mad-dev December 2003

mad-dev@lists.mars.org

8 participants
7 discussions

mad in motorola dsp?
by Kampschreur_FP 11 Dec '03

11 Dec '03

Hello, I am de student from the Netherlands and I am working on a mpeg project. I have to decode mpeg realtime thus not out of a file into de decoder and into a file, but from an satilite into (buffer) the decoder into a DAC so there comes sound out of it. My question is Does enyone know if de libmab source fits into a Motorola DSP 56362 ? This is a 24 bit DSP so I have some problems whit fitting de intergers. thank you, -François-

1 0

urgent: Questions about how to build a ARM based madplay
by Kzhang 11 Dec '03

11 Dec '03

Hi friends, I download both of mad-0.14.2b and madplay-0.15.0b, and I try to build a ARM based player, but I met problem to change gcc to arm-linux-gcc with ./configure command. I work on a ReHat 9.0 PC as host, and use arm cross compiler 2.95.3 toolchain. I try " CC=arm-linux-gcc ./configure" and "env CC=armlinux-gcc", it did not work. Please give me some tips how to build it, I very appreciate for your help. Thanks. Keji Zhang kzhang(a)neomagic.com

2 1

MAD 0.15 Accuracy
by Grigory A. 04 Dec '03

04 Dec '03

Hi! I tested my implementation of some math function for MAD 0.15 library. And I have found this: function mad_fixed_t III_requantize(unsigned int value, signed int exp) in layer3.c file input value=8 exp=-23; that means 8^(4/3)*2(-5.75) correct? result obtained by mad_f_todouble(requantized)is 0.297241 but calculator says >> 8^(4/3)*2^(-5.75) ans = 0.29730177875068 for Q15 format its means ~2 bits difference. >>(0.29730177875068-0.29724100000000)*32768 ans = 1.99159810228230 my Q15 arithmetics gives: 9743/32768=0.29733276367188 >>(0.29730177875068-0.29733276367188)*32768 ans = -1.01531389788215 -1 bit difference And total difference between MAD math and my = 3 bits! Maybe I have to compare my math not with MAD - but with some float point implementation? -- Best regards, Grigory mailto:Ryhor@tut.by

2 2

IMDCT accuracy
by Rafael Avila de Espíndola 03 Dec '03

03 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Testing the Rob Leslie implementation of the Szu-Wei Lee algorithm I notice that his modification (multiplying the input of the DCTIV by 2 and dividing the output) generally improves the accuracy but, in some cases, may cause a overflow. I have not found a .mp3 file that triggers a overflow, but the following data does. The present IMDCT implementation in libmad 0.15 return a valid result. 0x442D46F 0x217CC40 0x8E26853 0xF1888C6 0x921BF17 0xD282 0xF060CB8 0x3E82D68 0xBCA6124 0x1217700 0x9500F6D 0x3E9B4B6 0xC4B7675 0x6AE1645 0x56C266C 0x4463FD6 0x7EE5176 0xC408F90 Is there some evidence that no real mp3 may trigger an overflow? Thanks for any help. Rafael Espíndola -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.3 (GNU/Linux) iD8DBQE/y2NVLlrfGJ8JUHwRAibvAJ9OtOIAoFi3UlKMqscFYUZIquCeGwCcCFLC 563s6VSqX5vnPxdTZNZAR/g= =yIV7 -----END PGP SIGNATURE-----

2 1

MAD doesn't decode last frame in stream buffer unless it has 8 bytes extra
by Jedediah Smith 03 Dec '03

03 Dec '03

Pass a buffer to mad_stream_buffer which ends exactly on a frame boundary. mad_frame_decode will fail on the last frame and set the error to MAD_ERROR_BUFLEN even though entire frame is there. As much as 7 extra bytes can be added to the end of the buffer and the last frame decode will still fail. If 8 bytes are added, the frame decode will succeed. Is this a bug or by design? If by design, how can I decode the last frame without adding padding after it?

2 1

Problem with this mp3
by starz 02 Dec '03

02 Dec '03

OK, I'm having problems using MPD (http://musicpd.org) or mpg321 with this file http://thelinuxshow.com/archives/2003/tls-11-25-2003.mp3 (actually any of the mp3s in that archive). Alsaplayer plays it fine, and some others have said it plays fine in MPD, which makes me think this is a libao problem, but I have tried this in XMMS with the xmms-mad module and it doesn't seem to work there either. Any ideas? Is this a known bug? Many thanks, sbh P.S. mpg123 and mpg321 report different values for the bitrate and samples... please don't ask me anything too technical as I'm the definition of 'enduser' :) _____________________________________________ Free email with personality! Over 200 domains! http://www.MyOwnEmail.com Looking for friendships,romance and more? http://www.MyOwnFriends.com

3 3

random seeking with MAD
by Jedediah Smith 01 Dec '03

01 Dec '03

How would I implement fast, sample-resolution random seeking with MAD? In other words, I need to be able to reposition the stream to a specific sample and get resynced in constant time and without too much overhead. The MP3 also needs to be streamed from disk rather than loaded entirely in RAM. I imagine the main problem would be reconstructing the bit pool at the seek point. From my understanding of the MP3 format, the furthest back bit pool data can be stored is 9 frames. So here is my idea: 1. On initial load, create an index of all the frame offsets in the mp3 and cache it in RAM 2. When seeking to sample N, calculate the containing frame number F as F=N/1152 3. Use the cached frame index to find the file offset of frame F-9 4. Seek to that offset in the file and begin decoding 5. Decode 9 frames silently, ignoring any MAD_ERROR_BADDATAPTR 6. Begin playing at sample N which is within the current frame Will this work? Is there a faster way that avoids decoding the audio data for the extra 9 frames?

2 5

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

mad-dev December 2003