1027159b3361Sopenharmony_ci <li> 1028159b3361Sopenharmony_ci Gabriel Bouvigne: 1029159b3361Sopenharmony_ci <ul> 1030159b3361Sopenharmony_ci <li><font color="#ff0000">new quantization 1031159b3361Sopenharmony_ci selection mode (used in ABR/CBR)</font> 1032159b3361Sopenharmony_ci </li> 1033159b3361Sopenharmony_ci <li><font color="#ff0000">set sfscale for 1034159b3361Sopenharmony_ci ABR/CBR up to 160kbps</font> 1035159b3361Sopenharmony_ci </li> 1036159b3361Sopenharmony_ci </ul> 1037159b3361Sopenharmony_ci </li> 1038159b3361Sopenharmony_ci </ul> 1039159b3361Sopenharmony_ci <br> 1040159b3361Sopenharmony_ci <h3>LAME 3.96 beta 2 March 28 2004</h3> 1041159b3361Sopenharmony_ci <ul> 1042159b3361Sopenharmony_ci <li> 1043159b3361Sopenharmony_ci Takehiro Tominaga: 1044159b3361Sopenharmony_ci <ul> 1045159b3361Sopenharmony_ci <li><font color="#3366ff">removed unnecessary 1046159b3361Sopenharmony_ci integer convertion in resampling</font> 1047159b3361Sopenharmony_ci </li> 1048159b3361Sopenharmony_ci </ul> 1049159b3361Sopenharmony_ci </li> 1050159b3361Sopenharmony_ci <li> 1051159b3361Sopenharmony_ci Robert Hegemann: 1052159b3361Sopenharmony_ci <ul> 1053159b3361Sopenharmony_ci <li><font color="#ff0000">reworked scalefactor 1054159b3361Sopenharmony_ci allocation in vbr-new</font> 1055159b3361Sopenharmony_ci </li> 1056159b3361Sopenharmony_ci <li>fixed a freeformat decoding problem</li> 1057159b3361Sopenharmony_ci </ul> 1058159b3361Sopenharmony_ci </li> 1059159b3361Sopenharmony_ci <li> 1060159b3361Sopenharmony_ci Gabriel Bouvigne: 1061159b3361Sopenharmony_ci <ul> 1062159b3361Sopenharmony_ci <li><font color="#ff0000">updated minimal 1063159b3361Sopenharmony_ci bitrate for V1 and V2</font> 1064159b3361Sopenharmony_ci </li> 1065159b3361Sopenharmony_ci </ul> 1066159b3361Sopenharmony_ci </li> 1067159b3361Sopenharmony_ci <li> 1068159b3361Sopenharmony_ci Aleksander Korzynski: 1069159b3361Sopenharmony_ci <ul> 1070159b3361Sopenharmony_ci <li>added ability to disable ReplayGain analysis</li> 1071159b3361Sopenharmony_ci </ul> 1072159b3361Sopenharmony_ci </li> 1073159b3361Sopenharmony_ci </ul> 1074159b3361Sopenharmony_ci <br> 1075159b3361Sopenharmony_ci <h3>LAME 3.96 beta March 7 2004</h3> 1076159b3361Sopenharmony_ci <ul> 1077159b3361Sopenharmony_ci <li> 1078159b3361Sopenharmony_ci Takehiro Tominaga: 1079159b3361Sopenharmony_ci <ul> 1080159b3361Sopenharmony_ci <li>fixed decoding issue</li> 1081159b3361Sopenharmony_ci </ul> 1082159b3361Sopenharmony_ci </li> 1083159b3361Sopenharmony_ci <li> 1084159b3361Sopenharmony_ci Aleksander Korzynski: 1085159b3361Sopenharmony_ci <ul> 1086159b3361Sopenharmony_ci <li>changed internal ReplayGain handling</li> 1087159b3361Sopenharmony_ci <li>fixed some issues when ReplayGain is used with 1088159b3361Sopenharmony_ci resampling 1089159b3361Sopenharmony_ci </li> 1090159b3361Sopenharmony_ci </ul> 1091159b3361Sopenharmony_ci </li> 1092159b3361Sopenharmony_ci <li> 1093159b3361Sopenharmony_ci Robert Hegemann: 1094159b3361Sopenharmony_ci <ul> 1095159b3361Sopenharmony_ci <li>added standard ISO quantization for vbr-new, used at 1096159b3361Sopenharmony_ci lower quality settings 1097159b3361Sopenharmony_ci </li> 1098159b3361Sopenharmony_ci <li><font color="#3366ff">faster count_bits 1099159b3361Sopenharmony_ci for vbr-new</font> 1100159b3361Sopenharmony_ci </li> 1101159b3361Sopenharmony_ci <li><font color="#3366ff">faster 1102159b3361Sopenharmony_ci find_scalefac_ave function for vbr-new</font> 1103159b3361Sopenharmony_ci </li> 1104159b3361Sopenharmony_ci <li><font color="#ff0000">fixed an out of 1105159b3361Sopenharmony_ci array access in psychoacoustic models; this bug could make some psy 1106159b3361Sopenharmony_ci calculations worthless and sometimes let lame crash</font> 1107159b3361Sopenharmony_ci </li> 1108159b3361Sopenharmony_ci <li><font color="#ff0000">fixed an error on 1109159b3361Sopenharmony_ci silent scalefactor bands; this bug resulted in huffman data overrun 1110159b3361Sopenharmony_ci problems while decoding, resulting in audible glitches</font> 1111159b3361Sopenharmony_ci </li> 1112159b3361Sopenharmony_ci <li>fixed a freeformat decoding bug</li> 1113159b3361Sopenharmony_ci </ul> 1114159b3361Sopenharmony_ci </li> 1115159b3361Sopenharmony_ci <li> 1116159b3361Sopenharmony_ci Gabriel Bouvigne: 1117159b3361Sopenharmony_ci <ul> 1118159b3361Sopenharmony_ci <li><font color="#ff0000">adjusted short block 1119159b3361Sopenharmony_ci thresholds</font> 1120159b3361Sopenharmony_ci </li> 1121159b3361Sopenharmony_ci <li>fixed some array addressing bugs</li> 1122159b3361Sopenharmony_ci <li>made ReplayGain analysis reentrant</li> 1123159b3361Sopenharmony_ci </ul> 1124159b3361Sopenharmony_ci </li> 1125159b3361Sopenharmony_ci <li>David Chandler: fixed a crash in quantize_xrpow </li> 1126159b3361Sopenharmony_ci <li>Michal Bacik: fixed a crash when using 8kHz </li> 1127159b3361Sopenharmony_ci <li>Goran Markovic: fixed some decoding bugs </li> 1128159b3361Sopenharmony_ci <li>John Edwards: fixed a too small buffer in ReplayGain code</li> 1129159b3361Sopenharmony_ci </ul> 1130159b3361Sopenharmony_ci <br> 1131159b3361Sopenharmony_ci <h3>LAME 3.95.1 January 12 2004</h3> 1132159b3361Sopenharmony_ci <ul> 1133159b3361Sopenharmony_ci <li> 1134159b3361Sopenharmony_ci Gabriel Bouvigne: 1135159b3361Sopenharmony_ci <ul> 1136159b3361Sopenharmony_ci <li>fixed a crash when using vbr-new</li> 1137159b3361Sopenharmony_ci <li>changed ReplayGain reference level to 89dB</li> 1138159b3361Sopenharmony_ci </ul> 1139159b3361Sopenharmony_ci </li> 1140159b3361Sopenharmony_ci </ul> 1141159b3361Sopenharmony_ci <br> 1142159b3361Sopenharmony_ci <h3>LAME 3.95 January 11 2004</h3> 1143159b3361Sopenharmony_ci <ul> 1144159b3361Sopenharmony_ci <li> 1145159b3361Sopenharmony_ci Gabriel Bouvigne: 1146159b3361Sopenharmony_ci <ul> 1147159b3361Sopenharmony_ci <li><font color="#ff0000">fixed lowpass values 1148159b3361Sopenharmony_ci when using vbr with mono files</font> 1149159b3361Sopenharmony_ci </li> 1150159b3361Sopenharmony_ci <li><font color="#3366ff">faster quantization 1151159b3361Sopenharmony_ci loops</font> 1152159b3361Sopenharmony_ci </li> 1153159b3361Sopenharmony_ci <li><font color="#3366ff">faster count_bits</font></li> 1154159b3361Sopenharmony_ci <li>fixed a buffer requirement error in ACM codec</li> 1155159b3361Sopenharmony_ci </ul> 1156159b3361Sopenharmony_ci </li> 1157159b3361Sopenharmony_ci <li> 1158159b3361Sopenharmony_ci Takehiro TOMINAGA: 1159159b3361Sopenharmony_ci <ul> 1160159b3361Sopenharmony_ci <li>fixed mpglib and other decoding support code to prevent 1161159b3361Sopenharmony_ci the crash when invalid mp3 input 1162159b3361Sopenharmony_ci </li> 1163159b3361Sopenharmony_ci </ul> 1164159b3361Sopenharmony_ci </li> 1165159b3361Sopenharmony_ci <li>removed Layer I decoding support</li> 1166159b3361Sopenharmony_ci <li><font color="#3366ff">use FastLog and IEEE 754 1167159b3361Sopenharmony_ci hack on PowerPC too (approx. 10 percent faster)</font> 1168159b3361Sopenharmony_ci </li> 1169159b3361Sopenharmony_ci </ul> 1170159b3361Sopenharmony_ci <br> 1171159b3361Sopenharmony_ci <h3>LAME 3.94 beta December 15 2003</h3> 1172159b3361Sopenharmony_ci <ul> 1173159b3361Sopenharmony_ci <li> 1174159b3361Sopenharmony_ci Takehiro Tominaga: 1175159b3361Sopenharmony_ci <ul> 1176159b3361Sopenharmony_ci <li><font color="#ff0000">fixed block 1177159b3361Sopenharmony_ci switching of nspsytune</font> 1178159b3361Sopenharmony_ci </li> 1179159b3361Sopenharmony_ci <li><font color="#ff0000">best huffman divide 1180159b3361Sopenharmony_ci in the inner loop.</font> This should improve the quality, but 1181159b3361Sopenharmony_ci PAINFULLY slow. So it is not enabled by default. Use -q0 to use it. 1182159b3361Sopenharmony_ci </li> 1183159b3361Sopenharmony_ci <li>Changed -q option mapping. "-q2" until version 3.93 is 1184159b3361Sopenharmony_ci now "-q3". 1185159b3361Sopenharmony_ci </li> 1186159b3361Sopenharmony_ci <li><font color="#ff0000">saving bits by 1187159b3361Sopenharmony_ci better scalefactor storing</font> 1188159b3361Sopenharmony_ci </li> 1189159b3361Sopenharmony_ci <li>removed Vorbis support</li> 1190159b3361Sopenharmony_ci <li><font color="#ff0000">substep quantization.</font>This 1191159b3361Sopenharmony_ci should help breaking the SFB21 bloating problem 1192159b3361Sopenharmony_ci </li> 1193159b3361Sopenharmony_ci <li><font color="#ff0000">made psychoacoustic 1194159b3361Sopenharmony_ci model aware of ATH adjustements</font> 1195159b3361Sopenharmony_ci </li> 1196159b3361Sopenharmony_ci <li><font color="#ff0000">use ATH value as 1197159b3361Sopenharmony_ci short block masking lower limit</font> 1198159b3361Sopenharmony_ci </li> 1199159b3361Sopenharmony_ci <li><font color="#ff0000">several fixes in 1200159b3361Sopenharmony_ci psychoacoustic model</font> 1201159b3361Sopenharmony_ci </li> 1202159b3361Sopenharmony_ci <li>more robust decoding</li> 1203159b3361Sopenharmony_ci </ul> 1204159b3361Sopenharmony_ci </li> 1205159b3361Sopenharmony_ci <li>Mark Taylor / Gabriel Bouvigne: fixed issues in VBR header</li> 1206159b3361Sopenharmony_ci <li>Mark Taylor: workaround against some hardware decoder 1207159b3361Sopenharmony_ci defficiencies 1208159b3361Sopenharmony_ci </li> 1209159b3361Sopenharmony_ci <li>Aleksander Korzynski: ability to compute the "Radio" 1210159b3361Sopenharmony_ci ReplayGain and detect clipping on the fly. The ReplayGain value is 1211159b3361Sopenharmony_ci stored in the Lame tag. 1212159b3361Sopenharmony_ci </li> 1213159b3361Sopenharmony_ci <li> 1214159b3361Sopenharmony_ci Gabriel Bouvigne: 1215159b3361Sopenharmony_ci <ul> 1216159b3361Sopenharmony_ci <li><font color="#ff0000">work on presets</font></li> 1217159b3361Sopenharmony_ci <li><font color="#ff0000">use presets by 1218159b3361Sopenharmony_ci default for cbr/abr</font> 1219159b3361Sopenharmony_ci </li> 1220159b3361Sopenharmony_ci <li><font color="#ff0000">use presets by 1221159b3361Sopenharmony_ci default for vbr</font> 1222159b3361Sopenharmony_ci </li> 1223159b3361Sopenharmony_ci <li><font color="#ff0000">analog silence 1224159b3361Sopenharmony_ci detection in partitionned sfb21</font> 1225159b3361Sopenharmony_ci </li> 1226159b3361Sopenharmony_ci <li><font color="#3366ff">do not compute noise 1227159b3361Sopenharmony_ci in upper 0 part of the spectrum</font> 1228159b3361Sopenharmony_ci </li> 1229159b3361Sopenharmony_ci <li><font color="#3366ff">only compute noise 1230159b3361Sopenharmony_ci in modified scalefactor bands</font> 1231159b3361Sopenharmony_ci </li> 1232159b3361Sopenharmony_ci </ul> 1233159b3361Sopenharmony_ci </li> 1234159b3361Sopenharmony_ci <li> 1235159b3361Sopenharmony_ci Guillaume Lessard: 1236159b3361Sopenharmony_ci <ul> 1237159b3361Sopenharmony_ci <li>nogap related changes</li> 1238159b3361Sopenharmony_ci </ul> 1239159b3361Sopenharmony_ci </li> 1240159b3361Sopenharmony_ci <li> 1241159b3361Sopenharmony_ci Alexander Leidinger: 1242159b3361Sopenharmony_ci <ul> 1243159b3361Sopenharmony_ci <li>prevent closing the input fd prematurely if the input 1244159b3361Sopenharmony_ci is a named pipe 1245159b3361Sopenharmony_ci </li> 1246159b3361Sopenharmony_ci </ul> 1247159b3361Sopenharmony_ci </li> 1248159b3361Sopenharmony_ci </ul> 1249159b3361Sopenharmony_ci <br> 1250159b3361Sopenharmony_ci <h3>LAME 3.93.1 December 1 2002</h3> 1251159b3361Sopenharmony_ci <ul> 1252159b3361Sopenharmony_ci <li> 1253159b3361Sopenharmony_ci Gabriel Bouvigne: 1254159b3361Sopenharmony_ci <ul> 1255159b3361Sopenharmony_ci <li>preset medium added to the dll interface</li> 1256159b3361Sopenharmony_ci <li><font color="#ff0000">fix for abr/cbr 1257159b3361Sopenharmony_ci presets</font> 1258159b3361Sopenharmony_ci </li> 1259159b3361Sopenharmony_ci <li><font color="#ff0000">fix -q0 switch</font></li> 1260159b3361Sopenharmony_ci </ul> 1261159b3361Sopenharmony_ci </li> 1262159b3361Sopenharmony_ci <li>Alexander Leidinger: fix link problem on systems where 1263159b3361Sopenharmony_ci socket() resides in libsocket 1264159b3361Sopenharmony_ci </li> 1265159b3361Sopenharmony_ci </ul> 1266159b3361Sopenharmony_ci <br> 1267159b3361Sopenharmony_ci <h3>LAME 3.93 November 16 2002</h3> 1268159b3361Sopenharmony_ci <ul> 1269159b3361Sopenharmony_ci <li> 1270159b3361Sopenharmony_ci Takehiro Tominaga: 1271159b3361Sopenharmony_ci <ul> 1272159b3361Sopenharmony_ci <li><font color="#ff0000">bit allocation for 1273159b3361Sopenharmony_ci pre-echo control improved for single channel encodings</font> 1274159b3361Sopenharmony_ci </li> 1275159b3361Sopenharmony_ci <li><font color="#ff0000">substep noise shaping</font></li> 1276159b3361Sopenharmony_ci <li><font color="#3366ff">optimizations by 1277159b3361Sopenharmony_ci changing data structure</font> 1278159b3361Sopenharmony_ci </li> 1279159b3361Sopenharmony_ci <li><font color="#ff0000">noise shaping model 1280159b3361Sopenharmony_ci 2 fix</font> 1281159b3361Sopenharmony_ci </li> 1282159b3361Sopenharmony_ci <li><font color="#3366ff">nspsytune FIR filter 1283159b3361Sopenharmony_ci clean up</font> 1284159b3361Sopenharmony_ci </li> 1285159b3361Sopenharmony_ci <li><font color="#ff0000">fix small psymodel 1286159b3361Sopenharmony_ci bugs(DC current estimation, preecho detection of non-VBR mode, and 1287159b3361Sopenharmony_ci nspsymode initialization)</font> 1288159b3361Sopenharmony_ci </li> 1289159b3361Sopenharmony_ci <li>portability fixes for Tru64 UNIX</li> 1290159b3361Sopenharmony_ci </ul> 1291159b3361Sopenharmony_ci </li> 1292159b3361Sopenharmony_ci <li>Albert Faber: some fixes in the DLL</li> 1293159b3361Sopenharmony_ci <li>Simon Blandford: fixes for channel scaling in mono mode</li> 1294159b3361Sopenharmony_ci <li><font color="#3366ff">Dominique Duvivier: some 1295159b3361Sopenharmony_ci optimizations and a faster log10 function</font> 1296159b3361Sopenharmony_ci </li> 1297159b3361Sopenharmony_ci <li> 1298159b3361Sopenharmony_ci Mark Taylor: 1299159b3361Sopenharmony_ci <ul> 1300159b3361Sopenharmony_ci <li>some tag related fixes in the direct show filter and in 1301159b3361Sopenharmony_ci the ACM codec 1302159b3361Sopenharmony_ci </li> 1303159b3361Sopenharmony_ci <li><font color="#3366ff">fixed a mono 1304159b3361Sopenharmony_ci encoding bug found by Justin Schoeman</font> 1305159b3361Sopenharmony_ci </li> 1306159b3361Sopenharmony_ci <li>calc_noise bug fix</li> 1307159b3361Sopenharmony_ci <li>other fixes</li> 1308159b3361Sopenharmony_ci </ul> 1309159b3361Sopenharmony_ci </li> 1310159b3361Sopenharmony_ci <li> 1311159b3361Sopenharmony_ci Alexander Leidinger: 1312159b3361Sopenharmony_ci <ul> 1313159b3361Sopenharmony_ci <li>update to autoconf 2.53, rewrite some configure tests</li> 1314159b3361Sopenharmony_ci <li>Akos Maroy: determine gcc version even with gcc 3.1</li> 1315159b3361Sopenharmony_ci <li>Andrew Bachmann: compile shared libs on BeOS (and 1316159b3361Sopenharmony_ci perhaps other arches) 1317159b3361Sopenharmony_ci </li> 1318159b3361Sopenharmony_ci <li>ultrasparc switches for gcc 3.1</li> 1319159b3361Sopenharmony_ci <li>fixes for SunOS 4.x</li> 1320159b3361Sopenharmony_ci <li>fixes for 64bit arches</li> 1321159b3361Sopenharmony_ci <li>CFLAGS fix for IRIX</li> 1322159b3361Sopenharmony_ci <li>don't override CFLAGS if exptopt isn't requested</li> 1323159b3361Sopenharmony_ci </ul> 1324159b3361Sopenharmony_ci </li> 1325159b3361Sopenharmony_ci <li> 1326159b3361Sopenharmony_ci Robert Hegeman: 1327159b3361Sopenharmony_ci <ul> 1328159b3361Sopenharmony_ci <li><font color="#3366ff">some fixes</font></li> 1329159b3361Sopenharmony_ci <li><font color="#ff0000">some fixes for VBR</font></li> 1330159b3361Sopenharmony_ci </ul> 1331159b3361Sopenharmony_ci </li> 1332159b3361Sopenharmony_ci <li> 1333159b3361Sopenharmony_ci Gabriel Bouvigne: 1334159b3361Sopenharmony_ci <ul> 1335159b3361Sopenharmony_ci <li>--noasm switch. Might help Cyrix/Via users</li> 1336159b3361Sopenharmony_ci <li><font color="#ff0000">presets and 1337159b3361Sopenharmony_ci alt-presets merged</font> 1338159b3361Sopenharmony_ci </li> 1339159b3361Sopenharmony_ci </ul> 1340159b3361Sopenharmony_ci </li> 1341159b3361Sopenharmony_ci </ul> 1342159b3361Sopenharmony_ci <br> 1343159b3361Sopenharmony_ci <h3>LAME 3.92 April 14 2002</h3> 1344159b3361Sopenharmony_ci <ul> 1345159b3361Sopenharmony_ci <li><font color="#ff0000">Alexander 