mad-dev July 2003

mad-dev@lists.mars.org

16 participants
17 discussions

Re: How to know the frequency of a frame
by armin.gerritsen＠philips.com 01 Aug '03

01 Aug '03

mad-dev-request(a)lists.mars.org wrote: > Message: 1 > Date: Wed, 30 Jul 2003 16:56:15 -0700 (PDT) > From: Shiming Dong <dong_sm(a)yahoo.com> > To: mad-dev(a)lists.mars.org > Subject: [mad-dev] How to know the frequency of a frame > > --0-1603577723-1059609375=:7610 > Content-Type: text/plain; charset=us-ascii > > Hi there: > > I have one question: I was supposed to implement a program to detect > a specific marker tone. But my problem is how to know the frequency > of each frame? > > Thank you! > > shiming I guess you mean the samplerate? For instance 44,1kHz. In that case it is easy. Once you have decoded a frame, at some point you will synthesise it to PCM samples. struct mad_frame myFrame ; struct mad_synth mySynth ; [...lots of code...] mad_synth_frame( &mySynth, &myFrame ) ; After that mySynth.pcm.samplerate will contain the samplerate. For instance 44100 for 44,1kHz. Regards, Armin

3 2

How to know the frequency of a frame
by Shiming Dong 31 Jul '03

31 Jul '03

Hi there: I have one question: I was supposed to implement a program to detect a specific marker tone. But my problem is how to know the frequency of each frame? Thank you! shiming --------------------------------- Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software

2 1

Configure error
by hank.jan＠billionton.com.tw 31 Jul '03

31 Jul '03

Hi, I encounted some trouble when I did the cross compile. I am not familiar with linux environment. So I may ask a stupid question. Any help will be appreciated. I installed the arm cross compiler toolchian, and tried to compile madplay. But when I ran ./configure, I met an error with the message : checking whether byte ordering is bigendian... configure: error: can not run test program while cross compiling. I have ever succeeded the configuration, but I am not sure what to do. Can anybody help me? Thanks in advance.

2 1

Minimad: output PCM data overflow ?
by Yeh You-Ying 31 Jul '03

31 Jul '03

mad-dev-request(a)lists.mars.org mentioned: > Could be, but in my experience even something as simple as: > >signed int MAD_SimpleScale( mad_fixed_t sample ) >{ > // Round > sample += ( 1L << (MAD_F_FRACBITS - 16) ) ; > > // Clip > if ( sample >= MAD_F_ONE ) > sample = MAD_F_ONE - 1 ; > else > if ( sample < -MAD_F_ONE ) > sample = -MAD_F_ONE ; > > // Quantise > return sample >> (MAD_F_FRACBITS + 1 - 16) ; >} > > >... already produces nice results. >If you forget about the clipping though, the quality s#cks! >I'm not sure, but I believe the standard mad-example 'forgets' the clipping. >(Could be wrong though!) >I don't know what minimad.c uses ... > >If that is not it, check the compiler settings. Some very agressive >optimisations may cause rounding-errors, although this usually occures with >floating point numbers, and MAD's strength is that it doesn't use them ... > >I know, I'm just quessing, but maybe this helps somehow. > >Regards, > >Armin > > > > Dear Armin, Thanks for your information. I've check the scale function in minimad.c and it's the same as the one you mentioned. And I 've tried several ways to change the compiler flags, but still no work. I used dist10 decoder to generate another PCM raw data with the same demo.mp3 I also used Sound Forge to analyze these two PCM raw data. The one generated by minimad seems to be overflow in sample value, the maximum and minimun value are -32768 and 32768 (2^16). Any hint about this ? I can email to statistic information to you if you'd like. Thnaks in advance. Yeh You-Ying

2 2

Re: [mad-dev] Palm OS port
by Boris Raicheff 29 Jul '03

29 Jul '03

>From: Jesse McGrew <jmcgrew(a)hansprestige.com> >To: mad-dev(a)lists.mars.org >Subject: [mad-dev] Palm OS port >Date: Mon, 28 Jul 2003 21:19:48 -0700 > >I want to write a streaming audio player for Palm OS, so I'm thinking of >trying to port libmad to Palm > OS. Newer Palms have ARM processors but >user apps are mostly emulated 68k code, so theoretically > it'll be >possible as long as I can convert the whole library to an "ARMlet". > >I don't want to reinvent the wheel or step on anyone's toes, though. Is >there already a Palm port in > progress? Yes, there is. In fact, there are at least 4 people (the ones that I know of) working on different Palm OS ports. Unfortunately, your idea to convert the whole library to an ARMlet is not easy at all. So you won't be able to just include the sources in your Codewarrior or whatever you're using, and hit compile. Don't forget the 64k limit. There are many issues to deal with (like the lookup tables, which are too large, and the struct alignment thing, the name a few). Regards, Boris Raicheff _________________________________________________________________ MSN 8 with e-mail virus protection service: 2 months FREE* http://join.msn.com/?page=features/virus

