atracdenc/src/atrac/atrac1.cpp, branch 0.2.3 OpenSource ATRAC1 ATRAC3 Encoder https://code.mastervirt.ru/atracdenc/atom?h=0.2.3 2026-07-08T18:00:52Z Validate ATRAC1 decoder input and skip malformed frames 2026-07-08T18:00:52Z Daniil Cherednik dan.cherednik@gmail.com 2026-06-26T21:32:56Z urn:sha1:1c35c32838452997c304761c7f816d5604eaa7a6 The decoder now skips an invalid frame and emits silence. (cherry picked from commit 75b3d391652e8cb61a98685fcfc5185891007429) Simplify scaler code reusage. 2025-01-05T01:57:14Z Daniil Cherednik dan.cherednik@gmail.com 2025-01-04T23:30:02Z urn:sha1:0fe8e1a0d1a495e41a281f54af31d6713c0de01d [refactoring] Get rid of TFloat parameter. Use float. 2024-12-21T15:30:07Z Daniil Cherednik dan.cherednik@gmail.com 2024-12-20T21:31:20Z urn:sha1:74d6e04c21bddd435bd74c34dbe027b883772a76 [refactoring] Get rid of inheritance from data. 2024-12-12T22:52:19Z Daniil Cherednik dan.cherednik@gmail.com 2024-12-12T22:26:24Z urn:sha1:39986e653def9d1130235c1cd6a38c3251bb6d79 Allow to configure atracdenc to use float instead of double 2024-07-09T21:07:40Z Daniil Cherednik dan.cherednik@gmail.com 2024-07-08T21:42:48Z urn:sha1:c4b19af8391640f143aaf446658354dcf01432d1 The default is double for hystorical reason. Actually no any reason to use double precision float point calculation for all processing. AT3P encoding will use float except places were we really need double precision. So this default will be changed to float in near future and this option will be removed. Support for MS Windows platform 2019-05-14T22:14:42Z Daniil Cherednik dan.cherednik@gmail.com 2019-05-14T22:14:42Z urn:sha1:ed9629395ce9696164c1d617d573a47982e82169 - VS build (2017 tested) - Media Foundation Framework support to read/write pcm data (instead of libsndfile) Set LGPL v2.1 license 2017-02-12T13:14:56Z Daniil Cherednik dan.cherednik@gmail.com 2017-02-12T13:14:56Z urn:sha1:3740b709a39e4a1eca7a5fbd598e8554944ce0e7 Dirty implementation of atrac3 encoder: 2016-09-02T18:21:28Z Daniil Cherednik dan.cherednik@gmail.com 2016-03-13T06:49:33Z urn:sha1:cfaa2cd39b7256a868a4f5cd83aac207df6bd1b3 - no JS mode - constant quantiser for tonal components - gain controll implemented but produces some artifacts with real signals. - etc... experimental first implementation of ATRAC encoder 2015-10-30T23:14:04Z Daniil Cherednik dan.cherednik@gmail.com 2015-10-30T23:14:04Z urn:sha1:592b4bd68f3eb9d1cdbcda74feaffc7b1d5f0485 current limitations: - only long window - naive MDCT (O(n^2)) - bad table of fixed bit allocation - bad usage of CBR - time accuracy is not guaranteed - dirty, not optimized code