1346159b3361Sopenharmony_ci Leidinger: add non linear psymodel (compile time option, 1347159b3361Sopenharmony_ci disabled by default)</font>, workaround a bug in gcc 3.0.3 1348159b3361Sopenharmony_ci (compiler options, based upon suggestions from various people, see 1349159b3361Sopenharmony_ci archives and changelog for more) 1350159b3361Sopenharmony_ci </li> 1351159b3361Sopenharmony_ci <li>Steve Lhomme: ACM wrapper (MS-Windows codec)</li> 1352159b3361Sopenharmony_ci <li><font color="#3366ff">Steve Lhomme: 1353159b3361Sopenharmony_ci less memory copying on stereo (interleaved) input</font> 1354159b3361Sopenharmony_ci </li> 1355159b3361Sopenharmony_ci <li><font color="#ff0000">Takehiro Tominaga: 1356159b3361Sopenharmony_ci Inter-channel masking, enables with --interch x option</font> 1357159b3361Sopenharmony_ci </li> 1358159b3361Sopenharmony_ci <li>For buggy versions of gcc compiler (2.96*), back off on 1359159b3361Sopenharmony_ci some of the advanced compiler options<br> 1360159b3361Sopenharmony_ci </li> 1361159b3361Sopenharmony_ci </ul> 1362159b3361Sopenharmony_ci <br> 1363159b3361Sopenharmony_ci <h3>LAME 3.91 December 29 2001</h3> 1364159b3361Sopenharmony_ci <ul> 1365159b3361Sopenharmony_ci <li><font color="#ff0000">Darin 1366159b3361Sopenharmony_ci Morrison: Bugfix for --alt-preset (for content with low 1367159b3361Sopenharmony_ci volume, clean vocals), only important for the "fast standard" preset</font> 1368159b3361Sopenharmony_ci </li> 1369159b3361Sopenharmony_ci <li> 1370159b3361Sopenharmony_ci Alexander Leidinger: 1371159b3361Sopenharmony_ci <ul> 1372159b3361Sopenharmony_ci <li>add some missing files to the distribution</li> 1373159b3361Sopenharmony_ci <li>add --alt-preset to the man page</li> 1374159b3361Sopenharmony_ci </ul> 1375159b3361Sopenharmony_ci </li> 1376159b3361Sopenharmony_ci </ul> 1377159b3361Sopenharmony_ci <br> 1378159b3361Sopenharmony_ci <h3>LAME 3.90 December 21 2001</h3> 1379159b3361Sopenharmony_ci <ul> 1380159b3361Sopenharmony_ci <li><font color="#ff0000">Many small improvements 1381159b3361Sopenharmony_ci and bug fixes not added to history</font> 1382159b3361Sopenharmony_ci </li> 1383159b3361Sopenharmony_ci <li><font color="#ff0000">John 1384159b3361Sopenharmony_ci Dahlstrom: more fine tuning on the auto adjustment of the ATH</font> 1385159b3361Sopenharmony_ci </li> 1386159b3361Sopenharmony_ci <li><font color="#3366ff">Robert 1387159b3361Sopenharmony_ci Hegemann: small speed and quality improvements for the old 1388159b3361Sopenharmony_ci VBR code (--vbr-old).</font> 1389159b3361Sopenharmony_ci </li> 1390159b3361Sopenharmony_ci <li><font color="#ff0000">Robert 1391159b3361Sopenharmony_ci Hegemann: some short block bug fixes</font> 1392159b3361Sopenharmony_ci </li> 1393159b3361Sopenharmony_ci <li><font color="#ff0000">Robert 1394159b3361Sopenharmony_ci Hegemann: Big improvements to --vbr-mtrh, now encodes much 1395159b3361Sopenharmony_ci more frequencies over 16khz</font> 1396159b3361Sopenharmony_ci </li> 1397159b3361Sopenharmony_ci <li><font color="#ff0000">Robert 1398159b3361Sopenharmony_ci Hegemann: --vbr-new code disabled (outdated and lower 1399159b3361Sopenharmony_ci quality) and replaced with --vbr-mtrh (Both --vbr-new and --vbr-mtrh 1400159b3361Sopenharmony_ci now default to mtrh)</font> 1401159b3361Sopenharmony_ci </li> 1402159b3361Sopenharmony_ci <li>Robert Hegemann: reordering of --longhelp to give 1403159b3361Sopenharmony_ci more information, --extrahelp dropped 1404159b3361Sopenharmony_ci </li> 1405159b3361Sopenharmony_ci <li> 1406159b3361Sopenharmony_ci Darin Morrison: Totally revamped and extremely 1407159b3361Sopenharmony_ci high quality unified preset system and other general quality 1408159b3361Sopenharmony_ci improvements now available with --alt-presets: 1409159b3361Sopenharmony_ci <ul> 1410159b3361Sopenharmony_ci <li> 1411159b3361Sopenharmony_ci <font color="#ff0000">some improvements 1412159b3361Sopenharmony_ci to psychoacoustics (vast improvements over default L.A.M.E. modes) when 1413159b3361Sopenharmony_ci --alt-preset is used including:</font> 1414159b3361Sopenharmony_ci <ul> 1415159b3361Sopenharmony_ci <li><font color="#ff0000">Improved tuning 1416159b3361Sopenharmony_ci of short block usage.</font> 1417159b3361Sopenharmony_ci </li> 1418159b3361Sopenharmony_ci <li><font color="#ff0000">Improved 1419159b3361Sopenharmony_ci quantization selection usage (the -X modes), now adapts between 1420159b3361Sopenharmony_ci appropriate modes on the fly. Also helps on "dropout" problems and with 1421159b3361Sopenharmony_ci pre-echo cases.</font> 1422159b3361Sopenharmony_ci </li> 1423159b3361Sopenharmony_ci <li><font color="#ff0000">Improved joint 1424159b3361Sopenharmony_ci stereo usage. Thresholds are better tuned now and fix some "dropout" 1425159b3361Sopenharmony_ci problems L.A.M.E. suffers from on clips like serioustrouble.</font> 1426159b3361Sopenharmony_ci </li> 1427159b3361Sopenharmony_ci <li><font color="#ff0000">Improved noise 1428159b3361Sopenharmony_ci shaping usage. Now switches between noise shaping modes on the fly 1429159b3361Sopenharmony_ci (toggles -Z on and off when appropriate) which allows lower bitrates 1430159b3361Sopenharmony_ci but without the quality compromise.</font> 1431159b3361Sopenharmony_ci </li> 1432159b3361Sopenharmony_ci <li><font color="#ff0000">Clips vastly 1433159b3361Sopenharmony_ci improved over default L.A.M.E. modes (vbr/cbr/abr, including --r3mix): 1434159b3361Sopenharmony_ci castanets, florida_seq, death2, fatboy, spahm, gbtinc, ravebase, short, 1435159b3361Sopenharmony_ci florida_seq, hihat, bassdrum, 2nd_vent_clip, serioustrouble, bloodline, 1436159b3361Sopenharmony_ci and others. No degraded clips known.</font> 1437159b3361Sopenharmony_ci </li> 1438159b3361Sopenharmony_ci <li>VBR bitrates are now more "stable" with less 1439159b3361Sopenharmony_ci fluctuation -- not dipping too low on some music and not increasing too 1440159b3361Sopenharmony_ci high unnecessarily on other music. "--alt-preset standard" provides 1441159b3361Sopenharmony_ci bitrates roughly within the range of 180-220kbps, often averaging close 1442159b3361Sopenharmony_ci to 192kbps. 1443159b3361Sopenharmony_ci </li> 1444159b3361Sopenharmony_ci </ul> 1445159b3361Sopenharmony_ci </li> 1446159b3361Sopenharmony_ci <li>--alt-presets replace the --dm-presets and "metal" 1447159b3361Sopenharmony_ci preset is removed and replaced with generic abr and cbr presets. 1448159b3361Sopenharmony_ci </li> 1449159b3361Sopenharmony_ci <li>--alt-preset extreme (note the 'e') replaces xtreme to 1450159b3361Sopenharmony_ci help eliminate some confusion 1451159b3361Sopenharmony_ci </li> 1452159b3361Sopenharmony_ci <li>--alt-preset vbr modes now have a fast option which 1453159b3361Sopenharmony_ci offers almost no compromise in speed. 1454159b3361Sopenharmony_ci </li> 1455159b3361Sopenharmony_ci <li>--alt-preset standard (and "fast standard") are now 1456159b3361Sopenharmony_ci much lower in bitrate, matching --r3mix with an overall average, though 1457159b3361Sopenharmony_ci offering higher quality especially on difficult test samples. 1458159b3361Sopenharmony_ci </li> 1459159b3361Sopenharmony_ci <li>--alt-presets are no longer just "presets" as in a 1460159b3361Sopenharmony_ci collection of switches, instead they are now quality "modes" because of 1461159b3361Sopenharmony_ci special code level tunings (those mentioned above). 1462159b3361Sopenharmony_ci </li> 1463159b3361Sopenharmony_ci <li>Use --alt-preset help for more information.</li> 1464159b3361Sopenharmony_ci </ul> 1465159b3361Sopenharmony_ci </li> 1466159b3361Sopenharmony_ci <li>Roel VdB: more tuning on the --r3mix preset </li> 1467159b3361Sopenharmony_ci <li>Jon Dee, Roel VdB: INFO tag</li> 1468159b3361Sopenharmony_ci <li>Alexander Leidinger, mp3gain@hotmail.com: added 1469159b3361Sopenharmony_ci --scale-l and --scale-r to scale stereo channels independantly 1470159b3361Sopenharmony_ci </li> 1471159b3361Sopenharmony_ci <li>Takehiro Tominaga: <font color="#ff0000">new 1472159b3361Sopenharmony_ci noise shaping mode, offering more "cutting edge" shaping according to 1473159b3361Sopenharmony_ci masking, enabled via -q0</font> 1474159b3361Sopenharmony_ci </li> 1475159b3361Sopenharmony_ci <li>Mark Taylor: More work on --nogap </li> 1476159b3361Sopenharmony_ci <li>Gabriel Bouvigne: Small changes to abr code for 1477159b3361Sopenharmony_ci more accurate final bitrate 1478159b3361Sopenharmony_ci </li> 1479159b3361Sopenharmony_ci <li>Gabriel Bouvigne, mp3gain@hotmail.com: 1480159b3361Sopenharmony_ci Preliminary <a href="http://www.replaygain.org"> 1481159b3361Sopenharmony_ci ReplayGain</a> analysis code added (not functional yet) 1482159b3361Sopenharmony_ci </li> 1483159b3361Sopenharmony_ci <li>Gabriel Bouvigne, Alexander Leidinger: 1484159b3361Sopenharmony_ci Documentation updates 1485159b3361Sopenharmony_ci </li> 1486159b3361Sopenharmony_ci <li>John Dahlstrom, DSPguru@math.com: floating point 1487159b3361Sopenharmony_ci interface function in the Windows DLL 1488159b3361Sopenharmony_ci </li> 1489159b3361Sopenharmony_ci </ul> 1490159b3361Sopenharmony_ci <br> 1491159b3361Sopenharmony_ci <h3>LAME 3.89beta July 5 2001</h3> 1492159b3361Sopenharmony_ci <ul> 1493159b3361Sopenharmony_ci <li>John Stewart: long filename support for Win9x/NT.</li> 1494159b3361Sopenharmony_ci <li>Takehiro Tominaga: LAME can calculate the CRC of 1495159b3361Sopenharmony_ci VBR header, so now "lame -pv" works fine. 1496159b3361Sopenharmony_ci </li> 1497159b3361Sopenharmony_ci <li><font color="#ff0000">Robert 1498159b3361Sopenharmony_ci Hegemann: Improvements of the new VBR code (--vbr-mtrh).</font> 1499159b3361Sopenharmony_ci </li> 1500159b3361Sopenharmony_ci <li><font color="#3366ff">Robert Hegemann: New VBR 1501159b3361Sopenharmony_ci code (--vbr-mtrh) is now defaulted to get more feedback. The VBR speed 1502159b3361Sopenharmony_ci is now on par with CBR. We will use the old VBR code in the release.</font> 1503159b3361Sopenharmony_ci </li> 1504159b3361Sopenharmony_ci <li><font color="#ff0000">Gabriel Bouvigne: Change 1505159b3361Sopenharmony_ci of the maximum frame size limit. LAME should now be more friendly with 1506159b3361Sopenharmony_ci hardware players.</font> 1507159b3361Sopenharmony_ci </li> 1508159b3361Sopenharmony_ci <li>Gabriel Bouvigne: Size of VBR is now more balanced 1509159b3361Sopenharmony_ci according to the -V value. 1510159b3361Sopenharmony_ci </li> 1511159b3361Sopenharmony_ci <li>Alexander Leidinger: Finished the implementation of the 1512159b3361Sopenharmony_ci set/get functions. 1513159b3361Sopenharmony_ci </li> 1514159b3361Sopenharmony_ci <li>John Dahlstrom: LAME now handles 24bits input</li> 1515159b3361Sopenharmony_ci <li>Mark Taylor: bugs in lame --decode causing truncation of 1516159b3361Sopenharmony_ci mp3 file fixed 1517159b3361Sopenharmony_ci </li> 1518159b3361Sopenharmony_ci <li>Mark Taylor: preliminary --nogap support</li> 1519159b3361Sopenharmony_ci <li>"Final" API completed: shared library safe! This 1520159b3361Sopenharmony_ci API is frozen and should be backwords compatiable with future versions 1521159b3361Sopenharmony_ci of libmp3lame.so, but we will continue to add new functionality. 1522159b3361Sopenharmony_ci <br> 1523159b3361Sopenharmony_ci </li> 1524159b3361Sopenharmony_ci </ul> 1525159b3361Sopenharmony_ci <h3> LAME 3.88beta March 25 2001</h3> 1526159b3361Sopenharmony_ci <ul> 1527159b3361Sopenharmony_ci <li><font color="#ff0000">A lot of work that was 1528159b3361Sopenharmony_ci never added to the History!</font> 1529159b3361Sopenharmony_ci </li> 1530159b3361Sopenharmony_ci <li><font color="#ff0000">Frank Klemm and 1531159b3361Sopenharmony_ci Gabriel Bouvigne: New ATH formula. Big improvement 1532159b3361Sopenharmony_ci for high bitrate encodings.</font> 1533159b3361Sopenharmony_ci </li> 1534159b3361Sopenharmony_ci <li><font color="#ff0000">Takehiro Tominaga: 1535159b3361Sopenharmony_ci Temporal masking</font> 1536159b3361Sopenharmony_ci </li> 1537159b3361Sopenharmony_ci <li><font color="#ff0000">Gabriel Bouvigne/Mark 1538159b3361Sopenharmony_ci Taylor: auto adjustment of ATH</font> 1539159b3361Sopenharmony_ci </li> 1540159b3361Sopenharmony_ci <li><font color="#ff0000">Robert 1541159b3361Sopenharmony_ci Hegemann: Better outer_loop stopping criterion. 1542159b3361Sopenharmony_ci Enabled with -q2 or better.</font> 1543159b3361Sopenharmony_ci </li> 1544159b3361Sopenharmony_ci <li><font color="#ff0000">Robert Hegemann/Naoki 1545159b3361Sopenharmony_ci Shibata: slow/carefull noise shaping. 1546159b3361Sopenharmony_ci -q3..9: amplify all distorted bands. -q2: amplify 1547159b3361Sopenharmony_ci distorted bands within 50%. -q1-0: amplify only 1548159b3361Sopenharmony_ci most distorted band at each iteration.</font> 1549159b3361Sopenharmony_ci </li> 1550159b3361Sopenharmony_ci <li><font color="#ff0000">Takehiro Tominaga: 1551159b3361Sopenharmony_ci Interframe, shortblock temporal masking.</font> 1552159b3361Sopenharmony_ci </li> 1553159b3361Sopenharmony_ci <li>Takehiro Tominaga: LAME restructured into a 1554159b3361Sopenharmony_ci shared library and front end application. Slight changes to 1555159b3361Sopenharmony_ci the API. More changes are coming to turn LAME into a true shared 1556159b3361Sopenharmony_ci library (right now you have to recompile if you upgrade the library :-( 1557159b3361Sopenharmony_ci </li> 1558159b3361Sopenharmony_ci <li> 1559159b3361Sopenharmony_ci <font color="#000000">Naoki Shibata:</font> 1560159b3361Sopenharmony_ci <ul> 1561159b3361Sopenharmony_ci <li><font color="#ff0000">improvements to 1562159b3361Sopenharmony_ci psychoacoustics</font><font color="#000000"> 1563159b3361Sopenharmony_ci (--nspsytune)</font> 1564159b3361Sopenharmony_ci <li><font color="#ff0000">BUG in long block 1565159b3361Sopenharmony_ci pre echo control fixed </font><font color="#000000"> 1566159b3361Sopenharmony_ci (some out of range array access in M/S psychoacoustics)</font> 1567159b3361Sopenharmony_ci </li> 1568159b3361Sopenharmony_ci </ul> 1569159b3361Sopenharmony_ci </li> 1570159b3361Sopenharmony_ci <li><font color="#000000">Ralf 1571159b3361Sopenharmony_ci Kempkens: Visual Basic Script for lame, 1572159b3361Sopenharmony_ci suggested to put it on your Windows Desktop and you can drag'n'drop 1573159b3361Sopenharmony_ci Waves to encode on it.</font> 1574159b3361Sopenharmony_ci </li> 1575159b3361Sopenharmony_ci <li><font color="#000000">Alexander 1576159b3361Sopenharmony_ci Stumpf: improved lame.bat for 4Dos users</font> 1577159b3361Sopenharmony_ci </li> 1578159b3361Sopenharmony_ci <li><font color="#000000">Mark Taylor: Several 1579159b3361Sopenharmony_ci bugs fixed in the resampling code.</font> 1580159b3361Sopenharmony_ci </li> 1581159b3361Sopenharmony_ci <li><font color="#000000">Frank Klemm, Robert 1582159b3361Sopenharmony_ci Hegemann: added assembler code for CPU 1583159b3361Sopenharmony_ci feature detection on runtime (MMX, 3DNow, SIMD)</font> 1584159b3361Sopenharmony_ci </li> 1585159b3361Sopenharmony_ci <li><font color="#3366ff">Takehiro Tominaga: 1586159b3361Sopenharmony_ci 3DNow FFT code.</font> 1587159b3361Sopenharmony_ci </li> 1588159b3361Sopenharmony_ci <li><font color="#000000">Florian Bome, 1589159b3361Sopenharmony_ci Alexander Leidinger: more work on 1590159b3361Sopenharmony_ci configure stuff</font> 1591159b3361Sopenharmony_ci </li> 1592159b3361Sopenharmony_ci <li><font color="#000000">Alexander 1593159b3361Sopenharmony_ci Leidinger: automake/libtool generated Makefiles and 1594159b3361Sopenharmony_ci TONS of other work.</font> 1595159b3361Sopenharmony_ci </li> 1596159b3361Sopenharmony_ci <li><font color="#000000">Alexander 1597159b3361Sopenharmony_ci Leidinger: Much work towards shared library style 1598159b3361Sopenharmony_ci API.</font> 1599159b3361Sopenharmony_ci </li> 1600159b3361Sopenharmony_ci <li><font color="#000000">Anonymous: New more 1601159b3361Sopenharmony_ci efficient RTP code.</font> 1602159b3361Sopenharmony_ci </li> 1603159b3361Sopenharmony_ci <li><font color="#ff0000">Mark Taylor: 1604159b3361Sopenharmony_ci psycho-acoustic data now computed for all scalefactor bands (up to 24 1605159b3361Sopenharmony_ci kHz)</font> 1606159b3361Sopenharmony_ci </li> 1607159b3361Sopenharmony_ci <li><font color="#ff0000">Mark Taylor, Takehiro 1608159b3361Sopenharmony_ci Tominaga: All ISO table data replaced by formulas - should improve 1609159b3361Sopenharmony_ci MPEG2.5 results for which we never had correct table data.