2 1

Re: PCM output sample overflow ?
by Yeh You-Ying 29 Jul '03

29 Jul '03

1 0

Palm OS port
by Jesse McGrew 29 Jul '03

29 Jul '03

I want to write a streaming audio player for Palm OS, so I'm thinking of trying to port libmad to Palm OS. Newer Palms have ARM processors but user apps are mostly emulated 68k code, so theoretically it'll be possible as long as I can convert the whole library to an "ARMlet". I don't want to reinvent the wheel or step on anyone's toes, though. Is there already a Palm port in progress?

1 0

Re: mad-dev digest, Vol 1 #278 - 1 msg
by armin.gerritsen＠philips.com 28 Jul '03

28 Jul '03

> Thanks for your reply. > I knew that the output is raw PCM data. > And the Sound Forge can play raw PCM data no matter what extsion > filename is. > I can hear the PCM data via Sound Forge, just the quality is not good, > too much nosie. > As the decription of scale( ) in minimad.c, there is no noise shaping > for scaling. > It that the reason about the quality issue ? Could be, but in my experience even something as simple as: signed int MAD_SimpleScale( mad_fixed_t sample ) { // Round sample += ( 1L << (MAD_F_FRACBITS - 16) ) ; // Clip if ( sample >= MAD_F_ONE ) sample = MAD_F_ONE - 1 ; else if ( sample < -MAD_F_ONE ) sample = -MAD_F_ONE ; // Quantise return sample >> (MAD_F_FRACBITS + 1 - 16) ; } ... already produces nice results. If you forget about the clipping though, the quality s#cks! I'm not sure, but I believe the standard mad-example 'forgets' the clipping. (Could be wrong though!) I don't know what minimad.c uses ... If that is not it, check the compiler settings. Some very agressive optimisations may cause rounding-errors, although this usually occures with floating point numbers, and MAD's strength is that it doesn't use them ... I know, I'm just quessing, but maybe this helps somehow. Regards, Armin

1 0

Re: mad-dev digest, Vol 1 #277 - 1 msg
by Yeh You-Ying 26 Jul '03

26 Jul '03

>Today's Topics: > > 1. Re: PCM Outout Quality Issue (Rob Leslie) > > > > > Subject: > Re: [mad-dev] PCM Outout Quality Issue > (From): > Rob Leslie <rob(a)mars.org> > Date: > Fri, 25 Jul 2003 13:36:13 -0700 > (To): > mad-dev(a)lists.mars.org > > > On Wednesday, July 23, 2003, at 08:27 PM, Yeh You-Ying wrote: > >> What I did is just to decode a mp3 file with my minimad, and generate >> an output, for instance, pcm.wav. >> I played this raw pcm data with Sound Forge. >> Ohhh...it sounds terrible. > > > It's important to realize minimad doesn't output a WAV file. It just > writes raw PCM data in little-endian format. Make sure Sound Forge > isn't treating the file as a WAV file or reading the PCM as big-endian > samples. > Dear Rob, Thanks for your reply. I knew that the output is raw PCM data. And the Sound Forge can play raw PCM data no matter what extsion filename is. I can hear the PCM data via Sound Forge, just the quality is not good, too much nosie. As the decription of scale( ) in minimad.c, there is no noise shaping for scaling. It that the reason about the quality issue ? Thanks a lot.

1 0

PCM Outout Quality Issue
by Yeh You-Ying 25 Jul '03

25 Jul '03

Hello Everyone, I'm new for MEPG decode. I just finished struggling from set-up MinGW/GCC environmnet. I've successfully compiled and linked minimad. There must be something wrong with me. I got terrible PCM output. Any hint to improve it ? What I did is just to decode a mp3 file with my minimad, and generate an output, for instance, pcm.wav. I played this raw pcm data with Sound Forge. Ohhh...it sounds terrible. My platform is x86 (P4), so here is the compiling CFLAG in my makefile, HAVE_CONFIG_H FPM_INTEL. and in comfig.h #define inline __inline #define STDC_HEADERS 1 #define HAVE_ERRNO_H 1 #define HAVE_FCNTL_H 1 #define HAVE_LIMITS_H 1 #define HAVE_SYS_TYPES_H 1 #define SIZEOF_INT 4 #define SIZEOF_LONG 4 #define SIZEOF_LONG_LONG 8 Did I miss something important ? Thanks in advance.

2 1

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 July 2003