mad-user December 2000

mad-user@lists.mars.org

12 participants
35 discussions

Re: [mad-user] Auto Attenuation and decoding.
by Juan Covas 13 Dec '00

13 Dec '00

Greetings, Doh! :) heh sorry for the stupid question :| I meant to write about using the diskwriter plug-in to convert to a wave file, but I assume now that it does take place too. Heh, thanks. Best Regards, Juan _____________________________________________________________________________________ Get more from the Web. FREE MSN Explorer download : http://explorer.msn.com

2 1

Auto Attenuation and decoding.
by Juan Covas 13 Dec '00

13 Dec '00

Greetings, Simple question... auto attenuation also takes action while a file is being decoded, correct? Best Regards, Juan _____________________________________________________________________________________ Get more from the Web. FREE MSN Explorer download : http://explorer.msn.com

2 1

Winamp plug-in pre-release 0.12.3b-pre2
by Rob Leslie 13 Dec '00

13 Dec '00

It's not quite finished, but I thought some people might like another pre-release of the plug-in for Winamp: http://www.mars.org/home/rob/proj/mpeg/mad-plugin/in_mad-0.12.3b-pre2.zip You can at least now turn on (but not configure) the auto attenuation, and I also added some joint stereo statistics. You'll notice some other config options pending too. This version also detects a special (rare) kind of bitstream corruption and discards the corrupt frames instead of trying to play them. Cheers, -rob

1 0

MAD DLL
by FunkyMan 11 Dec '00

11 Dec '00

I wanted to use the MAD decoder in my MP3 Player, but as a matter of facts, i wasn't able to compile it under windows :-( has anyone build a DLL that can be used by a windows program (a kind of LIBMAD.DLL), or anyone made a VC++ project to compile MAD under windows ? thanx in advance. FunkyMan PS: my MP3 Player is mean't to be built with VC++ 6... so i someone could help me compiling the LIBMAD library ... ++ TBX Player by the TeufCorp (http://www.teufcorp.fr.st)

3 2

Sv: Sv: [mad-user] Attenuation, clipping
by Peter Olufsen 11 Dec '00

11 Dec '00

>> 3) I think it would be nice to have some form of auto-attenuation (and also >> auto normalize-to-peak or even amping/compressing), but to make it perfect >> wouldn't you have to scan all your files first, or could you make some setup >> for that ? > >The expected mode of use would be that an appropriate setting would be learned >the first time a song is played, and recalled when the song is played again. > Ok, but where would you store the info ? And would it be fast enough with large song-collections ? If you do this it would perhaps be good to put the plugin in a "learn"-mode so you could get the info from all your songs fast and write it in a file for later use. >> I have testet some files and they typically have clipping about 1db, but i >> found one file with 6db clipping in one sample (?), and if you >> auto-attenuate that the result is not good, so you should have some form of >> setup to prevent that. > >This is a good point, but also a tough decision... how much clipping is >acceptable? > >Are there any audio technicians on the list (Anthony? :) who could go into >details with me about sound energy levels, compression, normalization, etc.? I found another file with a few (160 ?) samples with 8db clipping so i think that some kind of setup would be nice! Thanks, Peter

4 4

MAD and insane bitrates
by whee＠e247.com 08 Dec '00

08 Dec '00

Apparently MAD handles 550kbps mp3s just fine, and winamp's default decoder does not. Just one more reason to use MAD .D (the CVS LAME I got allows bitrates up to 550kbps) -whee Get your free e24/7 email at http://www.e247.com/eyelandreg.jsp Send & receive large files with 100MB of storage

1 0

Attenuating pre-release
by Rob Leslie 07 Dec '00

07 Dec '00

Here's a mailing list exclusive: the following URL leads to a pre-release version of the MAD plug-in. http://www.mars.org/home/rob/proj/mpeg/mad-plugin/in_mad-0.12.3b-pre1.zip This version includes a special "Statistics" tab in the file info dialog for the currently playing song. Among other things, you'll find there an indication of how often and how much samples are being clipped. Furthermore, you can choose to attenuate the signal on-the-fly based on the amount of clipping. Attenuation is performed internally on high-precision pre-dithered samples, so it should not affect sound quality except to improve it by eliminating clipped samples. Further ideas I have in this direction include: - "Automatic clipping attenuation" configuration option - remembering the attenuation setting on a per-file basis Currently the attenuation setting is automatically reset at the beginning of each song. I'd like feedback on where this is going. Is this useful? If an automatic attenuation option is made available, would the manual buttons still be useful? Would there be any benefit in an even more manual configuration like, say, a slider? What about amplification rather than attenuation? Cheers, -rob