</font> 1610159b3361Sopenharmony_ci </li> 1611159b3361Sopenharmony_ci </ul> 1612159b3361Sopenharmony_ci <h3> LAME 3.87alpha September 25 2000</h3> 1613159b3361Sopenharmony_ci <ul> 1614159b3361Sopenharmony_ci <li>Mark Taylor: Bug fixed in LAME/mpglib error 1615159b3361Sopenharmony_ci recovery when encountering a corrupt MP3 frame during 1616159b3361Sopenharmony_ci *decoding*. 1617159b3361Sopenharmony_ci </li> 1618159b3361Sopenharmony_ci <li>Albert Faber: added LayerI+II decoding support</li> 1619159b3361Sopenharmony_ci <li><font color="#000000">Frank Klemm: 1620159b3361Sopenharmony_ci added improved CRC calculation</font> 1621159b3361Sopenharmony_ci </li> 1622159b3361Sopenharmony_ci <li><font color="#000000">Frank Klemm: 1623159b3361Sopenharmony_ci substantial code cleanup/improvements</font> 1624159b3361Sopenharmony_ci </li> 1625159b3361Sopenharmony_ci <li> 1626159b3361Sopenharmony_ci Robert Hegemann: Bug fixes 1627159b3361Sopenharmony_ci <ul> 1628159b3361Sopenharmony_ci <li><font color="#ff0000">in huffman_init</font>, 1629159b3361Sopenharmony_ci could lead to segmentation faults (only in rare cases, most likely at 1630159b3361Sopenharmony_ci lower sample rates) 1631159b3361Sopenharmony_ci </li> 1632159b3361Sopenharmony_ci <li><font color="#ff0000">M/S switching at 1633159b3361Sopenharmony_ci lower sample rates</font> (the fact there is no 2nd granule was 1634159b3361Sopenharmony_ci ignored) 1635159b3361Sopenharmony_ci </li> 1636159b3361Sopenharmony_ci </ul> 1637159b3361Sopenharmony_ci </li> 1638159b3361Sopenharmony_ci <li><font color="#3366ff">Robert 1639159b3361Sopenharmony_ci Hegemann: speed up in VBR</font> 1640159b3361Sopenharmony_ci </li> 1641159b3361Sopenharmony_ci <li>Jarmo Laakkonen: Amiga/GCC settings for 1642159b3361Sopenharmony_ci Makefile.unix. 1643159b3361Sopenharmony_ci </li> 1644159b3361Sopenharmony_ci <li>Magnus Holmgren: README and Makefile for (free) 1645159b3361Sopenharmony_ci Borland C++ compiler. Will also compile lame_enc.dll, but 1646159b3361Sopenharmony_ci this is untested. 1647159b3361Sopenharmony_ci </li> 1648159b3361Sopenharmony_ci <li>Florian Bome: LAME finally has 1649159b3361Sopenharmony_ci a ./configure script!! 1650159b3361Sopenharmony_ci </li> 1651159b3361Sopenharmony_ci </ul> 1652159b3361Sopenharmony_ci <h3> LAME 3.86beta August 6 2000</h3> 1653159b3361Sopenharmony_ci <ul> 1654159b3361Sopenharmony_ci <li>Christopher Wise: A makefile for DJGPP, the DOS 1655159b3361Sopenharmony_ci version of gcc. Now most windows users should be able to 1656159b3361Sopenharmony_ci compile LAME with minimal effort. 1657159b3361Sopenharmony_ci </li> 1658159b3361Sopenharmony_ci <li><font color="#ff0000">Robert 1659159b3361Sopenharmony_ci Hegemann: old VBR: fixed some bugs and 1660159b3361Sopenharmony_ci Takehiro's scalefac_scale feature (not yet on by 1661159b3361Sopenharmony_ci default.) older LAME versions did not allow to spent more 1662159b3361Sopenharmony_ci than 2500 bits of 4095 possible bits to a granule per channel, now 1663159b3361Sopenharmony_ci fixed.</font> 1664159b3361Sopenharmony_ci </li> 1665159b3361Sopenharmony_ci <li>Robert Hegemann: new VBR: 1666159b3361Sopenharmony_ci analog silence treatment like in old VBR 1667159b3361Sopenharmony_ci </li> 1668159b3361Sopenharmony_ci <li>William Welch: Improved options for Linux/Alpha 1669159b3361Sopenharmony_ci gcc and ccc compilers in Makefile. 1670159b3361Sopenharmony_ci </li> 1671159b3361Sopenharmony_ci <li>Mathew Hendry: setting appropriate CRC bit for 1672159b3361Sopenharmony_ci additional Xing-VBR tagging frame 1673159b3361Sopenharmony_ci </li> 1674159b3361Sopenharmony_ci <li>Don Melton: added ID3 version 2 TAG support</li> 1675159b3361Sopenharmony_ci <li><font color="#000000">John Dahlstrom: fixed 1676159b3361Sopenharmony_ci bug allowing timing information (for status in command line encoder) to 1677159b3361Sopenharmony_ci overflow.</font> 1678159b3361Sopenharmony_ci </li> 1679159b3361Sopenharmony_ci <li><font color="#000000">Tamito KAJIYAMA, Fixed 1680159b3361Sopenharmony_ci several bugs in the LAME/Vorbis interface.</font> 1681159b3361Sopenharmony_ci </li> 1682159b3361Sopenharmony_ci <li><font color="#000000">Mark Taylor: 1683159b3361Sopenharmony_ci lame --decode will recognize <a href="http://albumid.cjb.net">Album 1684159b3361Sopenharmony_ci ID tags</a></font> 1685159b3361Sopenharmony_ci </li> 1686159b3361Sopenharmony_ci <li><font color="#ff0000">Naoki 1687159b3361Sopenharmony_ci Shibata: Additive masking and other improvements to psycho 1688159b3361Sopenharmony_ci acoustics. (not yet on by default)</font> 1689159b3361Sopenharmony_ci </li> 1690159b3361Sopenharmony_ci </ul> 1691159b3361Sopenharmony_ci <h3> LAME 3.85beta July 3 2000</h3> 1692159b3361Sopenharmony_ci <ul> 1693159b3361Sopenharmony_ci <li><font color="#ff0000">Takehiro 1694159b3361Sopenharmony_ci Tominaga: mid/side stereo demasking thresholds updated.</font> 1695159b3361Sopenharmony_ci </li> 1696159b3361Sopenharmony_ci <li>Takehiro Tominaga: New short block MDCT coefficient data 1697159b3361Sopenharmony_ci structure. Should allow for future speed improvements. 1698159b3361Sopenharmony_ci </li> 1699159b3361Sopenharmony_ci <li>Robert Hegemann: fixed bug in old VBR routine, 1700159b3361Sopenharmony_ci the --noath mode messed up the VBR routine resulting in very large files 1701159b3361Sopenharmony_ci </li> 1702159b3361Sopenharmony_ci <li>Robert Hegemann: found bugs in some sections when using 32 1703159b3361Sopenharmony_ci bit floating point. Default is now back to 64bit floating 1704159b3361Sopenharmony_ci point. 1705159b3361Sopenharmony_ci </li> 1706159b3361Sopenharmony_ci <li><font color="#ff0000">Takehiro 1707159b3361Sopenharmony_ci Tominaga: Modified PE formula to use ATH.</font> 1708159b3361Sopenharmony_ci </li> 1709159b3361Sopenharmony_ci <li><font color="#000000">S.T.L.: 1710159b3361Sopenharmony_ci README.DJGPP - instructions for compiling LAME with DJGPP, the dos 1711159b3361Sopenharmony_ci version of gcc.</font> 1712159b3361Sopenharmony_ci </li> 1713159b3361Sopenharmony_ci </ul> 1714159b3361Sopenharmony_ci <h3> LAME 3.84beta June 30 2000</h3> 1715159b3361Sopenharmony_ci <ul> 1716159b3361Sopenharmony_ci <li>Mark Weinstein: .wav file output (with --decode 1717159b3361Sopenharmony_ci option) was writing the wrong filesize in the .wav file. Now 1718159b3361Sopenharmony_ci fixed. 1719159b3361Sopenharmony_ci </li> 1720159b3361Sopenharmony_ci <li>Mark Taylor: (optional) Vorbis support, both 1721159b3361Sopenharmony_ci encoding and decoding. LAME can now produce .ogg files, or 1722159b3361Sopenharmony_ci even re-encode your entire .ogg collection into 1723159b3361Sopenharmony_ci mp3. (Just kidding: it is always a bad idea to 1724159b3361Sopenharmony_ci convert from one lossy format to another) 1725159b3361Sopenharmony_ci </li> 1726159b3361Sopenharmony_ci <li>?: Bug fixed causing VBR to crash under 1727159b3361Sopenharmony_ci windows. (pretab[] array overflow) 1728159b3361Sopenharmony_ci </li> 1729159b3361Sopenharmony_ci <li>Sergey Sapelin: Another bug found in the mpg123 MPEG2 1730159b3361Sopenharmony_ci tables. Now fixed for the mpg123 based decoder in LAME. 1731159b3361Sopenharmony_ci </li> 1732159b3361Sopenharmony_ci <li>Marco Remondini: VBR histogram works in 1733159b3361Sopenharmony_ci win32. compile with -DBRHIST -DNOTERMCAP 1734159b3361Sopenharmony_ci </li> 1735159b3361Sopenharmony_ci <li><font color="#ff0000">Takehiro 1736159b3361Sopenharmony_ci Tominaga: LAME CBR will now use scalefac_scale to expand the 1737159b3361Sopenharmony_ci dynamic range of the scalefactors.</font> 1738159b3361Sopenharmony_ci </li> 1739159b3361Sopenharmony_ci <li><font color="#000000">Iwasa Kazmi: 1740159b3361Sopenharmony_ci Library improvements: exit()'s, printf, fprintf's are being 1741159b3361Sopenharmony_ci replaced by interceptable macros.</font> 1742159b3361Sopenharmony_ci </li> 1743159b3361Sopenharmony_ci </ul> 1744159b3361Sopenharmony_ci <h3> LAME 3.83beta May 19 2000</h3> 1745159b3361Sopenharmony_ci <ul> 1746159b3361Sopenharmony_ci <li><font color="#ff0000">Mark Taylor: 1747159b3361Sopenharmony_ci Bug in buffering routines: in some cases, could cause 1748159b3361Sopenharmony_ci MDCT to read past end of buffer. Rare in MPEG2, 1749159b3361Sopenharmony_ci even more rare for MPEG1, but potentially serious!</font> 1750159b3361Sopenharmony_ci </li> 1751159b3361Sopenharmony_ci <li>Mark Taylor: MDCT/polyphase filterbank was not 1752159b3361Sopenharmony_ci being "primed" properly. Does not effect output unless you 1753159b3361Sopenharmony_ci set the encoder delay lower than the default of 576 samples. 1754159b3361Sopenharmony_ci </li> 1755159b3361Sopenharmony_ci <li><font color="#ff0000">Mark Taylor: 1756159b3361Sopenharmony_ci "vdbj" and "Caster" found several VBR bugs (now 1757159b3361Sopenharmony_ci fixed): 1. Analog silence detection only 1758159b3361Sopenharmony_ci checked frequencies up to 16 kHz. 2. VBR mode could 1759159b3361Sopenharmony_ci still somehow avoid -F mode. 3. VBR mode would 1760159b3361Sopenharmony_ci ignore noise above 16 kHz (scalefactor band 22), Now calc_noise1 will 1761159b3361Sopenharmony_ci compute the noise in this band when in VBR mode. Not 1762159b3361Sopenharmony_ci calculated in CBR mode since CBR algorithm has no way of 1763159b3361Sopenharmony_ci using this information.</font> 1764159b3361Sopenharmony_ci </li> 1765159b3361Sopenharmony_ci <li>Mark Taylor: scalefactor band 22 info 1766159b3361Sopenharmony_ci (masking(=ATH), noise and energy) now displayed in frame 1767159b3361Sopenharmony_ci analyzer. 1768159b3361Sopenharmony_ci </li> 1769159b3361Sopenharmony_ci <li><font color="#ff0000">VBR code ATH tuning 1770159b3361Sopenharmony_ci was disabled by accident in 3.81, now fixed.</font> 1771159b3361Sopenharmony_ci </li> 1772159b3361Sopenharmony_ci <li><font color="#000000">Mark Taylor: 1773159b3361Sopenharmony_ci lame --decode will produce .wav files. (oops - size is off by 1774159b3361Sopenharmony_ci a factor of 4)</font> 1775159b3361Sopenharmony_ci </li> 1776159b3361Sopenharmony_ci </ul> 1777159b3361Sopenharmony_ci <h3> LAME 3.82beta May 11 2000</h3> 1778159b3361Sopenharmony_ci <ul> 1779159b3361Sopenharmony_ci <li>Robert Hegemann: Fixed bug in high bitrate joint 1780159b3361Sopenharmony_ci stereo encodings. 1781159b3361Sopenharmony_ci </li> 1782159b3361Sopenharmony_ci <li><font color="#3366ff">Naoki 1783159b3361Sopenharmony_ci Shibata: new long block MDCT routine</font> 1784159b3361Sopenharmony_ci </li> 1785159b3361Sopenharmony_ci </ul> 1786159b3361Sopenharmony_ci <h3> LAME 3.81beta May 8 2000</h3> 1787159b3361Sopenharmony_ci <ul> 1788159b3361Sopenharmony_ci <li>all ISO code removed!</li> 1789159b3361Sopenharmony_ci <li><font color="#3366ff">Takehiro Tominaga and 1790159b3361Sopenharmony_ci Naoki Shibata: new window subband routines.</font> 1791159b3361Sopenharmony_ci </li> 1792159b3361Sopenharmony_ci <li><font color="#000000">Naoki 1793159b3361Sopenharmony_ci Shibata: Bug fix in mpglib (decoding) lib: in some 1794159b3361Sopenharmony_ci cases, MDCT coefficients from previous granule was incorrectly used for 1795159b3361Sopenharmony_ci the next granule.</font> 1796159b3361Sopenharmony_ci </li> 1797159b3361Sopenharmony_ci <li><font color="#ff0000">ISO 7680 bit buffer 1798159b3361Sopenharmony_ci limitation removed. It can be reactivated with 1799159b3361Sopenharmony_ci "--strictly-enforce-ISO" Please report any trouble with high 1800159b3361Sopenharmony_ci bitrates.</font> 1801159b3361Sopenharmony_ci </li> 1802159b3361Sopenharmony_ci </ul> 1803159b3361Sopenharmony_ci <h3> LAME 3.80beta May 6 2000</h3> 1804159b3361Sopenharmony_ci <ul> 1805159b3361Sopenharmony_ci <li><font color="#ff0000">Takehiro 1806159b3361Sopenharmony_ci Tominaga: more efficient and faster huffman encoding!</font> 1807159b3361Sopenharmony_ci </li> 1808159b3361Sopenharmony_ci <li><font color="#ff0000">Takehiro Tominaga and 1809159b3361Sopenharmony_ci Mark Taylor: much improved short block compression!</font> 1810159b3361Sopenharmony_ci </li> 1811159b3361Sopenharmony_ci <li><font color="#000000">Tomasz Motylewski and 1812159b3361Sopenharmony_ci Mark Taylor: MPEG2.5 now supported!</font> 1813159b3361Sopenharmony_ci </li> 1814159b3361Sopenharmony_ci <li><font color="#000000">Mark Taylor: 1815159b3361Sopenharmony_ci incorporated Takehiro's bitstream.c! bitstream.c used by 1816159b3361Sopenharmony_ci default, but old ISO bitstream code can also be used.</font> 1817159b3361Sopenharmony_ci </li> 1818159b3361Sopenharmony_ci <li><font color="#ff0000">Scott Manley 1819159b3361Sopenharmony_ci and Mark Taylor: good resampling routine finaly in 1820159b3361Sopenharmony_ci LAME. uses a 19 point FIR filter with Blackman 1821159b3361Sopenharmony_ci window. Very slow for non integer resampling ratios.</font> 1822159b3361Sopenharmony_ci </li> 1823159b3361Sopenharmony_ci <li><font color="#000000">Iwasa Kazmi: 1824159b3361Sopenharmony_ci fixed SIGBUS error: VBR and id3 tags were using data after it 1825159b3361Sopenharmony_ci was free()'d.</font> 1826159b3361Sopenharmony_ci </li> 1827159b3361Sopenharmony_ci <li><font color="#ff0000">Robert 1828159b3361Sopenharmony_ci Hegemann: Improved VBR tuning. #define 1829159b3361Sopenharmony_ci RH_QUALITY_CONTROL and #RH_SIDE_VBR now the defaults.</font> 1830159b3361Sopenharmony_ci </li> 1831159b3361Sopenharmony_ci <li><font color="#000000">Robert 1832159b3361Sopenharmony_ci Hegemann: LAME version string now added to 1833159b3361Sopenharmony_ci ancillary data.</font> 1834159b3361Sopenharmony_ci </li> 1835159b3361Sopenharmony_ci <li>Kimmo Mustonen: VBR histogram support for Amiga.</li> 1836159b3361Sopenharmony_ci <li>Casper Gripenberg: VBR stats (but not histogram) 1837159b3361Sopenharmony_ci for DOS verson. 1838159b3361Sopenharmony_ci </li> 1839159b3361Sopenharmony_ci <li>Robert Hegemann: rare VBR overflow bug fixed.</li> 1840159b3361Sopenharmony_ci <li>Zack: -F option strictly enforces the VBR min 1841159b3361Sopenharmony_ci bitrate. Without -F, LAME will ignore the minimum bitrate 1842159b3361Sopenharmony_ci when encoding analog silence. 1843159b3361Sopenharmony_ci </li> 1844159b3361Sopenharmony_ci <li>Shawn Riley: User can now specify a compression 1845159b3361Sopenharmony_ci ratio (--comp <arg>) instead of a bit rate. 1846159b3361Sopenharmony_ci Default settings based on a compression ratio of 11.0 1847159b3361Sopenharmony_ci </li> 1848159b3361Sopenharmony_ci <li>Mark Taylor: free format bitstreams can be 1849159b3361Sopenharmony_ci created with --freeformat, and specify any integer bitrate from 8 to 1850159b3361Sopenharmony_ci 320kbs with -b. 1851159b3361Sopenharmony_ci </li> 1852159b3361Sopenharmony_ci <li>Mark Taylor: lame be used as a decoder (output raw pcm 1853159b3361Sopenharmony_ci only): lame --decode input.mp3 output.pcm 1854159b3361Sopenharmony_ci </li> 1855159b3361Sopenharmony_ci </ul> 1856159b3361Sopenharmony_ci <h3> LAME 3.70 April 6 2000</h3> 1857159b3361Sopenharmony_ci <ul> 1858159b3361Sopenharmony_ci <li>"LAME 3.69beta" becomes LAME 3.70 "stable"</li> 1859159b3361Sopenharmony_ci </ul> 1860159b3361Sopenharmony_ci <h3> LAME 3.69beta April 6 2000</h3> 1861159b3361Sopenharmony_ci <ul> 1862159b3361Sopenharmony_ci <li>"spahm": default mode selection bug 1863159b3361Sopenharmony_ci fixed. In some cases, lame was defaulting to regular stereo 1864159b3361Sopenharmony_ci instead of jstereo when the user did not specify a mode. 1865159b3361Sopenharmony_ci </li> 1866159b3361Sopenharmony_ci </ul> 1867159b3361Sopenharmony_ci <h3> LAME 3.68beta April 4 2000</h3> 1868159b3361Sopenharmony_ci <ul> 1869159b3361Sopenharmony_ci <li>Mark Taylor: mono encoding bug in DLL fixed.</li> 1870159b3361Sopenharmony_ci <li>Ingo Saitz: bug in --cwlimit argument parsing fixed.</li> 1871159b3361Sopenharmony_ci <li><font color="#ff0000">Scott Manly: bug in 1872159b3361Sopenharmony_ci 4-point resample code fixed.