rockbox

History

Michael Giacomelli c38378eff1 Remove some benchmarking code accidentally included in the previous commit. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@28550 a1c6a512-1295-4272-9138-f99709370657		2010-11-10 18:34:27 +00:00
..
atrac3.c	Remove some benchmarking code accidentally included in the previous commit.	2010-11-10 18:34:27 +00:00
atrac3.h	Commit FS#11709 by me. Introduces ARMv5E optimized iQMF for atrac3 based on packed multiply instructions. Additionally, improves scheduling on arm9 and arm11 and forces cache alignment of buffers on all targets. Accuracy is slightly reduced, but still greater then 16 bit. Clip+ CPU clock required for LP2 files decreases by 13MHz and ARM11 by 18MHz. No performance or accuracy changes on armv4 or non-arm.	2010-11-10 18:29:29 +00:00
atrac3_arm.S	Introduce macros to ARM asm atrac3 filterbank for better readability.	2010-11-10 07:16:04 +00:00
atrac3_armv5e.S	Commit FS#11709 by me. Introduces ARMv5E optimized iQMF for atrac3 based on packed multiply instructions. Additionally, improves scheduling on arm9 and arm11 and forces cache alignment of buffers on all targets. Accuracy is slightly reduced, but still greater then 16 bit. Clip+ CPU clock required for LP2 files decreases by 13MHz and ARM11 by 18MHz. No performance or accuracy changes on armv4 or non-arm.	2010-11-10 18:29:29 +00:00
atrac3data.h	Further optimization and minor clean up of atrac codec: Unroll iqmf_dewindowing for non-ARM speeds up decoder by +10% on H300. Removed some non-used arrays. Codec is still not fully realtime on Coldfire targets.	2010-02-14 14:38:18 +00:00
atrac3data_fixed.h	Reinstate ICONST_ATTR for matrixCoeffs_fix ; remove (silly) gain_tab1 and replace with a simple bitshift in the code	2010-02-20 12:55:35 +00:00
fixp_math.h	Submit FS#11502. Minor optimization to Coldfire assembler in libatrac's fixmul16(). Thanks to Nils Willménius.	2010-07-28 18:17:56 +00:00
libatrac.make
main.c	set mode argument for open() where O_CREAT flag is/can be set.	2010-11-07 11:45:20 +00:00
Makefile.test
README.rockbox
SOURCES	Commit FS#11709 by me. Introduces ARMv5E optimized iQMF for atrac3 based on packed multiply instructions. Additionally, improves scheduling on arm9 and arm11 and forces cache alignment of buffers on all targets. Accuracy is slightly reduced, but still greater then 16 bit. Clip+ CPU clock required for LP2 files decreases by 13MHz and ARM11 by 18MHz. No performance or accuracy changes on armv4 or non-arm.	2010-11-10 18:29:29 +00:00

README.rockbox

Library: libatrac
Imported by : Mohamed Tarek
Import date : 10-August-2009

LICENSING INFORMATION

ffmpeg is licensed under the Lesser GNU General Public License.

IMPORT DETAILS

The decoder is based on ffmpeg-svn r18110 : Mar 21 2009.
Some changes were done on in order to use static VLC tables
according to this commit :
http://git.ffmpeg.org/?p=ffmpeg;a=commit;h=4c20cf13a166577d93f5b2b0abb4609c60104d33

The decoder had been modified to use fixed-point arithmetic.

TESTING

The test program should compile in any Unix-like environment using the
command "make -f Makefile.test".

For ARM targets add -DCPU_ARM to CFLAGS in Makefile.test to make use of 
the asm ARM optimisations in rockbox's mdct library.

For Big-endian targets, change -D"ROCKBOX_LITTLE_ENDIAN=1" 
to -D"ROCKBOX_BIG_ENDIAN=1" in Makefile.test.

Running "./atractest file.rm" will decode the audio data to a WAV file
called "output.wav" in the current directory.