4 5

Sv: [mad-user] Attenuation, clipping
by Peter Olufsen 06 Dec '00

06 Dec '00

It was nice to have that information (and it's a great plug-in) ! 1) Is it correct that the clipping appears in the decoder even though the file was encoded "correct" without clipped samples ? 2) What is the "Other"-errors in the info-box, it seems that there is 1 og 2 in every file ? 3) I think it would be nice to have some form of auto-attenuation (and also auto normalize-to-peak or even amping/compressing), but to make it perfect wouldn't you have to scan all your files first, or could you make some setup for that ? I have testet some files and they typically have clipping about 1db, but i found one file with 6db clipping in one sample (?), and if you auto-attenuate that the result is not good, so you should have some form of setup to prevent that. I have also tested some auto-amp-plugin and the result wasen't good because you could hear that the sound-level was adjusted when the songs changed in level. 4) Would it be possible to use MAD's 24/32-bit output with some DSP-plugin like DFX ? Or do you loose the advantage from MAD when you use DSP ? Thanks, Peter

2 1

Sv: Sv: [mad-user] Attenuation, clipping
by Peter Olufsen 06 Dec '00

06 Dec '00

From: Anthony Airon Oetzmann <Anthony.Oetzmann(a)epost.de> To: Peter Olufsen <po(a)dsinet.dk> Dato: 6. december 2000 01:49 Emne: Re: Sv: [mad-user] Attenuation, clipping >>1) Is it correct that the clipping appears in the decoder even though the file was encoded "correct" without clipped samples ? > >You're thinking lossless compression here, 'cause you won't get a signal >different from the original with lossless compression such as Monkeyaudio. >However, MP3 reconstructs an audio signal from less than the original signal >plus noise. Add for example one frequency x and another frequency 2*x, then >you'll have some areas that are boosted, and some that are attenuated. When 2*x >is filtered out again(or gets discarded in some places, because the encoding >algorythem thought you won't hear it) then some places will get attenuated that >were peaks and some will be boosted that weren't peaks before. And sometimes in >complex signals that are mastered to use the maximum of the 16 bit resolution on >a CD and are mastered to extreme loudness(single releases are mangled this way a >LOT), the signal's reconstruction overshoots, because the stuff that got >filtered out in the encoding process meant that the tightly controlled >frequencies of the signal will have some things missing. You get some peaks and >attenuations you previously didn't have. You can check this easily, by taking a >ripped track and using an EQ in a sample editor to EQ 100Hz by -6 dB for >example. You almost always get clipped samples, because of the great energy that >lower frequencies take and thus command more amplitude and heavier changes to >the signal. Ok, i get it ! >>3) I think it would be nice to have some form of auto-attenuation (and also auto normalize-to-peak or even amping/compressing), but to make it perfect wouldn't you have to scan all your files first, or could you make some setup for that ? >>I have testet some files and they typically have clipping about 1db, but i found one file with 6db clipping in one sample (?), and if you auto-attenuate that the result is not good, so you should have some form of setup to prevent that. >>I have also tested some auto-amp-plugin and the result wasen't good because you could hear that the sound-level was adjusted when the songs changed in level. > >I propose to set a minimum attenuation level of -2 dB. That seems reasonable for >most material. Automatic attenuation would set in with a look-ahead algorhythem, >and perhaps a good limiter could be used in cases of less than five clipped >samples. I agree. >>4) Would it be possible to use MAD's 24/32-bit output with some DSP-plugin like DFX ? >>Or do you loose the advantage from MAD when you use DSP ? > >MAD needs to do this itself. DSP plugins get the decoded signal. >The signal path is DECODER PLUGIN -> (optional DPS Plugins->) OUTPUT PLUGIN But it would seem like a good idea that MAD delivered its 32-bit output to DFX (or some other DSP-plugin) instead of MAD going 16->32->16 and DFX doing the same thing, or am i wrong here ? Peter

1 0

Attenuation, clipping
by Rob Leslie 05 Dec '00

05 Dec '00

I just want to provide a little background for anyone confused by all this clipping and attenuation nonsense. :-) During the decoding process, PCM samples are reconstructed from MPEG audio frames. The nominal output range of the PCM samples can be considered to be a fraction number between -1.0 and +1.0. This is eventually scaled to a number with certain resolution, for example 0..255 for 8-bit output, or -32768..32767 for 16-bit output, etc. Because of the nature of the decoding process, it sometimes happens that PCM samples are calculated to have a value less than -1.0 or greater than +1.0. This isn't really an error, it's just an artifact of the lossy compression and depends a lot on the encoder and the strength of the original signal. Most decoders have little choice but to clip the output signal to the maximum representable values. This adds distortion to the signal. By "attenuating" the signal, the entire signal can be weakened proportionally so that the peaks fall exactly on the -1.0 or +1.0 full scale limits rather than beyond them. It is a digital volume adjustment to prevent clipping. Internally, MAD represents signals with about 169 dB of dynamic range and enough headroom to calculate samples 18 dB above full scale. It's therefore possible to further manipulate the signal without losing quality before it is finally scaled and dithered to the output resolution. The "Attenuate" button in the pre-release activates an attenuation filter to weaken the output signal based on the amount of clipping experienced. An automatic version of this would do the same thing as soon as clipping is detected. Cheers, -rob

1 0

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-user December 2000