</font> 1873159b3361Sopenharmony_ci </li> 1874159b3361Sopenharmony_ci </ul> 1875159b3361Sopenharmony_ci <h3> LAME 3.67beta March 27 2000</h3> 1876159b3361Sopenharmony_ci <ul> 1877159b3361Sopenharmony_ci <li><font color="#ff0000">Robert 1878159b3361Sopenharmony_ci Hegemann: jstereo now enabled for MPEG2 encodings</font> 1879159b3361Sopenharmony_ci </li> 1880159b3361Sopenharmony_ci <li>Mark Taylor: old M/S stereo mode which used L/R maskings 1881159b3361Sopenharmony_ci has been removed. 1882159b3361Sopenharmony_ci </li> 1883159b3361Sopenharmony_ci <li>Mark Taylor: Xing MPEG2 VBR headers now working.</li> 1884159b3361Sopenharmony_ci <li><font color="#ff0000">Mark Taylor: 1885159b3361Sopenharmony_ci When quantized coefficients are all 0 in a band, set scalefactors to 0 1886159b3361Sopenharmony_ci also to save a few bits.</font> 1887159b3361Sopenharmony_ci </li> 1888159b3361Sopenharmony_ci <li><font color="#000000">Ingo Saitz: 1889159b3361Sopenharmony_ci Problems with framesize calculation when using -f fast-math option 1890159b3361Sopenharmony_ci fixed.</font> 1891159b3361Sopenharmony_ci </li> 1892159b3361Sopenharmony_ci </ul> 1893159b3361Sopenharmony_ci <h3> LAME 3.66beta March 21 2000</h3> 1894159b3361Sopenharmony_ci <ul> 1895159b3361Sopenharmony_ci <li>Bug fixes in BladeEnc DLL, possible click in last mp3 1896159b3361Sopenharmony_ci frame, VBR historgram display, byteswapping option, ASM quantize 1897159b3361Sopenharmony_ci routines work for both float and double. 1898159b3361Sopenharmony_ci </li> 1899159b3361Sopenharmony_ci </ul> 1900159b3361Sopenharmony_ci <h3> LAME 3.65beta March 17 2000</h3> 1901159b3361Sopenharmony_ci <ul> 1902159b3361Sopenharmony_ci <li>Enabled ASM version of quantize_xrpow() - accidently 1903159b3361Sopenharmony_ci disabled in lame3.64. 1904159b3361Sopenharmony_ci </li> 1905159b3361Sopenharmony_ci </ul> 1906159b3361Sopenharmony_ci <h3> LAME 3.64beta March 16 2000</h3> 1907159b3361Sopenharmony_ci <ul> 1908159b3361Sopenharmony_ci <li>Don Melton: id3v1.1 tags & id3 bugfixes</li> 1909159b3361Sopenharmony_ci <li><font color="#ff0000">Gabriel 1910159b3361Sopenharmony_ci Bouvigne: L/R matching block type fix</font> 1911159b3361Sopenharmony_ci </li> 1912159b3361Sopenharmony_ci <li><font color="#ff0000">Bug fixed which was 1913159b3361Sopenharmony_ci allowing quantized values to exceed the maximum when not using -h</font> 1914159b3361Sopenharmony_ci </li> 1915159b3361Sopenharmony_ci <li><font color="#3366ff">Mark Taylor: Fitlers 1916159b3361Sopenharmony_ci based on polyphase filterbank. should be slightly better 1917159b3361Sopenharmony_ci since the responce is independent of the blocktype, and they are 1918159b3361Sopenharmony_ci slightly faster.</font> 1919159b3361Sopenharmony_ci </li> 1920159b3361Sopenharmony_ci <li>Mark Taylor: API: the API changed slightly - and 1921159b3361Sopenharmony_ci this should be the final version. There is a new routine: 1922159b3361Sopenharmony_ci lame_encode_buffer() which takes an arbritray sized input buffer, 1923159b3361Sopenharmony_ci resamples & filters if necessary, encodes, and returns the 1924159b3361Sopenharmony_ci mp3buffer. There are also several new #defines, so it is 1925159b3361Sopenharmony_ci possible to compile a simple encoding library with no decoding or file 1926159b3361Sopenharmony_ci I/O or command line parsing. see the file API for details. 1927159b3361Sopenharmony_ci </li> 1928159b3361Sopenharmony_ci <li>Mark Taylor: MSVC stuff: lame.exe (with and 1929159b3361Sopenharmony_ci without the frame analyzer) and the CDex lame_enc.dll 1930159b3361Sopenharmony_ci should compile under MSVC. The MSVC5 project files may need 1931159b3361Sopenharmony_ci some tweaking. In particular, 1932159b3361Sopenharmony_ci you need to make sure LAMEPARSE, LAMESNDFILE and HAVEMPGLIB 1933159b3361Sopenharmony_ci are defined. (and HAVEGTK for the GTK stuff). 1934159b3361Sopenharmony_ci </li> 1935159b3361Sopenharmony_ci </ul> 1936159b3361Sopenharmony_ci <h3>LAME 3.63beta February 20 2000</h3> 1937159b3361Sopenharmony_ci <ul> 1938159b3361Sopenharmony_ci <li>Robert Hegemann: FPE with -h fixed?</li> 1939159b3361Sopenharmony_ci <li>Mathey Hendry: FPE error catching for Cygwin, 1940159b3361Sopenharmony_ci FPE fix for vbr mode and output to /dev/null 1941159b3361Sopenharmony_ci </li> 1942159b3361Sopenharmony_ci <li>Jeremy Hall: Fixed problems with input files 1943159b3361Sopenharmony_ci where the number of samples is not known. 1944159b3361Sopenharmony_ci </li> 1945159b3361Sopenharmony_ci <li><font color="#3366ff">Mathew 1946159b3361Sopenharmony_ci Hendry: ASM quantize_xrpow() for GNU i386</font> 1947159b3361Sopenharmony_ci </li> 1948159b3361Sopenharmony_ci <li><font color="#3366ff">Wilfried 1949159b3361Sopenharmony_ci Behne quantize_xrpow ()for PowerPC and non-ASM</font> 1950159b3361Sopenharmony_ci </li> 1951159b3361Sopenharmony_ci <li><font color="#3366ff">Takehiro 1952159b3361Sopenharmony_ci Tominaga: GOGO FFTs (not yet used?)</font> 1953159b3361Sopenharmony_ci </li> 1954159b3361Sopenharmony_ci </ul> 1955159b3361Sopenharmony_ci <h3>LAME 3.62beta February 9 2000</h3> 1956159b3361Sopenharmony_ci <ul> 1957159b3361Sopenharmony_ci <li><font color="#000000">Iwasa Kazmi: 1958159b3361Sopenharmony_ci frame analyzer short block display of single subblocks (press 1959159b3361Sopenharmony_ci 1,2 or 3)</font> 1960159b3361Sopenharmony_ci </li> 1961159b3361Sopenharmony_ci <li><font color="#000000">Ingo Saitz: 1962159b3361Sopenharmony_ci --help option added, with output to stdout</font> 1963159b3361Sopenharmony_ci </li> 1964159b3361Sopenharmony_ci <li><font color="#ff0000">Alfred Weyers: short 1965159b3361Sopenharmony_ci block AAC spreading function bug fixed</font> 1966159b3361Sopenharmony_ci </li> 1967159b3361Sopenharmony_ci <li><font color="#3366ff">Takehiro 1968159b3361Sopenharmony_ci Tominaga: new scalefac data structure - improves performance!</font> 1969159b3361Sopenharmony_ci </li> 1970159b3361Sopenharmony_ci <li><font color="#ff0000">Lionel 1971159b3361Sopenharmony_ci Bonnet: Bug fixed in MPEG2 scalefactor routine: scalefactors 1972159b3361Sopenharmony_ci were being severly limited.</font> 1973159b3361Sopenharmony_ci </li> 1974159b3361Sopenharmony_ci <li><font color="#3366ff">Takehiro 1975159b3361Sopenharmony_ci Tominaga: faster FFT routines from. These routines 1976159b3361Sopenharmony_ci are also compatible with the GOGO routines, in case someone is 1977159b3361Sopenharmony_ci interested in porting them back to LAME.</font> 1978159b3361Sopenharmony_ci </li> 1979159b3361Sopenharmony_ci <li><font color="#3366ff">Sigbjørn 1980159b3361Sopenharmony_ci Skjæret, Takehiro Tominaga: faster pow() code.</font> 1981159b3361Sopenharmony_ci </li> 1982159b3361Sopenharmony_ci <li><font color="#ff0000">Joachim 1983159b3361Sopenharmony_ci Kuebart: Found some unitialized variables that were effecting 1984159b3361Sopenharmony_ci quality for encodings which did not use the -h option (now fixed).</font> 1985159b3361Sopenharmony_ci </li> 1986159b3361Sopenharmony_ci <li>Mark Taylor: More modularization work. 1987159b3361Sopenharmony_ci It is now possible to use LAME as a library where you can set the 1988159b3361Sopenharmony_ci encoding parameters directly and do your own file 1989159b3361Sopenharmony_ci i/o. The calling program is now it's own mp3 1990159b3361Sopenharmony_ci output. For an example of the LAME API, see main.c, or 1991159b3361Sopenharmony_ci mp3rtp.c or mp3x.c. These can all be compiled as stand alone 1992159b3361Sopenharmony_ci programs which link with libmp3lame.a. 1993159b3361Sopenharmony_ci </li> 1994159b3361Sopenharmony_ci <li>Felix vos Leitner: mp3rtp fixes. 1995159b3361Sopenharmony_ci mp3rtp is a standalone program which will encode and stream with RTP. 1996159b3361Sopenharmony_ci </li> 1997159b3361Sopenharmony_ci <li>Robert Hegemann: Information written to stderr 1998159b3361Sopenharmony_ci displaying exactly which type of lowpass filter (if any) is being used. 1999159b3361Sopenharmony_ci </li> 2000159b3361Sopenharmony_ci <li>Iwasa Kazmi: mpglib (the mpg123 decoder) scsfi 2001159b3361Sopenharmony_ci decoding fixes. 2002159b3361Sopenharmony_ci </li> 2003159b3361Sopenharmony_ci <li>Takehiro Tominaga: More mpglib scsfi decoding 2004159b3361Sopenharmony_ci fixes. 2005159b3361Sopenharmony_ci </li> 2006159b3361Sopenharmony_ci </ul> 2007159b3361Sopenharmony_ci <h3>LAME 3.61beta January 14 2000</h3> 2008159b3361Sopenharmony_ci <ul> 2009159b3361Sopenharmony_ci <li><font color="#ff0000">Mark Taylor: Fixed bug 2010159b3361Sopenharmony_ci with lowpass filters when using VBR with a 64kbs or lower min bitrate 2011159b3361Sopenharmony_ci setting.</font> 2012159b3361Sopenharmony_ci </li> 2013159b3361Sopenharmony_ci <li><font color="#ff0000">Takehiro 2014159b3361Sopenharmony_ci Tominaga: more efficient huffman encoding splitting.</font> 2015159b3361Sopenharmony_ci </li> 2016159b3361Sopenharmony_ci </ul> 2017159b3361Sopenharmony_ci <h3> LAME 3.60beta January 9 2000</h3> 2018159b3361Sopenharmony_ci <ul> 2019159b3361Sopenharmony_ci <li>Mark Taylor: Distribution now comes with self 2020159b3361Sopenharmony_ci test. Needs work to be automated, see 'make test' in Makefile. 2021159b3361Sopenharmony_ci </li> 2022159b3361Sopenharmony_ci <li><font color="#ff0000">Mark Taylor: AAC 2023159b3361Sopenharmony_ci spreading function now the default</font> 2024159b3361Sopenharmony_ci </li> 2025159b3361Sopenharmony_ci <li>Gabriel Bouvigne: updated HTML docs</li> 2026159b3361Sopenharmony_ci <li>Felix von Leitner: compute correct file length from Xing 2027159b3361Sopenharmony_ci header (if present) when input file is a mp3 file 2028159b3361Sopenharmony_ci </li> 2029159b3361Sopenharmony_ci <li>Felix von Leitner: mp3rtp (standalone) program now 2030159b3361Sopenharmony_ci included. Not yet tested. mp3rtp 2031159b3361Sopenharmony_ci ip:port:ttl <infile> 2032159b3361Sopenharmony_ci /dev/null will stream directly to ip:port using RTP. 2033159b3361Sopenharmony_ci </li> 2034159b3361Sopenharmony_ci </ul> 2035159b3361Sopenharmony_ci <h3>LAME 3.59beta January 4 2000</h3> 2036159b3361Sopenharmony_ci <ul> 2037159b3361Sopenharmony_ci <li>Takehiro Tominaga: --noath option. 2038159b3361Sopenharmony_ci Disables ATH maskings. 2039159b3361Sopenharmony_ci </li> 2040159b3361Sopenharmony_ci <li>Gabriel Bouvigne: updated HTML docs.</li> 2041159b3361Sopenharmony_ci <li>Iwasa Kazmi: makefile fixes</li> 2042159b3361Sopenharmony_ci <li>Mark Taylor: Fixed bug where first frame of data 2043159b3361Sopenharmony_ci was always overwritten with 0's. Thanks to 'gol' 2044159b3361Sopenharmony_ci </li> 2045159b3361Sopenharmony_ci <li><font color="#ff0000">Mark Taylor: 2046159b3361Sopenharmony_ci bug fixes in mid/side masking ratios (thanks to Menno Bakker)</font> 2047159b3361Sopenharmony_ci </li> 2048159b3361Sopenharmony_ci <li>Mark Taylor: replaced norm_l, norm_s table data 2049159b3361Sopenharmony_ci with formulas. 2050159b3361Sopenharmony_ci </li> 2051159b3361Sopenharmony_ci </ul> 2052159b3361Sopenharmony_ci <h3>LAME 3.58beta December 13 1999</h3> 2053159b3361Sopenharmony_ci <ul> 2054159b3361Sopenharmony_ci <li><font color="#ff0000">Segher 2055159b3361Sopenharmony_ci Boessenkool: More accurate quantization procedure! 2056159b3361Sopenharmony_ci Enabled with -h.</font> 2057159b3361Sopenharmony_ci </li> 2058159b3361Sopenharmony_ci <li><font color="#3366ff">Mathew Hendry, Acy 2059159b3361Sopenharmony_ci Stapp and Takehiro Tominaga: ASM optimizations for quantize_xrpow and 2060159b3361Sopenharmony_ci quantize_xrpow_ISO.</font> 2061159b3361Sopenharmony_ci </li> 2062159b3361Sopenharmony_ci <li>Chuck Zenkus: "encoder inside" logo on web page</li> 2063159b3361Sopenharmony_ci <li>Mark Taylor: a couple people have 2064159b3361Sopenharmony_ci asked for this. Allow LAME to overide 2065159b3361Sopenharmony_ci VBR_min_bitrate if analog_silence detected. 2066159b3361Sopenharmony_ci Analog_silence defined a la Robert: 2067159b3361Sopenharmony_ci energy < ATH. 2068159b3361Sopenharmony_ci </li> 2069159b3361Sopenharmony_ci <li>An Van Lam: Valid bitrates were being printed for layer 2, 2070159b3361Sopenharmony_ci not layer 3! 2071159b3361Sopenharmony_ci </li> 2072159b3361Sopenharmony_ci <li>Ethan Yeo: Makefile.MSVC updated</li> 2073159b3361Sopenharmony_ci <li>Mark Stephens: updated all MSVC project files</li> 2074159b3361Sopenharmony_ci <li>Robert Hegemann: lowpass and highpass filters 2075159b3361Sopenharmony_ci can be enabled with --lowpass, --highpass 2076159b3361Sopenharmony_ci </li> 2077159b3361Sopenharmony_ci <li><font color="#ff0000">Mark Taylor: 2078159b3361Sopenharmony_ci MS switching is now smoother: ms_ratio average over 4 granules</font> 2079159b3361Sopenharmony_ci </li> 2080159b3361Sopenharmony_ci <li><font color="#ff0000">Takehiro 2081159b3361Sopenharmony_ci Tominaga: Scalefactor pre-emphasis fixed (and now turned back 2082159b3361Sopenharmony_ci on)</font> 2083159b3361Sopenharmony_ci </li> 2084159b3361Sopenharmony_ci <li><font color="#ff0000">Takehiro 2085159b3361Sopenharmony_ci Tominaga: Bug in M/S maskings: switch to turn on 2086159b3361Sopenharmony_ci stereo demasking code was buggy.</font> 2087159b3361Sopenharmony_ci </li> 2088159b3361Sopenharmony_ci </ul> 2089159b3361Sopenharmony_ci <h3>LAME 3.57beta November 22 1999</h3> 2090159b3361Sopenharmony_ci <ul> 2091159b3361Sopenharmony_ci <li>Sigbjørn Skjæret, patch to allow 2092159b3361Sopenharmony_ci encoding from 8bit input files when using LIBSNDFILE 2093159b3361Sopenharmony_ci </li> 2094159b3361Sopenharmony_ci <li>Mark Taylor: Automatic downsampling to nearest valid 2095159b3361Sopenharmony_ci samplerate. 2096159b3361Sopenharmony_ci </li> 2097159b3361Sopenharmony_ci <li>Mark Taylor: Scalefactor bands demarked on MDCT plot in 2098159b3361Sopenharmony_ci frameanalyzer 2099159b3361Sopenharmony_ci </li> 2100159b3361Sopenharmony_ci <li>Mark Taylor: Scalefactor preemphasis disabled for 2101159b3361Sopenharmony_ci now. The algorithm was often doing more harm than 2102159b3361Sopenharmony_ci good. 2103159b3361Sopenharmony_ci </li> 2104159b3361Sopenharmony_ci </ul> 2105159b3361Sopenharmony_ci <h3>LAME 3.56beta November 19 1999</h3> 2106159b3361Sopenharmony_ci <ul> 2107159b3361Sopenharmony_ci <li>Kimmo Mustonen: portabilty code cleanup.</li> 2108159b3361Sopenharmony_ci <li>Vladimir Marek: id3 genre patch.</li> 2109159b3361Sopenharmony_ci <li>Conrad Sanderson: new applypatch script.</li> 2110159b3361Sopenharmony_ci <li>Mark Taylor: Initial window type now "STOP_TYPE" to reduce 2111159b3361Sopenharmony_ci initial attenuation. This is needed because the new encoder 2112159b3361Sopenharmony_ci delay is so short. With a NORM_TYPE, the first 240 samples 2113159b3361Sopenharmony_ci would be attenuated. 2114159b3361Sopenharmony_ci </li> 2115159b3361Sopenharmony_ci <li>Mark Taylor: Padding at end of file now adjusted 2116159b3361Sopenharmony_ci (hopefully!) to produce as little padding as possible while still 2117159b3361Sopenharmony_ci guarantee all input samples are encoded. 2118159b3361Sopenharmony_ci </li> 2119159b3361Sopenharmony_ci <li><font color="#ff0000">Takehiro 2120159b3361Sopenharmony_ci Tominaga: Reduced shortblock extra bit allocation formulas by 2121159b3361Sopenharmony_ci 10% since new huffman coding is at least 10% more efficient.</font> 2122159b3361Sopenharmony_ci </li> 2123159b3361Sopenharmony_ci </ul> 2124159b3361Sopenharmony_ci <h3>LAME 3.55beta November 11 1999</h3> 2125159b3361Sopenharmony_ci <ul> 2126159b3361Sopenharmony_ci <li>Albert Faber: updated BladeEnc.dll</li> 2127159b3361Sopenharmony_ci <li> 2128159b3361Sopenharmony_ci Mark Taylor: Simple lowpass filter added to linear 2129159b3361Sopenharmony_ci downsampling routine. 2130159b3361Sopenharmony_ci </li> 2131159b3361Sopenharmony_ci <li>Nils Faerber: updated man page.</li> 2132159b3361Sopenharmony_ci <li> 2133159b3361Sopenharmony_ci Mark Taylor: All floating point variables are delcared 2134159b3361Sopenharmony_ci FLOAT or FLOAT8. Change the definition of FLOAT8 in 2135159b3361Sopenharmony_ci machine.h to run at 32bit preceision. 2136159b3361Sopenharmony_ci </li> 2137159b3361Sopenharmony_ci <li> 2138159b3361Sopenharmony_ci Mark Taylor: Bug (introduced in 3.54beta) in 2139159b3361Sopenharmony_ci stereo->mono downsampling fixed. 2140159b3361Sopenharmony_ci </li> 2141159b3361Sopenharmony_ci </ul> 2142159b3361Sopenharmony_ci <h3>LAME 3.54beta November 8 1999</h3> 2143159b3361Sopenharmony_ci <ul> 2144159b3361Sopenharmony_ci <li> 2145159b3361Sopenharmony_ci Mark Taylor: Encoder delay is now 48 samples. Can be adjusted to 2146159b3361Sopenharmony_ci 1160 to sync with FhG (see ENCDELAY in encoder.h) This is kind 2147159b3361Sopenharmony_ci of amazing, since if Takehiro put his MDCT/filterbank routine in 2148159b3361Sopenharmony_ci a decoder, we could have a total delay of only 96 samples. 2149159b3361Sopenharmony_ci </li> 2150159b3361Sopenharmony_ci <li> 2151159b3361Sopenharmony_ci <font color="#ff0000">Mark Taylor: More inconstancies found and 2152159b3361Sopenharmony_ci fixed in MPEG2 tables.</font> 2153159b3361Sopenharmony_ci </li> 2154159b3361Sopenharmony_ci <li> 2155159b3361Sopenharmony_ci Mark Taylor: Resampling from an MP3 input file now works. But we 2156159b3361Sopenharmony_ci still dont have a lowpass filter so dont expect good results. 2157159b3361Sopenharmony_ci </li> 2158159b3361Sopenharmony_ci </ul> 2159159b3361Sopenharmony_ci <h3>LAME 3.53beta November 8 1999</h3> 2160159b3361Sopenharmony_ci <ul> 2161159b3361Sopenharmony_ci <li> 2162159b3361Sopenharmony_ci <font color="#3366ff">Takehiro Tominaga: Fixed MPEG2 problem in 2163159b3361Sopenharmony_ci new MDCT routines. Takehiro's combined filterbank/MDCT routine is 2164159b3361Sopenharmony_ci now the default. Removes all buffering from psymodel.c and the 2165159b3361Sopenharmony_ci filterbanks/MDCT routines.</font> 2166159b3361Sopenharmony_ci </li> 2167159b3361Sopenharmony_ci </ul> 2168159b3361Sopenharmony_ci <h3>LAME 3.52beta November 8 1999</h3> 2169159b3361Sopenharmony_ci <ul> 2170159b3361Sopenharmony_ci <li> 2171159b3361Sopenharmony_ci By permission of copyright holders of all GPL code in 2172159b3361Sopenharmony_ci LAME, all GPL code is now released under a modified version 2173159b3361Sopenharmony_ci of the LGPL (see the README file) 2174159b3361Sopenharmony_ci </li> 2175159b3361Sopenharmony_ci <li> 2176159b3361Sopenharmony_ci By popular demand, all C++ comments changed to C style 2177159b3361Sopenharmony_ci comments 2178159b3361Sopenharmony_ci </li> 2179159b3361Sopenharmony_ci <li> 2180159b3361Sopenharmony_ci Mark Taylor: Linear resampling now works. Use 2181159b3361Sopenharmony_ci --resample to set an output samplerate different from the input 2182159b3361Sopenharmony_ci samplerate. (doesn't seem to work with mp3 input files, and 2183159b3361Sopenharmony_ci there is no lowpass filter, so dont expect good results just yet) 2184159b3361Sopenharmony_ci </li> 2185159b3361Sopenharmony_ci <li> 2186159b3361Sopenharmony_ci <font color="#3366ff">Takehiro Tominaga: Faster Huffman encoding 2187159b3361Sopenharmony_ci routines</font> 2188159b3361Sopenharmony_ci </li> 2189159b3361Sopenharmony_ci </ul> 2190159b3361Sopenharmony_ci <font color="#3366ff">The following changes are disabled 2191159b3361Sopenharmony_ci because of MPEG2 problems. But to try them, set MDCTDELAY=48 2192159b3361Sopenharmony_ci in encoder.h, instead of MDCTDELAY=528.:</font> 2193159b3361Sopenharmony_ci <ul> 2194159b3361Sopenharmony_ci <li> 2195159b3361Sopenharmony_ci <font color="#3366ff">Takehiro Tominaga: New MDCT routines with 2196159b3361Sopenharmony_ci shorter delay (48 samples instead of 528) and even faster than 2197159b3361Sopenharmony_ci the old routines.</font> 2198159b3361Sopenharmony_ci </li> 2199159b3361Sopenharmony_ci <li> 2200159b3361Sopenharmony_ci <font color="#3366ff">Takehiro Tominaga: Removed extra buffering 2201159b3361Sopenharmony_ci in psymodel.c</font> 2202159b3361Sopenharmony_ci </li> 2203159b3361Sopenharmony_ci </ul> 2204159b3361Sopenharmony_ci <h3>LAME 3.51 November 7 1999</h3> 2205159b3361Sopenharmony_ci <ul> 2206159b3361Sopenharmony_ci <li> 2207159b3361Sopenharmony_ci Takehiro Tominaga: Bug in quantize.c absolute threshold of 2208159b3361Sopenharmony_ci hearing calculation for non-44.1 kHz input files. 2209159b3361Sopenharmony_ci </li> 2210159b3361Sopenharmony_ci </ul> 2211159b3361Sopenharmony_ci <h3>LAME 3.50 November 1 1999</h3> 2212159b3361Sopenharmony_ci <ul> 2213159b3361Sopenharmony_ci <li>LAME 3.37beta becomes official LAME 3.50 release</li> 2214159b3361Sopenharmony_ci </ul> 2215159b3361Sopenharmony_ci <h3>LAME 3.37beta November 1 1999</h3> 2216159b3361Sopenharmony_ci <ul> 2217159b3361Sopenharmony_ci <li> 2218159b3361Sopenharmony_ci <font color="#ff0000">Lionel Bonnet: Found severe bug in MPEG2 2219159b3361Sopenharmony_ci Short block SNR.</font> 2220159b3361Sopenharmony_ci </li> 2221159b3361Sopenharmony_ci <li>Sergey Sapelin: VBR Toc improvement.</li> 2222159b3361Sopenharmony_ci <li>Sergey Dubov: fskip() routine</li> 2223159b3361Sopenharmony_ci <li> 2224159b3361Sopenharmony_ci Conrad Sanderson: replacement for filterbank.c. Not much faster 2225159b3361Sopenharmony_ci but amazingly simpler. 2226159b3361Sopenharmony_ci </li> 2227159b3361Sopenharmony_ci </ul> 2228159b3361Sopenharmony_ci <h3>LAME 3.36beta October 25 1999</h3> 2229159b3361Sopenharmony_ci <ul> 2230159b3361Sopenharmony_ci <li>Albert Faber: more MSVC and BladeDLL updates</li> 2231159b3361Sopenharmony_ci <li>Kimmo Mustonen: Much code cleanup and Amiga updates</li> 2232159b3361Sopenharmony_ci <li>Anton Oleynikov: Borland C updates</li> 2233159b3361Sopenharmony_ci <li> 2234159b3361Sopenharmony_ci Mark Taylor: More stdin fixes: For some reason, forward 2235159b3361Sopenharmony_ci fseek()'s would fail when used on pipes even though it is okay 2236159b3361Sopenharmony_ci with redirection from "<". So I changed all the forward 2237159b3361Sopenharmony_ci fseek()'s to use fread(). This should improve stdin support for 2238159b3361Sopenharmony_ci wav/aiff files. If you know the input file is raw pcm, you can 2239159b3361Sopenharmony_ci still use the '-r' option to avoid *all* seeking of any kind. 2240159b3361Sopenharmony_ci </li> 2241159b3361Sopenharmony_ci </ul> 2242159b3361Sopenharmony_ci <h3>LAME 3.35beta October 21 1999</h3> 2243159b3361Sopenharmony_ci <ul> 2244159b3361Sopenharmony_ci <li> 2245159b3361Sopenharmony_ci <font color="#ff0000">Leonid Kulakov: Serious bug in MPEG2 2246159b3361Sopenharmony_ci scalefactor band tables fixed.</font> 2247159b3361Sopenharmony_ci </li> 2248159b3361Sopenharmony_ci <li> 2249159b3361Sopenharmony_ci Portability patches from: Anton Oleynikov, Sigbjørn 2250159b3361Sopenharmony_ci Skjæret, Mathew Hendry, Richard Gorton 2251159b3361Sopenharmony_ci </li> 2252159b3361Sopenharmony_ci <li>Alfred Weyers: compiler options, updated timestatus.</li> 2253159b3361Sopenharmony_ci <li>Albert Faber: BladeDll and other updates (new machine.h).</li> 2254159b3361Sopenharmony_ci <li>Monty: updated Makefile to fix gcc inline math bug.</li> 2255159b3361Sopenharmony_ci </ul> 2256159b3361Sopenharmony_ci <h3>LAME 3.34beta October 12 1999</h3> 2257159b3361Sopenharmony_ci <ul> 2258159b3361Sopenharmony_ci <li> 2259159b3361Sopenharmony_ci <font color="#ff0000">Mark Taylor: Bug fixed: minimum bitrate in 2260159b3361Sopenharmony_ci VBR mode could be ignored for a few frames.</font> 2261159b3361Sopenharmony_ci </li> 2262159b3361Sopenharmony_ci <li><font color="#ff0000">Mark Taylor: New (minor) VBR tunings.</font></li> 2263159b3361Sopenharmony_ci <li> 2264159b3361Sopenharmony_ci Tim Ruddick: New wav/aiff header parsing routines. Better parsing 2265159b3361Sopenharmony_ci and fewer fseek()'s. 2266159b3361Sopenharmony_ci </li> 2267159b3361Sopenharmony_ci <li>Anton Oleynikov: patches to work with Borland C</li> 2268159b3361Sopenharmony_ci <li> 2269159b3361Sopenharmony_ci <font color="#ff0000">Gabriel Bouvigne: Experimental voice option 2270159b3361Sopenharmony_ci enabled with --voice</font> 2271159b3361Sopenharmony_ci </li> 2272159b3361Sopenharmony_ci </ul> 2273159b3361Sopenharmony_ci <h3>LAME 3.33beta October 11 1999</h3> 2274159b3361Sopenharmony_ci <ul> 2275159b3361Sopenharmony_ci <li> 2276159b3361Sopenharmony_ci <font color="#ff0000">Robert Hegemann: RH VBR mode now the 2277159b3361Sopenharmony_ci default and only VBR mode. The new code will always quantize to 2278159b3361Sopenharmony_ci 0 distortion and the quality is increased by reducing the 2279159b3361Sopenharmony_ci masking from the psy-model. -X0 is still the default for now.</font> 2280159b3361Sopenharmony_ci </li> 2281159b3361Sopenharmony_ci <li><font color="#ff0000">Robert Hegemann: new -X5 mode</font></li> 2282159b3361Sopenharmony_ci <li>Mathew Hendry: New timing code, removes the need for HAVETIMES</li> 2283159b3361Sopenharmony_ci <li> 2284159b3361Sopenharmony_ci <font color="#3366ff">Mathew 2285159b3361Sopenharmony_ci Hendry: assembler quantize_xrpow for Windows</font> 2286159b3361Sopenharmony_ci </li> 2287159b3361Sopenharmony_ci <li>Iwasa Kazmi: stdin/stdout patch for Windows</li> 2288159b3361Sopenharmony_ci <li> 2289159b3361Sopenharmony_ci Mark Taylor: New option: "--athonly" will ignore the psy-model 2290159b3361Sopenharmony_ci output and use only the absolute threshold of hearing for the 2291159b3361Sopenharmony_ci masking. 2292159b3361Sopenharmony_ci </li> 2293159b3361Sopenharmony_ci </ul> 2294159b3361Sopenharmony_ci <h3>LAME 3.32beta October 8 1999</h3> 2295159b3361Sopenharmony_ci <ul> 2296159b3361Sopenharmony_ci <li> 2297159b3361Sopenharmony_ci <font color="#3366ff">Takehiro Tominaga: faster long block 2298159b3361Sopenharmony_ci spreading function convolution for non 44.1 kHz sampling 2299159b3361Sopenharmony_ci frequencies, and faster short block spreading function 2300159b3361Sopenharmony_ci convolution for all sampling frequencies.</font> 2301159b3361Sopenharmony_ci </li> 2302159b3361Sopenharmony_ci <li> 2303159b3361Sopenharmony_ci <font color="#ff0000">Takehiro Tominaga: Completly rewritten 2304159b3361Sopenharmony_ci huffman table selection and count_bits(). More efficient table 2305159b3361Sopenharmony_ci selection results in many more bits per frame.</font> 2306159b3361Sopenharmony_ci </li> 2307159b3361Sopenharmony_ci <li> 2308159b3361Sopenharmony_ci <font color="#ff0000">Takehiro Tominaga: More efficient scalefac 2309159b3361Sopenharmony_ci compress setting.</font> 2310159b3361Sopenharmony_ci </li> 2311159b3361Sopenharmony_ci <li><font color="#3366ff">Mike Cheng: new calc_noise2()</font></li> 2312159b3361Sopenharmony_ci <li>Alfred Weyers: patch for timestatus() seconds rollover</li> 2313159b3361Sopenharmony_ci </ul> 2314159b3361Sopenharmony_ci <h3>LAME 3.31beta September 28 1999</h3> 2315159b3361Sopenharmony_ci <ul> 2316159b3361Sopenharmony_ci <li> 2317159b3361Sopenharmony_ci Albert Faber: updated his BladeDLL code. This allows LAME to be 2318159b3361Sopenharmony_ci compiled into a BladeEnc compatiable .dll. 2319159b3361Sopenharmony_ci </li> 2320159b3361Sopenharmony_ci <li> 2321159b3361Sopenharmony_ci <font color="#3366ff">Mike Cheng: faster l3psycho_ener() 2322159b3361Sopenharmony_ci routine.</font> 2323159b3361Sopenharmony_ci </li> 2324159b3361Sopenharmony_ci <li>Sigbjørn Skjæret: more code cleanup.</li> 2325159b3361Sopenharmony_ci </ul> 2326159b3361Sopenharmony_ci <h3>LAME 3.30beta September 27 1999</h3> 2327159b3361Sopenharmony_ci <ul> 2328159b3361Sopenharmony_ci <li> 2329159b3361Sopenharmony_ci Conrad Sanderson: ID3 tag code added (type 'lame' for 2330159b3361Sopenharmony_ci instructions) 2331159b3361Sopenharmony_ci </li> 2332159b3361Sopenharmony_ci <li>new mdct.c from Mike Cheng (no faster, but much cleaner code)</li> 2333159b3361Sopenharmony_ci <li> 2334159b3361Sopenharmony_ci Mathew Hendry: Microsoft nmake makefile and a couple other 2335159b3361Sopenharmony_ci changes for MSVC 2336159b3361Sopenharmony_ci </li> 2337159b3361Sopenharmony_ci <li> 2338159b3361Sopenharmony_ci More modulization work: One input sound file interface handles 2339159b3361Sopenharmony_ci mp3's, uncompressed audio, with or without LIBSNDFILE. Fixes 2340159b3361Sopenharmony_ci (hopefully) a bunch of file I/O bugs introduced in 3.29 (Mark Taylor) 2341159b3361Sopenharmony_ci </li> 2342159b3361Sopenharmony_ci <li> 2343159b3361Sopenharmony_ci LAME will now print valid samplerate/bitrate combinations 2344159b3361Sopenharmony_ci (Mark Taylor) 2345159b3361Sopenharmony_ci </li> 2346159b3361Sopenharmony_ci <li>stdin/stdout fix for OS/2 (Paul Hartman)</li> 2347159b3361Sopenharmony_ci <li> 2348159b3361Sopenharmony_ci For mp3 input files, total frames estimated based on 2349159b3361Sopenharmony_ci filesize and first frame bitrate. (Mark Taylor) 2350159b3361Sopenharmony_ci </li> 2351159b3361Sopenharmony_ci <li> 2352159b3361Sopenharmony_ci Updated all functions with new style prototypes. 2353159b3361Sopenharmony_ci (Sigbjørn Skjæret) 2354159b3361Sopenharmony_ci </li> 2355159b3361Sopenharmony_ci </ul> 2356159b3361Sopenharmony_ci <h3>LAME 3.29beta September 21 1999</h3> 2357159b3361Sopenharmony_ci <ul> 2358159b3361Sopenharmony_ci <li> 2359159b3361Sopenharmony_ci <font color="#ff0000">Bug in bigv_bitcount fixed. Loop.c was 2360159b3361Sopenharmony_ci overestimating the number of bits needed, resulting in wasted 2361159b3361Sopenharmony_ci bits every frame. (Leonid A. Kulakov)</font> 2362159b3361Sopenharmony_ci </li> 2363159b3361Sopenharmony_ci <li> 2364159b3361Sopenharmony_ci <font color="#ff0000">Bug in *_choose_table() fixed These 2365159b3361Sopenharmony_ci routines would not sellect the optimal Huffman table in some 2366159b3361Sopenharmony_ci cases. (Leonid A. Kulakov)</font> 2367159b3361Sopenharmony_ci </li> 2368159b3361Sopenharmony_ci <li><font color="#ff0000">Tuning of ATH normalization (macik)</font></li> 2369159b3361Sopenharmony_ci <li> 2370159b3361Sopenharmony_ci Removed unused variables and fixed function prototypes 2371159b3361Sopenharmony_ci (Sigbjørn Skjæret) 2372159b3361Sopenharmony_ci </li> 2373159b3361Sopenharmony_ci <li> 2374159b3361Sopenharmony_ci Sami Farin sent a .wav file that LAME built in support choked 2375159b3361Sopenharmony_ci on. I added a slightly more sophisticated wav header parsing to 2376159b3361Sopenharmony_ci handle this, but if you have trouble, use libsndfile. 2377159b3361Sopenharmony_ci </li> 2378159b3361Sopenharmony_ci <li> 2379159b3361Sopenharmony_ci Resampling hooks and options added. Buffering 2380159b3361Sopenharmony_ci and resampling routines need to be written. 2381159b3361Sopenharmony_ci </li> 2382159b3361Sopenharmony_ci <li> 2383159b3361Sopenharmony_ci LAME will now take an mp3 file as input. When resampling code 2384159b3361Sopenharmony_ci is working, LAME will be able to (for example) convert a high 2385159b3361Sopenharmony_ci bitrate stereo mp3 to a low bitrate mono mp3 for streaming. 2386159b3361Sopenharmony_ci </li> 2387159b3361Sopenharmony_ci </ul> 2388159b3361Sopenharmony_ci <h3>LAME 3.28beta September 15 1999</h3> 2389159b3361Sopenharmony_ci <ul> 2390159b3361Sopenharmony_ci <li> 2391159b3361Sopenharmony_ci <font color="#ff0000">Serious bug fixed in 2392159b3361Sopenharmony_ci high frequency MDCT coefficients. Huffman coding was 2393159b3361Sopenharmony_ci reversing the order of the count1 block quadruples. 2394159b3361Sopenharmony_ci (Leonid A. Kulakov)</font> 2395159b3361Sopenharmony_ci </li> 2396159b3361Sopenharmony_ci <li> 2397159b3361Sopenharmony_ci nint() problems under Tru64 unix fixed and preprocessor 2398159b3361Sopenharmony_ci variable HAVE_NINT removed. (Bob Bell) 2399159b3361Sopenharmony_ci </li> 2400159b3361Sopenharmony_ci <li> 2401159b3361Sopenharmony_ci Compiler warning fixes and code cleanup (Sigbjørn 2402159b3361Sopenharmony_ci Skjæret, Lionel Bonnet) 2403159b3361Sopenharmony_ci </li> 2404159b3361Sopenharmony_ci <li> 2405159b3361Sopenharmony_ci USAGE file now includes suggestions for 2406159b3361Sopenharmony_ci downsampling. For low bitrate encodings, proper downsampling 2407159b3361Sopenharmony_ci can give dramatically better results. (John Hayward-Warburton) 2408159b3361Sopenharmony_ci </li> 2409159b3361Sopenharmony_ci </ul> 2410159b3361Sopenharmony_ci <h3>LAME 3.27beta September 12 1999</h3> 2411159b3361Sopenharmony_ci <ul> 2412159b3361Sopenharmony_ci <li> 2413159b3361Sopenharmony_ci Several bugs in encode.c and l3bitstream.c fixed by Lionel 2414159b3361Sopenharmony_ci Bonnet. 2415159b3361Sopenharmony_ci </li> 2416159b3361Sopenharmony_ci <li> 2417159b3361Sopenharmony_ci Bugs in new VBR (#define RH) formula for mono input file 2418159b3361Sopenharmony_ci and mid/side encoding fixed. 2419159b3361Sopenharmony_ci </li> 2420159b3361Sopenharmony_ci </ul> 2421159b3361Sopenharmony_ci <h3>LAME 3.26beta September 10 1999</h3> 2422159b3361Sopenharmony_ci <ul> 2423159b3361Sopenharmony_ci <li> 2424159b3361Sopenharmony_ci The "-m m" option (mono .mp3 file) will automatically mix 2425159b3361Sopenharmony_ci left and right channels if the input file is stereo. (Alfred 2426159b3361Sopenharmony_ci Weyers) 2427159b3361Sopenharmony_ci </li> 2428159b3361Sopenharmony_ci <li> 2429159b3361Sopenharmony_ci <font color="#ff0000">New quant_compare algorithm (method for 2430159b3361Sopenharmony_ci deciding which of two quantizations is better) enabled with -X4 2431159b3361Sopenharmony_ci (Greg Maxwell)</font> 2432159b3361Sopenharmony_ci </li> 2433159b3361Sopenharmony_ci <li> 2434159b3361Sopenharmony_ci <font color="#ff0000">New mid/side VBR bit allocation formula. 2435159b3361Sopenharmony_ci Mid channel bits are set by the quality requirements, and then 2436159b3361Sopenharmony_ci the side channel uses a reduced number of bits (in a proportion 2437159b3361Sopenharmony_ci coming from the fixed bitrate code). This might not be optimal, 2438159b3361Sopenharmony_ci but it should be pretty good and no one knows what the optimal 2439159b3361Sopenharmony_ci solution should be. (Greg Maxwell)</font> 2440159b3361Sopenharmony_ci </li> 2441159b3361Sopenharmony_ci <li> 2442159b3361Sopenharmony_ci <font color="#ff0000">New VBR (#define RH) tunings based on 2443159b3361Sopenharmony_ci detailed listening tests by Macik and Greg Maxwell.</font> 2444159b3361Sopenharmony_ci </li> 2445159b3361Sopenharmony_ci <li> 2446159b3361Sopenharmony_ci Sigbjørn Skjæret fixed several 2447159b3361Sopenharmony_ci compiler warnings (which turned out to be potential bugs) 2448159b3361Sopenharmony_ci </li> 2449159b3361Sopenharmony_ci <li>Takehiro Tominaga fixed a low bitrate bug in reduce_side()</li> 2450159b3361Sopenharmony_ci <li>Alfred Weyers fixed some buffer overflows.</li> 2451159b3361Sopenharmony_ci <li> 2452159b3361Sopenharmony_ci <font color="#ff0000">New ATH (absolute threshold of hearing) 2453159b3361Sopenharmony_ci formula replaces buggy ISO code, and adds analog silence 2454159b3361Sopenharmony_ci treatment (removal of coefficients below below ATH). These are 2455159b3361Sopenharmony_ci turned on by default but have not been fully tested. (Robert 2456159b3361Sopenharmony_ci Hegemann)</font> 2457159b3361Sopenharmony_ci </li> 2458159b3361Sopenharmony_ci <li> 2459159b3361Sopenharmony_ci <font color="#ff0000">Bug in short block 2460159b3361Sopenharmony_ci spreading function fixed. (Robert Hegemann)</font> 2461159b3361Sopenharmony_ci </li> 2462159b3361Sopenharmony_ci </ul> 2463159b3361Sopenharmony_ci <h3>LAME 3.25beta August 22 1999</h3> 2464159b3361Sopenharmony_ci <ul> 2465159b3361Sopenharmony_ci <li> 2466159b3361Sopenharmony_ci Sigbjørn Skjæret fixed a zero byte malloc call. 2467159b3361Sopenharmony_ci This bug was introduced in 3.24 and causes problems on non 2468159b3361Sopenharmony_ci Linux systems. 2469159b3361Sopenharmony_ci </li> 2470159b3361Sopenharmony_ci <li> 2471159b3361Sopenharmony_ci Bit allocation routines would sometimes allocate more than 2472159b3361Sopenharmony_ci 4095 bits to one channel of one granule. A couple of people 2473159b3361Sopenharmony_ci reported problems that might be caused by this, especially at 2474159b3361Sopenharmony_ci higher bitrates. 2475159b3361Sopenharmony_ci </li> 2476159b3361Sopenharmony_ci <li> 2477159b3361Sopenharmony_ci Nils Faerber updated the man page and fixed many of the 2478159b3361Sopenharmony_ci compiler warnings. 2479159b3361Sopenharmony_ci </li> 2480159b3361Sopenharmony_ci </ul> 2481159b3361Sopenharmony_ci <h3>LAME 3.24beta August 15 1999</h3> 2482159b3361Sopenharmony_ci <ul> 2483159b3361Sopenharmony_ci <li> 2484159b3361Sopenharmony_ci This release contains the following new code (for 2485159b3361Sopenharmony_ci developers) which is disabled by default: 2486159b3361Sopenharmony_ci </li> 2487159b3361Sopenharmony_ci <li> 2488159b3361Sopenharmony_ci Hegemann: Completely overhauled VBR code. Now computes exact 2489159b3361Sopenharmony_ci number of bits required for the given qualty and then quantized 2490159b3361Sopenharmony_ci with the appropriate bitrate. 2491159b3361Sopenharmony_ci </li> 2492159b3361Sopenharmony_ci <li>Several new quantization quality measures.</li> 2493159b3361Sopenharmony_ci </ul> 2494159b3361Sopenharmony_ci <h3>LAME 3.23beta August 8 1999</h3> 2495159b3361Sopenharmony_ci <ul> 2496159b3361Sopenharmony_ci <li> 2497159b3361Sopenharmony_ci Very nice continuously updated VBR histogram display from 2498159b3361Sopenharmony_ci Iwasa Kazmi. (disabled with --nohist). 2499159b3361Sopenharmony_ci </li> 2500159b3361Sopenharmony_ci <li> 2501159b3361Sopenharmony_ci More modulerization work. The encoding engine can now be 2502159b3361Sopenharmony_ci compiled into libmp3lame, but the interface is awkward. 2503159b3361Sopenharmony_ci </li> 2504159b3361Sopenharmony_ci <li> 2505159b3361Sopenharmony_ci <font color="#ff0000">Bug fixed in FFT Hann window formula 2506159b3361Sopenharmony_ci (Leonid A. Kulakov).</font> 2507159b3361Sopenharmony_ci </li> 2508159b3361Sopenharmony_ci <li> 2509159b3361Sopenharmony_ci New LAME logo on the download page. Created by Chris Michalisles. 2510159b3361Sopenharmony_ci </li> 2511159b3361Sopenharmony_ci <li> 2512159b3361Sopenharmony_ci <font color="#ff0000">Several VBR algorithm 2513159b3361Sopenharmony_ci improvements from Robert Hegemann. New quantization noise 2514159b3361Sopenharmony_ci metrics and VBR quality measure takes into account mid/side 2515159b3361Sopenharmony_ci encoding. Should produce smaller files with the same quality, 2516159b3361Sopenharmony_ci especially when using jstereo.</font> 2517159b3361Sopenharmony_ci </li> 2518159b3361Sopenharmony_ci </ul> 2519159b3361Sopenharmony_ci <h3>LAME 3.22beta July 27 1999</h3> 2520159b3361Sopenharmony_ci <ul> 2521159b3361Sopenharmony_ci <li> 2522159b3361Sopenharmony_ci Downsampling (stereo to mono) bug with MPEG2 fixed. (Mike 2523159b3361Sopenharmony_ci Oliphant) 2524159b3361Sopenharmony_ci </li> 2525159b3361Sopenharmony_ci <li> 2526159b3361Sopenharmony_ci Downsampling now merges L & R channels - before it 2527159b3361Sopenharmony_ci only took the L channel. 2528159b3361Sopenharmony_ci </li> 2529159b3361Sopenharmony_ci <li> 2530159b3361Sopenharmony_ci More modularization and code cleanup from Albert Faber and 2531159b3361Sopenharmony_ci myself. 2532159b3361Sopenharmony_ci </li> 2533159b3361Sopenharmony_ci <li> 2534159b3361Sopenharmony_ci Input filesize limit removed for raw pcm input 2535159b3361Sopenharmony_ci files. For other file types, LAME will still only read the 2536159b3361Sopenharmony_ci first 2^32 samples, (27 hours of playing time at 44.1 kHz). 2537159b3361Sopenharmony_ci </li> 2538159b3361Sopenharmony_ci </ul> 2539159b3361Sopenharmony_ci <h3>LAME 3.21beta July 26 1999</h3> 2540159b3361Sopenharmony_ci <ul> 2541159b3361Sopenharmony_ci <li> 2542159b3361Sopenharmony_ci <font color="#ff0000">Correct Mid/Side masking thresholds for 2543159b3361Sopenharmony_ci JSTEREO mode! This is enabled with-h. It makes LAME about 20% 2544159b3361Sopenharmony_ci slower since it computes psycho-acoustics for L,R Mid and Side 2545159b3361Sopenharmony_ci channels.</font> 2546159b3361Sopenharmony_ci </li> 2547159b3361Sopenharmony_ci <li> 2548159b3361Sopenharmony_ci <font color="#ff0000">"Analog silence" 2549159b3361Sopenharmony_ci threshold added. Keeps VBR from upping the bitrate during 2550159b3361Sopenharmony_ci very quite passages. (Robert.Hegemann)</font> 2551159b3361Sopenharmony_ci </li> 2552159b3361Sopenharmony_ci <li> 2553159b3361Sopenharmony_ci <font color="#ff0000">New VBR quality setting from Robert 2554159b3361Sopenharmony_ci Hegemann. It is based on the idea that distortion at lower bit 2555159b3361Sopenharmony_ci rates sounds worse than at higher bitrates, and so the allowed 2556159b3361Sopenharmony_ci distortion (VBR quality setting) is proportional to the bitrate. 2557159b3361Sopenharmony_ci Because of this, default minimum bitrate is now 32kbs.</font> 2558159b3361Sopenharmony_ci </li> 2559159b3361Sopenharmony_ci <li> 2560159b3361Sopenharmony_ci <font color="#ff0000">Expermental subblock gain code enabled 2561159b3361Sopenharmony_ci with -Z.</font> 2562159b3361Sopenharmony_ci </li> 2563159b3361Sopenharmony_ci <li> 2564159b3361Sopenharmony_ci New "-r" option for raw pcm input files. With -r, LAME will not 2565159b3361Sopenharmony_ci do any fseek()'s or look for wav and aiff headers on 2566159b3361Sopenharmony_ci the input file. 2567159b3361Sopenharmony_ci </li> 2568159b3361Sopenharmony_ci <li> 2569159b3361Sopenharmony_ci Bug fixes in mp3x (frame analyzer) for viewing frames near 2570159b3361Sopenharmony_ci end of the file. 2571159b3361Sopenharmony_ci </li> 2572159b3361Sopenharmony_ci <li> 2573159b3361Sopenharmony_ci Bug fixed to allow setting the sampling rate of raw pcm 2574159b3361Sopenharmony_ci input files. 2575159b3361Sopenharmony_ci </li> 2576159b3361Sopenharmony_ci </ul> 2577159b3361Sopenharmony_ci <h3>LAME 3.20beta July 19 1999</h3> 2578159b3361Sopenharmony_ci <ul> 2579159b3361Sopenharmony_ci <li> 2580159b3361Sopenharmony_ci Bug in get_audio.c fixed. Libsndfile wrappers would not compile 2581159b3361Sopenharmony_ci (Miguel Revilla Rodriguez) 2582159b3361Sopenharmony_ci </li> 2583159b3361Sopenharmony_ci <li> 2584159b3361Sopenharmony_ci Nils Faerber found some unitialized variables and some 2585159b3361Sopenharmony_ci wierd extranous computations in filter_subband, now fixed. 2586159b3361Sopenharmony_ci This was causing seg faults on some machines. 2587159b3361Sopenharmony_ci </li> 2588159b3361Sopenharmony_ci </ul> 2589159b3361Sopenharmony_ci <h3> LAME 3.19beta July 18 1999</h3> 2590159b3361Sopenharmony_ci <ul> 2591159b3361Sopenharmony_ci <li> 2592159b3361Sopenharmony_ci <font color="#ff0000">Oops! Robert Hegemann immediatly found a 2593159b3361Sopenharmony_ci bug in the new (old -Z option) quantization code. calc_noise1 2594159b3361Sopenharmony_ci was not returning tot_noise, so non ms-stereo frames were 2595159b3361Sopenharmony_ci buggy.</font> 2596159b3361Sopenharmony_ci </li> 2597159b3361Sopenharmony_ci </ul> 2598159b3361Sopenharmony_ci <h3>LAME 3.18beta July 17 1999</h3> 2599159b3361Sopenharmony_ci <ul> 2600159b3361Sopenharmony_ci <li> 2601159b3361Sopenharmony_ci <font color="#ff0000">Many psycho-acoustic bug fixes. Dan 2602159b3361Sopenharmony_ci Nelson discovered a bug in MPEG2: For short blocks, the code 2603159b3361Sopenharmony_ci assumes 42 partition bands. MPEG1 sometimes has less, MPEG2 2604159b3361Sopenharmony_ci can have more. In MPEG1, this bug would not have effected the 2605159b3361Sopenharmony_ci output if your compiler initializes static variables to 0 on 2606159b3361Sopenharmony_ci creation. In MPEG2 it leads to array out-of-bounds access 2607159b3361Sopenharmony_ci errors. Finally, there was a related bug in MPEG1/MPEG2, short 2608159b3361Sopenharmony_ci & long blocks where the energy above 16 kHz was all added to 2609159b3361Sopenharmony_ci partition band 0. (the lowest frequeny partition band!)</font> 2610159b3361Sopenharmony_ci </li> 2611159b3361Sopenharmony_ci <li> 2612159b3361Sopenharmony_ci <font color="#ff0000">The -Z option (Gabriel Bouvigne's idea of 2613159b3361Sopenharmony_ci using total quantization noise to choose between two 2614159b3361Sopenharmony_ci quantizations with the same value of "over") is now the default. 2615159b3361Sopenharmony_ci I believe this helps remove the trilling sound in Jan's 2616159b3361Sopenharmony_ci testsignal4.wav. The quality of testsignal2.wav and 2617159b3361Sopenharmony_ci testsignal4.wav are now better than Xing and getting closer to FhG.</font> 2618159b3361Sopenharmony_ci </li> 2619159b3361Sopenharmony_ci <li> 2620159b3361Sopenharmony_ci Bug fixes in frame & sample count for downsampling 2621159b3361Sopenharmony_ci mode. (ben "jacobs") 2622159b3361Sopenharmony_ci </li> 2623159b3361Sopenharmony_ci <li>Patches to improve modulization. (ben "jacobs")</li> 2624159b3361Sopenharmony_ci </ul> 2625159b3361Sopenharmony_ci <h3>LAME 3.17beta July 11 1999</h3> 2626159b3361Sopenharmony_ci <ul> 2627159b3361Sopenharmony_ci <li> 2628159b3361Sopenharmony_ci substantial code cleanup towards goal of making LAME more 2629159b3361Sopenharmony_ci modular. 2630159b3361Sopenharmony_ci </li> 2631159b3361Sopenharmony_ci </ul> 2632159b3361Sopenharmony_ci <h3>LAME 3.16beta July 11 1999</h3> 2633159b3361Sopenharmony_ci <ul> 2634159b3361Sopenharmony_ci <li> 2635159b3361Sopenharmony_ci <font color="#ff0000">New tunings of window switching, and 2636159b3361Sopenharmony_ci better bit allocation based on pe. (Jan Rafaj. improves both 2637159b3361Sopenharmony_ci testsignal2.wav and testsignal4.wav).</font> 2638159b3361Sopenharmony_ci </li> 2639159b3361Sopenharmony_ci <li> 2640159b3361Sopenharmony_ci <font color="#ff0000">Bug in mid/side quantization when side 2641159b3361Sopenharmony_ci channel was zero fixed. (Albert Faber)</font> 2642159b3361Sopenharmony_ci </li> 2643159b3361Sopenharmony_ci <li> 2644159b3361Sopenharmony_ci Removed some extranous computations in l3psy.c (Robert 2645159b3361Sopenharmony_ci Hegemann) 2646159b3361Sopenharmony_ci </li> 2647159b3361Sopenharmony_ci <li> 2648159b3361Sopenharmony_ci More detailed timing status info, including hours display. 2649159b3361Sopenharmony_ci (Sakari Ailus) and percentage indicator (Conrad Sanderson). 2650159b3361Sopenharmony_ci </li> 2651159b3361Sopenharmony_ci <li> 2652159b3361Sopenharmony_ci <font color="#3366ff">Window_subband and calc_noise1, 2653159b3361Sopenharmony_ci calc_noise2 speedups. Quantize_xrpow speedup should be 2654159b3361Sopenharmony_ci significant on non GNU/intel systems. (Mike Cheng)</font> 2655159b3361Sopenharmony_ci </li> 2656159b3361Sopenharmony_ci <li> 2657159b3361Sopenharmony_ci <font color="#3366ff">Better initial guess for VBR bitrate. 2658159b3361Sopenharmony_ci Should speed up VBR encoding. (Gabriel Bouvigne)</font> 2659159b3361Sopenharmony_ci </li> 2660159b3361Sopenharmony_ci <li> 2661159b3361Sopenharmony_ci More advanced .wav header parsing. fixes bugs involving click 2662159b3361Sopenharmony_ci in first frame. (Robert.Hegemann) 2663159b3361Sopenharmony_ci </li> 2664159b3361Sopenharmony_ci <li> 2665159b3361Sopenharmony_ci Correct filesize and total frame computation when using 2666159b3361Sopenharmony_ci LIBSNDFILE (ben "jacobs") 2667159b3361Sopenharmony_ci </li> 2668159b3361Sopenharmony_ci <li> 2669159b3361Sopenharmony_ci Click in last frame (buffering problem) when using 2670159b3361Sopenharmony_ci libsndfile fixed. 2671159b3361Sopenharmony_ci </li> 2672159b3361Sopenharmony_ci <li> 2673159b3361Sopenharmony_ci Audio I/O code overhauled. There is now a uniform audio i/o 2674159b3361Sopenharmony_ci interface to libsndfile or the LAME built in wav/aiff 2675159b3361Sopenharmony_ci routines. All audio i/o code localized to get_audio.c. 2676159b3361Sopenharmony_ci </li> 2677159b3361Sopenharmony_ci </ul> 2678159b3361Sopenharmony_ci <h3>LAME 3.15beta</h3> 2679159b3361Sopenharmony_ci <ul> 2680159b3361Sopenharmony_ci <li> 2681159b3361Sopenharmony_ci times()/clock() problem fixed for non-unix OS. (Ben "Jacobs") 2682159b3361Sopenharmony_ci </li> 2683159b3361Sopenharmony_ci <li> 2684159b3361Sopenharmony_ci Fixed uninitialized pe[] when using fast mode. (Ben "Jacobs") 2685159b3361Sopenharmony_ci </li> 2686159b3361Sopenharmony_ci </ul> 2687159b3361Sopenharmony_ci <h3>LAME 3.13 June 24 1999</h3> 2688159b3361Sopenharmony_ci <ul> 2689159b3361Sopenharmony_ci <li>Patches for BeOS from Gertjan van Ratingen.</li> 2690159b3361Sopenharmony_ci <li>Makefile info for OS/2 Warp 4.0 (from dink.org).</li> 2691159b3361Sopenharmony_ci <li>Status display now based on wall clock time, not cpu time.</li> 2692159b3361Sopenharmony_ci <li> 2693159b3361Sopenharmony_ci mem_alloc no longer allocates twice as much memory as 2694159b3361Sopenharmony_ci needed (Jan Peman). 2695159b3361Sopenharmony_ci </li> 2696159b3361Sopenharmony_ci </ul> 2697159b3361Sopenharmony_ci <h3>LAME 3.12pre9</h3> 2698159b3361Sopenharmony_ci <ul> 2699159b3361Sopenharmony_ci <li> 2700159b3361Sopenharmony_ci Updated BLADEDLL code to handle recent changes (Albert 2701159b3361Sopenharmony_ci Faber). 2702159b3361Sopenharmony_ci </li> 2703159b3361Sopenharmony_ci <li> 2704159b3361Sopenharmony_ci Bug fixed in parsing options when not using GTK (Albert 2705159b3361Sopenharmony_ci Faber). 2706159b3361Sopenharmony_ci </li> 2707159b3361Sopenharmony_ci <li> 2708159b3361Sopenharmony_ci <font color="#ff0000">MPEG2 Layer III psycho acoustics now 2709159b3361Sopenharmony_ci working.</font> 2710159b3361Sopenharmony_ci </li> 2711159b3361Sopenharmony_ci <li> 2712159b3361Sopenharmony_ci <font color="#3366ff">Improved huffman encoding Chris 2713159b3361Sopenharmony_ci Matrakidis. (10% faster). I dont know how he finds these 2714159b3361Sopenharmony_ci improvements! LAME with full quality now encodes 2715159b3361Sopenharmony_ci faster than real time on my PII 266.</font> 2716159b3361Sopenharmony_ci </li> 2717159b3361Sopenharmony_ci <li> 2718159b3361Sopenharmony_ci Fixed time display when encoding takes more than 60 2719159b3361Sopenharmony_ci minutes. 2720159b3361Sopenharmony_ci </li> 2721159b3361Sopenharmony_ci </ul> 2722159b3361Sopenharmony_ci <h3>LAME 3.12pre8</h3> 2723159b3361Sopenharmony_ci <ul> 2724159b3361Sopenharmony_ci <li> 2725159b3361Sopenharmony_ci <font color="#ff0000">New <a href="gpsycho/ms_stereo.html"> 2726159b3361Sopenharmony_ci mid/side stereo</a> criterion. LAME will use mid/side stereo 2727159b3361Sopenharmony_ci only when the difference between L & R masking thresholds 2728159b3361Sopenharmony_ci (averaged over all scalefactors) is less then 5db. In several 2729159b3361Sopenharmony_ci test samples it does a very good job mimicking the FhG 2730159b3361Sopenharmony_ci encoder.</font> 2731159b3361Sopenharmony_ci </li> 2732159b3361Sopenharmony_ci <li> 2733159b3361Sopenharmony_ci <font color="#ff0000">Bug in mid/side stereo fixed: independent 2734159b3361Sopenharmony_ci variation of mid & side channel scalefactors disabled. 2735159b3361Sopenharmony_ci Because of the way outer_loop is currently coded, when encoding 2736159b3361Sopenharmony_ci mid/side coefficietns using left/right thresholds, you have 2737159b3361Sopenharmony_ci to vary the scalefactors simultaneously.</font> 2738159b3361Sopenharmony_ci </li> 2739159b3361Sopenharmony_ci <li> 2740159b3361Sopenharmony_ci <font color="#ff0000">Bug in side/mid energy 2741159b3361Sopenharmony_ci ratio calculation fixed. (Thanks to Robert Hegemann)</font> 2742159b3361Sopenharmony_ci </li> 2743159b3361Sopenharmony_ci <li> 2744159b3361Sopenharmony_ci Default mode is stereo (not jstereo) if bitrate is chosen 2745159b3361Sopenharmony_ci as 192kbs or higher. Tero Auvinen first pointed out that FhG 2746159b3361Sopenharmony_ci seems to think at 160kbs, their encoder is so good it doesn't 2747159b3361Sopenharmony_ci need jstereo tricks. Since LAME is not as good as FhG, I am 2748159b3361Sopenharmony_ci going to claim that 192kbs LAME is so good it doens't need 2749159b3361Sopenharmony_ci jstereo tricks, and thus it is disabled by default. 2750159b3361Sopenharmony_ci </li> 2751159b3361Sopenharmony_ci <li> 2752159b3361Sopenharmony_ci WAV header parsing for big-endian machines, and automatic 2753159b3361Sopenharmony_ci detection of big-endian machines. (Thanks to Sigbjørn 2754159b3361Sopenharmony_ci Skjæret). 2755159b3361Sopenharmony_ci </li> 2756159b3361Sopenharmony_ci <li>added 56 sample delay to sync LAME with FhG.</li> 2757159b3361Sopenharmony_ci <li>MP3x (frame analyzer) can now handle MPEG2 streams.</li> 2758159b3361Sopenharmony_ci </ul> 2759159b3361Sopenharmony_ci <h3>LAME 3.12pre7</h3> 2760159b3361Sopenharmony_ci <ul> 2761159b3361Sopenharmony_ci <li> 2762159b3361Sopenharmony_ci MPEG2 layer III now working! lower bit rates (down to 8kbs) 2763159b3361Sopenharmony_ci and 3 more sampling frequencies: 16000, 22050, 24000Hz. Quality 2764159b3361Sopenharmony_ci is poor - the psy-model does not yet work with these 2765159b3361Sopenharmony_ci sampling frequencies. 2766159b3361Sopenharmony_ci </li> 2767159b3361Sopenharmony_ci <li> 2768159b3361Sopenharmony_ci Fixed "ERROR: outer_loop(): huff_bits < 0." bug 2769159b3361Sopenharmony_ci when using VBR. 2770159b3361Sopenharmony_ci </li> 2771159b3361Sopenharmony_ci <li> 2772159b3361Sopenharmony_ci bash and sh scripts to run LAME on multiple files now 2773159b3361Sopenharmony_ci included. (from Robert Hegemann and Gerhard Wesp respectively) 2774159b3361Sopenharmony_ci </li> 2775159b3361Sopenharmony_ci <li> 2776159b3361Sopenharmony_ci bug fix in encoding times for longer files from 2777159b3361Sopenharmony_ci (Alvaro Martinez Echevarria) 2778159b3361Sopenharmony_ci </li> 2779159b3361Sopenharmony_ci <li>yet another segfault in the frame analyzer fixed.</li> 2780159b3361Sopenharmony_ci <li> 2781159b3361Sopenharmony_ci ISO psy-model/bit allocation routines removed. This allowed 2782159b3361Sopenharmony_ci makeframe() to be made much simpler, and most of the complicated 2783159b3361Sopenharmony_ci buffering is now gone. Eventually I would like the encoding 2784159b3361Sopenharmony_ci engine to be a stand alone library. 2785159b3361Sopenharmony_ci </li> 2786159b3361Sopenharmony_ci </ul> 2787159b3361Sopenharmony_ci <h3>LAME 3.12pre6</h3> 2788159b3361Sopenharmony_ci <ul> 2789159b3361Sopenharmony_ci <li> 2790159b3361Sopenharmony_ci <font color="#ff0000">Better VBR 2791159b3361Sopenharmony_ci tuning. Find minimum bitrate with distortion less than the 2792159b3361Sopenharmony_ci allows maximum. A minimum bit rate is imposed on frames with 2793159b3361Sopenharmony_ci short blocks (where the measured distortion can not be 2794159b3361Sopenharmony_ci trusted). A minimum frame bitrate can be specified 2795159b3361Sopenharmony_ci with -b, default=64kbs.</font> 2796159b3361Sopenharmony_ci </li> 2797159b3361Sopenharmony_ci <li> 2798159b3361Sopenharmony_ci <a href="http://www.zip.com.au/%7Eerikd/libsndfile">LIBSNDFILE</a> 2799159b3361Sopenharmony_ci support. With libsndfile, LAME can encode almost all sound 2800159b3361Sopenharmony_ci formats. Albert Faber did the work for this, including 2801159b3361Sopenharmony_ci getting libsndfile running under win32. 2802159b3361Sopenharmony_ci </li> 2803159b3361Sopenharmony_ci <li>CRC checksum now working! (Thanks to Johannes Overmann)</li> 2804159b3361Sopenharmony_ci <li>frame analyzer will now work with mono .mp3 files</li> 2805159b3361Sopenharmony_ci <li><font color="#3366ff">more code tweeks from Jan Peman.</font></li> 2806159b3361Sopenharmony_ci <li> 2807159b3361Sopenharmony_ci <font color="#3366ff">Compaq-Alpha(Linux) fixes and speedups 2808159b3361Sopenharmony_ci from Nils Faerber.</font> 2809159b3361Sopenharmony_ci </li> 2810159b3361Sopenharmony_ci <li> 2811159b3361Sopenharmony_ci <font color="#3366ff">Faster bin_search_StepSize from Juha 2812159b3361Sopenharmony_ci Laukala.</font> 2813159b3361Sopenharmony_ci </li> 2814159b3361Sopenharmony_ci <li> 2815159b3361Sopenharmony_ci <font color="#3366ff">Faster quantize() from Mike Cheng</font> 2816159b3361Sopenharmony_ci </li> 2817159b3361Sopenharmony_ci <li> 2818159b3361Sopenharmony_ci <font color="#3366ff">Faster 2819159b3361Sopenharmony_ci quantize_xrpow() from Chris Matrakidis. xrpow_flag removed 2820159b3361Sopenharmony_ci since this option is now on by default.</font> 2821159b3361Sopenharmony_ci </li> 2822159b3361Sopenharmony_ci <li>Fixed .wav header parsing from Nils Faerber.</li> 2823159b3361Sopenharmony_ci <li> 2824159b3361Sopenharmony_ci Xing VBR frame info header code from Albert 2825159b3361Sopenharmony_ci Faber. "Xing" and "LAME 3.12" embedded in first 2826159b3361Sopenharmony_ci frame. 2827159b3361Sopenharmony_ci </li> 2828159b3361Sopenharmony_ci <li> 2829159b3361Sopenharmony_ci <font color="#ff0000">Bug in VBR bit allocation based on "over" 2830159b3361Sopenharmony_ci value fixed.</font> 2831159b3361Sopenharmony_ci </li> 2832159b3361Sopenharmony_ci </ul> 2833159b3361Sopenharmony_ci <h3>LAME 3.11 June 3 1999</h3> 2834159b3361Sopenharmony_ci <ul> 2835159b3361Sopenharmony_ci <li>Almost all warnings (-Wall) now fixed! (Thanks to Jan Peman)</li> 2836159b3361Sopenharmony_ci <li> 2837159b3361Sopenharmony_ci More coding improvements from Gabriel Bouvigne and Warren 2838159b3361Sopenharmony_ci Toomey. 2839159b3361Sopenharmony_ci </li> 2840159b3361Sopenharmony_ci <li> 2841159b3361Sopenharmony_ci <font color="#ff0000">VBR (variable bit rate). Increases bit 2842159b3361Sopenharmony_ci rate for short blocks and for frames where the number of bands 2843159b3361Sopenharmony_ci containing audible distortion is greater than a given value. 2844159b3361Sopenharmony_ci Much tuning needs to be done.</font> 2845159b3361Sopenharmony_ci </li> 2846159b3361Sopenharmony_ci <li>Patch to remove all atan() calls from James Droppo.</li> 2847159b3361Sopenharmony_ci </ul> 2848159b3361Sopenharmony_ci <h3>LAME 3.10 May 30 1999</h3> 2849159b3361Sopenharmony_ci <ul> 2850159b3361Sopenharmony_ci <li> 2851159b3361Sopenharmony_ci <font color="#3366ff">Fast mode 2852159b3361Sopenharmony_ci (-f) disables psycho-acoustic model for real time encoding on 2853159b3361Sopenharmony_ci older machines. Thanks to Lauri Ahonen who first sent a patch 2854159b3361Sopenharmony_ci for this.</font> 2855159b3361Sopenharmony_ci </li> 2856159b3361Sopenharmony_ci <li> 2857159b3361Sopenharmony_ci <font color="#ff0000">New bit reservoir usage scheme to 2858159b3361Sopenharmony_ci accommodate the new pre-echo detection formulas.</font> 2859159b3361Sopenharmony_ci </li> 2860159b3361Sopenharmony_ci <li> 2861159b3361Sopenharmony_ci <font color="#ff0000">Tuning of AWS and ENER_AWS pre-echo 2862159b3361Sopenharmony_ci formulas by Gabriel Bouvigne and myself. They work great! now 2863159b3361Sopenharmony_ci on by default.</font> 2864159b3361Sopenharmony_ci </li> 2865159b3361Sopenharmony_ci <li> 2866159b3361Sopenharmony_ci In jstereo, force blocktypes for left & right 2867159b3361Sopenharmony_ci channels to be identical. FhG seems to do this. It 2868159b3361Sopenharmony_ci can be disabled with "-d". 2869159b3361Sopenharmony_ci </li> 2870159b3361Sopenharmony_ci <li>Patches to compile MP3x under win32 (Thanks to Albert Faber).</li> 2871159b3361Sopenharmony_ci <li> 2872159b3361Sopenharmony_ci <font color="#3366ff">bin_serach_stepsize 2873159b3361Sopenharmony_ci limited to a quantizationStepSize of -210 through 45.</font> 2874159b3361Sopenharmony_ci </li> 2875159b3361Sopenharmony_ci <li> 2876159b3361Sopenharmony_ci <font color="#ff0000">outer_loop() will now vary Mid & Side 2877159b3361Sopenharmony_ci scalefactors independently. Can lead to better quantizations, 2878159b3361Sopenharmony_ci but it is slower (twice as many quantizations to look at). 2879159b3361Sopenharmony_ci Running with "-m f" does not need this and will run at the 2880159b3361Sopenharmony_ci old speed</font> 2881159b3361Sopenharmony_ci </li> 2882159b3361Sopenharmony_ci <li> 2883159b3361Sopenharmony_ci <font color="#ff0000">Bug in inner_loop 2884159b3361Sopenharmony_ci would allow quantizations larger than allowed. (introduced in 2885159b3361Sopenharmony_ci lame3.04, now fixed.)</font> 2886159b3361Sopenharmony_ci </li> 2887159b3361Sopenharmony_ci <li>Updated HTML documentation from Gabriel Bouvigne.</li> 2888159b3361Sopenharmony_ci <li>Unix man page from William Schelter.</li> 2889159b3361Sopenharmony_ci <li> 2890159b3361Sopenharmony_ci <font color="#ff0000">numlines[] bug fixed. (Thanks to Rafael 2891159b3361Sopenharmony_ci Luebbert, MPecker author).</font> 2892159b3361Sopenharmony_ci </li> 2893159b3361Sopenharmony_ci <li> 2894159b3361Sopenharmony_ci <font color="#3366ff">Quantization speed improvements from Chris 2895159b3361Sopenharmony_ci Matrakidis.</font> 2896159b3361Sopenharmony_ci </li> 2897159b3361Sopenharmony_ci <li> 2898159b3361Sopenharmony_ci <font color="#ff0000">When comparing quantizations with the same 2899159b3361Sopenharmony_ci number of bands with audible distortion, use the one with the 2900159b3361Sopenharmony_ci largest scalefactors, not the first one outer_loop 2901159b3361Sopenharmony_ci happened to find.</font> 2902159b3361Sopenharmony_ci </li> 2903159b3361Sopenharmony_ci <li> 2904159b3361Sopenharmony_ci Improved defination of best quantization when using -f 2905159b3361Sopenharmony_ci (fast mode). 2906159b3361Sopenharmony_ci </li> 2907159b3361Sopenharmony_ci <li> 2908159b3361Sopenharmony_ci subblock code now working. But no algorithm to 2909159b3361Sopenharmony_ci choose subblock gains yet. 2910159b3361Sopenharmony_ci </li> 2911159b3361Sopenharmony_ci <li> 2912159b3361Sopenharmony_ci Linux now segfaults on floating point exceptions. Should 2913159b3361Sopenharmony_ci prevent me from releasing binaries that crash on other 2914159b3361Sopenharmony_ci operating systems. 2915159b3361Sopenharmony_ci </li> 2916159b3361Sopenharmony_ci </ul> 2917159b3361Sopenharmony_ci <h3>LAME 3.04 May 22 1999</h3> 2918159b3361Sopenharmony_ci <ul> 2919159b3361Sopenharmony_ci <li>Preliminary documentation from Gabriel Bouvigne.</li> 2920159b3361Sopenharmony_ci <li> 2921159b3361Sopenharmony_ci <font color="#3366ff">I wouldn't have thought it was possible, 2922159b3361Sopenharmony_ci but now there are even more speed improvements from Chris 2923159b3361Sopenharmony_ci Matrakidis! Removed one FFT when using joint stereo, and many 2924159b3361Sopenharmony_ci improvements in loop.c.</font> 2925159b3361Sopenharmony_ci </li> 2926159b3361Sopenharmony_ci <li> 2927159b3361Sopenharmony_ci "Fake" ms_stereo mode renamed "Force" ms_stereo since it forces 2928159b3361Sopenharmony_ci mid/side stereo on all frames. For some music this is said to be 2929159b3361Sopenharmony_ci a problem, but for most music mode is probably better than the 2930159b3361Sopenharmony_ci default jstereo because it uses specialized mid/side channel 2931159b3361Sopenharmony_ci masking thresholds. 2932159b3361Sopenharmony_ci </li> 2933159b3361Sopenharmony_ci <li>Small bugs in Force ms_stereo mode fixed.</li> 2934159b3361Sopenharmony_ci <li>Compaq Alpha fixes from Nathan Slingerland.</li> 2935159b3361Sopenharmony_ci <li> 2936159b3361Sopenharmony_ci <font color="#ff0000">Some new experimental pre-echo detection 2937159b3361Sopenharmony_ci formulas in l3psy.c (#ifdef AWS and #ifdef ENER_AWS, 2938159b3361Sopenharmony_ci both off by default. Thanks to Gabriel Bouvigne and Andre 2939159b3361Sopenharmony_ci Osterhues)</font> 2940159b3361Sopenharmony_ci </li> 2941159b3361Sopenharmony_ci <li> 2942159b3361Sopenharmony_ci Several bugs in the syncing of data displayed by mp3x (the 2943159b3361Sopenharmony_ci frame analyzer) were fixed. 2944159b3361Sopenharmony_ci </li> 2945159b3361Sopenharmony_ci <li> 2946159b3361Sopenharmony_ci highq (-h) option added. This turns on things 2947159b3361Sopenharmony_ci (just one so far) that should sound better but slow down LAME. 2948159b3361Sopenharmony_ci </li> 2949159b3361Sopenharmony_ci </ul> 2950159b3361Sopenharmony_ci <h3>LAME 3.03 May 18 1999 </h3> 2951159b3361Sopenharmony_ci <ul> 2952159b3361Sopenharmony_ci <li> 2953159b3361Sopenharmony_ci <font color="#3366ff">Faster (20%) & 2954159b3361Sopenharmony_ci cleaner FFT (Thanks to Chris Matrakidis 2955159b3361Sopenharmony_ci http://www.geocities.com/ResearchTriangle/8869/fft_summary.html)</font> 2956159b3361Sopenharmony_ci </li> 2957159b3361Sopenharmony_ci <li> 2958159b3361Sopenharmony_ci mods so it works with VC++ (Thanks to Gabriel Bouvigne, 2959159b3361Sopenharmony_ci www.mp3tech.org) 2960159b3361Sopenharmony_ci </li> 2961159b3361Sopenharmony_ci <li> 2962159b3361Sopenharmony_ci MP3s marked "original" by default (Thanks to 2963159b3361Sopenharmony_ci Gabriel Bouvigne, www.mp3tech.org) 2964159b3361Sopenharmony_ci </li> 2965159b3361Sopenharmony_ci <li> 2966159b3361Sopenharmony_ci Can now be compiled into a BladeEnc compatible .DLL (Thanks to 2967159b3361Sopenharmony_ci Albert Faber, CDex author) 2968159b3361Sopenharmony_ci </li> 2969159b3361Sopenharmony_ci <li> 2970159b3361Sopenharmony_ci Patches for "silent mode" and stdin/stdout (Thanks to Lars Magne 2971159b3361Sopenharmony_ci Ingebrigtsen) 2972159b3361Sopenharmony_ci </li> 2973159b3361Sopenharmony_ci <li> 2974159b3361Sopenharmony_ci <font color="#ff0000">Fixed rare bug: if a long_block is 2975159b3361Sopenharmony_ci sandwiched between two short_blocks, it must be changed to a 2976159b3361Sopenharmony_ci short_block, but the short_block ratios have not been computed in 2977159b3361Sopenharmony_ci l3psy.c. Now always compute short_block ratios just in case.</font> 2978159b3361Sopenharmony_ci </li> 2979159b3361Sopenharmony_ci <li> 2980159b3361Sopenharmony_ci <font color="#ff0000">Fixed bug with initial quantize step size 2981159b3361Sopenharmony_ci when many coefficients are zero. (Thanks to Martin Weghofer).</font> 2982159b3361Sopenharmony_ci </li> 2983159b3361Sopenharmony_ci <li>Bug fixed in MP3x display of audible distortion.</li> 2984159b3361Sopenharmony_ci <li>improved status display (Thanks to Lauri Ahonen).</li> 2985159b3361Sopenharmony_ci </ul> 2986159b3361Sopenharmony_ci <h3>LAME 3.02 May 12 1999</h3> 2987159b3361Sopenharmony_ci <ul> 2988159b3361Sopenharmony_ci <li> 2989159b3361Sopenharmony_ci <font color="#ff0000">encoder could use ms_stereo even if 2990159b3361Sopenharmony_ci channel 0 and 1 block types were different. (Thanks to Jan 2991159b3361Sopenharmony_ci Rafaj)</font> 2992159b3361Sopenharmony_ci </li> 2993159b3361Sopenharmony_ci <li> 2994159b3361Sopenharmony_ci <font color="#ff0000">added -k option to 2995159b3361Sopenharmony_ci disable the 16 kHz cutoff at 128kbs. This cutoff is never 2996159b3361Sopenharmony_ci used at higher bitrates. (Thanks to Jan Rafaj)</font> 2997159b3361Sopenharmony_ci </li> 2998159b3361Sopenharmony_ci <li> 2999159b3361Sopenharmony_ci <font color="#ff0000">modified pe bit allocation formula to 3000159b3361Sopenharmony_ci make sense at bit rates other than 128kbs.</font> 3001159b3361Sopenharmony_ci </li> 3002159b3361Sopenharmony_ci <li> 3003159b3361Sopenharmony_ci fixed l3_xmin initialization problem which showed up under 3004159b3361Sopenharmony_ci FreeBSD. (Thanks to Warren Toomey) 3005159b3361Sopenharmony_ci </li> 3006159b3361Sopenharmony_ci </ul> 3007159b3361Sopenharmony_ci <h3><b>LAME 3.01 May 11 1999</b> </h3> 3008159b3361Sopenharmony_ci <ul> 3009159b3361Sopenharmony_ci <li>max_name_size increased to 300 (Thanks to Mike Oliphant)</li> 3010159b3361Sopenharmony_ci <li>patch to allow seeks on input file (Thanks to Scott Manley)</li> 3011159b3361Sopenharmony_ci <li> 3012159b3361Sopenharmony_ci fixes for mono modes (Thanks to everyone who pointed this out) 3013159b3361Sopenharmony_ci </li> 3014159b3361Sopenharmony_ci <li>overflow in calc_noise2 fixed</li> 3015159b3361Sopenharmony_ci <li> 3016159b3361Sopenharmony_ci it reservoir overflow when encoding lots of frames with 3017159b3361Sopenharmony_ci all zeros (Thanks to Jani Frilander) 3018159b3361Sopenharmony_ci </li> 3019159b3361Sopenharmony_ci </ul> 3020159b3361Sopenharmony_ci <hr> 3021159b3361Sopenharmony_ci <h3>LAME 3.0 May 10 1999</h3> 3022159b3361Sopenharmony_ci <ul> 3023159b3361Sopenharmony_ci <li> 3024159b3361Sopenharmony_ci <font color="#ff0000">added GPSYCHO (developed by Mark Taylor) 3025159b3361Sopenharmony_ci </font> 3026159b3361Sopenharmony_ci </li> 3027159b3361Sopenharmony_ci <li> 3028159b3361Sopenharmony_ci <font color="#000000">added MP3x (developed by Mark Taylor)</font> 3029159b3361Sopenharmony_ci </li> 3030159b3361Sopenharmony_ci <li>LAME now maintained by Mark Taylor</li> 3031159b3361Sopenharmony_ci </ul> 3032159b3361Sopenharmony_ci <h3>November 8 1998</h3> 3033159b3361Sopenharmony_ci <ul> 3034159b3361Sopenharmony_ci <li>Version 2.1f released</li> 3035159b3361Sopenharmony_ci <li> 3036159b3361Sopenharmony_ci 50% faster filter_subband() routine in encode.c 3037159b3361Sopenharmony_ci contributed by James Droppo 3038159b3361Sopenharmony_ci </li> 3039159b3361Sopenharmony_ci </ul> 3040159b3361Sopenharmony_ci <h3>November 2 1998</h3> 3041159b3361Sopenharmony_ci <ul> 3042159b3361Sopenharmony_ci <li>Version 2.1e released.</li> 3043159b3361Sopenharmony_ci <li> 3044159b3361Sopenharmony_ci New command line switch <b>-a</b> auto-resamples a stereo input 3045159b3361Sopenharmony_ci file to mono. 3046159b3361Sopenharmony_ci </li> 3047159b3361Sopenharmony_ci <li> 3048159b3361Sopenharmony_ci New command line switch <b>-r</b> resamples from 44.1 kHz to 32 3049159b3361Sopenharmony_ci kHz [this switch doesn't work really well. Very tinny sounding 3050159b3361Sopenharmony_ci output files. Has to do with the way I do the resampling 3051159b3361Sopenharmony_ci probably] 3052159b3361Sopenharmony_ci </li> 3053159b3361Sopenharmony_ci <li> 3054159b3361Sopenharmony_ci Both of these were put into the ISO code in the encode.c file, 3055159b3361Sopenharmony_ci and are simply different ways of filling the input buffers from 3056159b3361Sopenharmony_ci a file. 3057159b3361Sopenharmony_ci </li> 3058159b3361Sopenharmony_ci </ul> 3059159b3361Sopenharmony_ci <h3>October 31 1998</h3> 3060159b3361Sopenharmony_ci <ul> 3061159b3361Sopenharmony_ci <li>Version 2.1d released</li> 3062159b3361Sopenharmony_ci <li>Fixed memory alloc in musicin.c (for l3_sb_sample)</li> 3063159b3361Sopenharmony_ci <li> 3064159b3361Sopenharmony_ci Added new command line switch (-x) to force swapping of 3065159b3361Sopenharmony_ci byte order 3066159b3361Sopenharmony_ci </li> 3067159b3361Sopenharmony_ci <li> 3068159b3361Sopenharmony_ci Cleaned up memory routines in l3psy.c. All the mem_alloc() and 3069159b3361Sopenharmony_ci free() routines where changed so that it was only done <i>once</i> 3070159b3361Sopenharmony_ci and not every single time the routine was called. 3071159b3361Sopenharmony_ci </li> 3072159b3361Sopenharmony_ci <li> 3073159b3361Sopenharmony_ci Added a compile time switch -DTIMER that includes all timing 3074159b3361Sopenharmony_ci info. It's a switch for the time being until some other people 3075159b3361Sopenharmony_ci have tested on their system. Timing code has a tendency to do 3076159b3361Sopenharmony_ci different things on different platforms. 3077159b3361Sopenharmony_ci </li> 3078159b3361Sopenharmony_ci </ul> 3079159b3361Sopenharmony_ci <h3>October 18 1998</h3> 3080159b3361Sopenharmony_ci <ul> 3081159b3361Sopenharmony_ci <li>Version 2.1b released.</li> 3082159b3361Sopenharmony_ci <li>Fixed up bug: all PCM files were being read as WAV.</li> 3083159b3361Sopenharmony_ci <li> 3084159b3361Sopenharmony_ci Played with the mem_alloc routine to fix crash under amigaos 3085159b3361Sopenharmony_ci (just allocating twice as much memory as needed). Might see if 3086159b3361Sopenharmony_ci we can totally do without this routine. Individual malloc()s 3087159b3361Sopenharmony_ci where they are needed instead 3088159b3361Sopenharmony_ci </li> 3089159b3361Sopenharmony_ci <li> 3090159b3361Sopenharmony_ci Put Jan Peman's quality switch back in. This reduces quality 3091159b3361Sopenharmony_ci via the '-q <int>' switch. Fun speedup which is 3092159b3361Sopenharmony_ci mostly harmless if you're not concerned with quality. 3093159b3361Sopenharmony_ci </li> 3094159b3361Sopenharmony_ci <li>Compiling with amiga-gcc works fine</li> 3095159b3361Sopenharmony_ci </ul> 3096159b3361Sopenharmony_ci <h3>October 16 1998</h3> 3097159b3361Sopenharmony_ci <ul> 3098159b3361Sopenharmony_ci <li> 3099159b3361Sopenharmony_ci Version 2.1a released. User input/output has been cleaned up a 3100159b3361Sopenharmony_ci bit. WAV file reading is there in a very rudimentary sense ie 3101159b3361Sopenharmony_ci the program will recognize the header and skip it, but not read 3102159b3361Sopenharmony_ci it. The WAV file is assumed to be 16bit stereo 44.1 kHz. 3103159b3361Sopenharmony_ci </li> 3104159b3361Sopenharmony_ci </ul> 3105159b3361Sopenharmony_ci <h3>October 6 1998</h3> 3106159b3361Sopenharmony_ci <ul> 3107159b3361Sopenharmony_ci <li> 3108159b3361Sopenharmony_ci Version 2.1 released with all tables now incorporated into 3109159b3361Sopenharmony_ci the exe. Thanks to <b>Lars Magne Ingebrigtseni</b> 3110159b3361Sopenharmony_ci </li> 3111159b3361Sopenharmony_ci </ul> 3112159b3361Sopenharmony_ci <h3>October 4 1998</h3> 3113159b3361Sopenharmony_ci <ul> 3114159b3361Sopenharmony_ci <li> 3115159b3361Sopenharmony_ci In response to some concerns about the quality of the encoder, I 3116159b3361Sopenharmony_ci have rebuilt the encoder from scratch and carefully compared output 3117159b3361Sopenharmony_ci at all stages with the output of the unmodified ISO encoder. 3118159b3361Sopenharmony_ci </li> 3119159b3361Sopenharmony_ci <li> 3120159b3361Sopenharmony_ci <a href="http://www.uq.net.au/%7Ezzmcheng/lame/download.html"> 3121159b3361Sopenharmony_ci Version 2.0</a> of LAME is built from the ISO source code (dist10), 3122159b3361Sopenharmony_ci and incorporates modifications from myself and the 8hz effort. The 3123159b3361Sopenharmony_ci output file from LAME v2.0 is <em>identical</em> to the output of 3124159b3361Sopenharmony_ci the ISO encoder for my test file. Since I do not have heaps of time, 3125159b3361Sopenharmony_ci I left the ISO AIFF file reader in the code, and did not incorporate 3126159b3361Sopenharmony_ci a WAV file reader. 3127159b3361Sopenharmony_ci </li> 3128159b3361Sopenharmony_ci <li> 3129159b3361Sopenharmony_ci Added section 3130159b3361Sopenharmony_ci on <a href="http://www.uq.net.au/%7Ezzmcheng/lame/quality.html"> 3131159b3361Sopenharmony_ci quality</a>. 3132159b3361Sopenharmony_ci </li> 3133159b3361Sopenharmony_ci </ul> 3134159b3361Sopenharmony_ci <h3> October 1 1998</h3> 3135159b3361Sopenharmony_ci <ul> 3136159b3361Sopenharmony_ci <li>Updated web page and released LAME v1.0</li> 3137159b3361Sopenharmony_ci </ul> 3138159b3361Sopenharmony_ci <hr> 3139159b3361Sopenharmony_ci <h3>Up to September 1998</h3> 3140159b3361Sopenharmony_ci <p>Working on the 8hz source code...</p> 3141159b3361Sopenharmony_ci <ul> 3142159b3361Sopenharmony_ci <li> 3143159b3361Sopenharmony_ci Patched the <a href="http://www.8hz.com/">8hz</a> source 3144159b3361Sopenharmony_ci code 3145159b3361Sopenharmony_ci </li> 3146159b3361Sopenharmony_ci <li> 3147159b3361Sopenharmony_ci 45% faster than original source (on my freebsd p166). 3148159b3361Sopenharmony_ci <ul> 3149159b3361Sopenharmony_ci <li> 3150159b3361Sopenharmony_ci m1 - sped up the mdct.c and quantize() functions [MDCTD, 3151159b3361Sopenharmony_ci MDCTD2, LOOPD] 3152159b3361Sopenharmony_ci </li> 3153159b3361Sopenharmony_ci <li> 3154159b3361Sopenharmony_ci m2 - sped up the filter_subband routine using <b>Stephane 3155159b3361Sopenharmony_ci Tavenard</b>'s work from musicin [FILTST] 3156159b3361Sopenharmony_ci </li> 3157159b3361Sopenharmony_ci <li>m2 - minor cleanup of window_subband [WINDST2]</li> 3158159b3361Sopenharmony_ci <li> 3159159b3361Sopenharmony_ci m2 - Cleaned up a few bits in l3psy.c. Replaced a sparse 3160159b3361Sopenharmony_ci matrix multiply with a hand configured unrolling [PSYD] 3161159b3361Sopenharmony_ci </li> 3162159b3361Sopenharmony_ci <li> 3163159b3361Sopenharmony_ci m3 - (amiga only) Added in the asm FFT for m68k (based on 3164159b3361Sopenharmony_ci sources from <b>Henryk Richter</b> and <b>Stephane Tavenard</b>) 3165159b3361Sopenharmony_ci </li> 3166159b3361Sopenharmony_ci <li>m4 - raw pcm support back in</li> 3167159b3361Sopenharmony_ci <li> 3168159b3361Sopenharmony_ci m5 - put in a byte-ordering switch for raw PCM reading 3169159b3361Sopenharmony_ci (just in case) 3170159b3361Sopenharmony_ci </li> 3171159b3361Sopenharmony_ci <li> 3172159b3361Sopenharmony_ci m6 - reworked the whole fft.c file. fft now 10-15% 3173159b3361Sopenharmony_ci faster. 3174159b3361Sopenharmony_ci </li> 3175159b3361Sopenharmony_ci <li> 3176159b3361Sopenharmony_ci m7 - totally new fft routine. exploits fact that this is 3177159b3361Sopenharmony_ci a real->complex fft. About twice as fast as previous 3178159b3361Sopenharmony_ci fastest fft (in m6). (C fft routine is faster than the 3179159b3361Sopenharmony_ci asm one on an m68k!) 3180159b3361Sopenharmony_ci </li> 3181159b3361Sopenharmony_ci <li> 3182159b3361Sopenharmony_ci m8 3183159b3361Sopenharmony_ci <ul> 3184159b3361Sopenharmony_ci <li> 3185159b3361Sopenharmony_ci Now encodes from stdin. Use '-' as the input 3186159b3361Sopenharmony_ci filename. Thanks to <b>Brad Threatt</b> 3187159b3361Sopenharmony_ci </li> 3188159b3361Sopenharmony_ci <li> 3189159b3361Sopenharmony_ci Worked out that the 1024point FFT only ever uses 3190159b3361Sopenharmony_ci the first 6 phi values, and the first 465 energy 3191159b3361Sopenharmony_ci values. Saves a bunch of calculations. 3192159b3361Sopenharmony_ci </li> 3193159b3361Sopenharmony_ci <li> 3194159b3361Sopenharmony_ci Added a speed-up/quality switch. Speed is 3195159b3361Sopenharmony_ci increased but quality is decreased <i>slightly</i>. 3196159b3361Sopenharmony_ci My ears are bad enough not to be able to notice 3197159b3361Sopenharmony_ci the difference in quality at low settings :). 3198159b3361Sopenharmony_ci Setting '-q 1' improves speed by about 10%. '-q 3199159b3361Sopenharmony_ci 100' improves speed by about 26%. Enoding of my 3200159b3361Sopenharmony_ci test track goes from 111s (at default '-q 0') 3201159b3361Sopenharmony_ci to 82s (at -q 100). Thanks to <b>Jan Peman</b> 3202159b3361Sopenharmony_ci for this tip. 3203159b3361Sopenharmony_ci </li> 3204159b3361Sopenharmony_ci </ul> 3205159b3361Sopenharmony_ci </li> 3206159b3361Sopenharmony_ci <li> 3207159b3361Sopenharmony_ci m9 - fixed an error in l3psy.c. numlines[] is overwritten 3208159b3361Sopenharmony_ci with incorrect data. Added a new variable numlines_s[] 3209159b3361Sopenharmony_ci to fix this. Thanks again to <b>Jan Peman</b>. 3210159b3361Sopenharmony_ci </li> 3211159b3361Sopenharmony_ci <li> 3212159b3361Sopenharmony_ci m10 - Down to 106 seconds by selecting a few more compiler 3213159b3361Sopenharmony_ci options. Also added a pow20() function in l3loop.c to 3214159b3361Sopenharmony_ci speed up (ever so slightly) calls to pow(2.0, x) 3215159b3361Sopenharmony_ci </li> 3216159b3361Sopenharmony_ci <li> 3217159b3361Sopenharmony_ci m11 3218159b3361Sopenharmony_ci <ul> 3219159b3361Sopenharmony_ci <li> 3220159b3361Sopenharmony_ci No speedups. Just cleaned up some bits of the code. 3221159b3361Sopenharmony_ci </li> 3222159b3361Sopenharmony_ci <li> 3223159b3361Sopenharmony_ci Changed K&R prototyping to 'normal' format. 3224159b3361Sopenharmony_ci Thanks to <b>Steffan Haeuser</b> for his help here. 3225159b3361Sopenharmony_ci </li> 3226159b3361Sopenharmony_ci <li> 3227159b3361Sopenharmony_ci Changed some C++ style comments to normal C comments in 3228159b3361Sopenharmony_ci huffman.c 3229159b3361Sopenharmony_ci </li> 3230159b3361Sopenharmony_ci <li> 3231159b3361Sopenharmony_ci Removed the #warning from psy_data.h (it was getting 3232159b3361Sopenharmony_ci annoying!) 3233159b3361Sopenharmony_ci </li> 3234159b3361Sopenharmony_ci <li> 3235159b3361Sopenharmony_ci Removed reference in bitstream.c to malloc.h. Is there a 3236159b3361Sopenharmony_ci system left where malloc.h hasn't been superceded by 3237159b3361Sopenharmony_ci stdlib.h? 3238159b3361Sopenharmony_ci </li> 3239159b3361Sopenharmony_ci </ul> 3240159b3361Sopenharmony_ci </li> 3241159b3361Sopenharmony_ci </ul> 3242159b3361Sopenharmony_ci <li> 3243159b3361Sopenharmony_ci In Progess: 3244159b3361Sopenharmony_ci <ul> 3245159b3361Sopenharmony_ci <li> 3246159b3361Sopenharmony_ci my PSYD hack for the spreading functions is only valid for 3247159b3361Sopenharmony_ci 44.1 kHz - Should really put in a "if freq = 44.1 kHz" 3248159b3361Sopenharmony_ci switch for it. Someone might want to extend the speedup for 3249159b3361Sopenharmony_ci 48 and 32 kHz. 3250159b3361Sopenharmony_ci </li> 3251159b3361Sopenharmony_ci <li> 3252159b3361Sopenharmony_ci Putting in Jan Peman's quantanf_init speedup. 3253159b3361Sopenharmony_ci </li> 3254159b3361Sopenharmony_ci </ul> 3255159b3361Sopenharmony_ci </li> 3256159b3361Sopenharmony_ci </ul> 3257159b3361Sopenharmony_ci <hr> 3258159b3361Sopenharmony_ci <center> 3259159b3361Sopenharmony_ci <p> 3260159b3361Sopenharmony_ci <a href="http://validator.w3.org/check?uri=referer"> 3261159b3361Sopenharmony_ci <img src="images/valid-html401.png" alt="Valid HTML 4.01 3262159b3361Sopenharmony_ci Transitional" height="31" width="88"> 3263159b3361Sopenharmony_ci </a> 3264159b3361Sopenharmony_ci </p> 3265159b3361Sopenharmony_ci </center> 3266159b3361Sopenharmony_ci </body> 3267159b3361Sopenharmony_ci</html> 3268
