Mp3 Gain and Wave Gain

Topic: Mp3 Gain and Wave Gain (Read 14645 times) previous topic - next topic

0 Members and 1 Guest are viewing this topic.

Mp3 Gain and Wave Gain

Reply #25 – 2002-07-26 00:29:43

Blaaargh

Fixed.

Mp3 Gain and Wave Gain

Reply #26 – 2002-07-26 05:35:41

john33,

I checked the help, so to be clear to apply max no clip gain/remove clipping I would:

volumax -c=input.wav -a -l -y output.wav

BTW keep in mind it's the first time I use commandline so excuse me if I'm very wrong

Mp3 Gain and Wave Gain

Reply #27 – 2002-07-26 12:04:20

@westgroveg

The options are:

Code: [Select]

 Volumax v0.1 - Compiled Jun 27 2002. Copyright (c) 2002 John Edwards

 Usage: volumax [options] input.wav [...]



 OPTIONS

  -h, --help       Prints this help information.

  -a, --album      Use Album volume setting, or

  -t, --title      Use Track volume setting(DEFAULT).

  -c, --calculate  Calculates and prints proposed volume adjustments, but

                   DOES NOT APPLY THEM. This is the DEFAULT.

  -y, --apply      Calculates and APPLIES proposed volume adjustments.

  -l, --level X    Apply reduction to maximum non-clipping volume in dB, where

              X =  any floating point number between 0.0 and -12.0(DEFAULT 0.0)

 FORMAT OPTIONS (One option ONLY may be used)

  -b, --bits X    Set output sample format, where X =

             1    for        8 bit PCM data.

             2    for       16 bit PCM data(DEFAULT).

             3    for       24 bit PCM data.

             4    for       32 bit PCM data.

             5    for       32 bit floats.

 INPUT FILES

  Volumax input files must be 16 bit integer wave files with

  1 or 2 channels and a sample rate of 48000Hz, 44100Hz,

  32000Hz, 24000Hz, 22050Hz, 16000Hz, 12000Hz, 11025Hz or 8000Hz.

  Wildcards (?, *) can be used in the filename.

A simple command line would, therefore, be:

Code: [Select]

volumax -a -y *.wav

This will adjust the volume on all .wav files in the directory on an album basis to a max volume no-clipping level. The output overwrites the input, so if you want to keep the original wav file, you'll need to copy it first. The '-l' options takes a value of between 0.00 and -12.00dB as a reduction in the max volume level to be applied; eg. '-l -6.25'. The '-c' option, which is the default, calculates, but does not apply the adjustment. The '-y' option causes the adjustment to be calculated AND applied.

Mp3 Gain and Wave Gain

Reply #28 – 2002-07-26 12:14:02

Can I remind people of something here:

If you're changing the volume of a .wav file, the only thing you can do is decrease the quality (slightly) - there is no way that you are increasing the quality.

So, unless you have a reason for making each album as loud as possible (e.g. crappy soundcard with lots of noise), this is a pointless exercise.

If you're trying to match the loudness of several albums, then arguably this brings about a big advantage (consistant volume) for a small disadvantage (slight decrease in quality). However, if you're just trying to make them all as loud as possible, then (with ideal equpiment) there's no advantage.

In contrast, mp3gain (and all the other format specific implementations of replaygain) create a win/win situation - you equalise the volume without sacrificing quality.

I don't want to stop anyone using this software who has a genuine reason for it (e.g. portable player is too quiet, sound card isn't great etc etc) - but it's certainly NOT something that everyone should do to every album.

Cheers,
David.

Mp3 Gain and Wave Gain

Reply #29 – 2002-07-26 12:20:55

May I just endorse what David has just said. I only produced this originally for a guy who had precisely the problem David described, ie, portable player with low output volume. I certainly didn't intend giving the impression this was a good thing to do in the normal course of things.

Mp3 Gain and Wave Gain

Reply #30 – 2002-07-26 14:49:11

are 2Bdecided's comments in reference to Volumax or wavegain, or both?

Mp3 Gain and Wave Gain

Reply #31 – 2002-07-26 15:00:26

Let me put this in context: we're talking about a very very small reduction in quality. It's much smaller than the reduction in quality due to the very very best audio codec (e.g. mpc braindead). But it's there, so I thought I'd mention it.

If you want to use replaygain with a format that is totally unsupported, then use Wavgain. However, it's better to use the format dependent versions.

Therefore:
.wav > mp3 > mp3gain
is better than
.wav > wavgain > mp3

Cheers,
David.

Mp3 Gain and Wave Gain

Reply #32 – 2002-07-28 11:56:20

Thanks 2Bdecided, but what about when going the other way?

I have mp3 files (never mind where I got them from...) and when I select some of them in order to make a compilation CD I would like to normalize them relative to this specific selection.

I don't want to alter the original mp3 files because I might be using some of them later in another selection where a different normalization level would be more appropriate.

So I thought that normalizing the intermediate wave files (which I delete after a successful burn) is the right choice.

Does this make any sense?

Thanks!

Mp3 Gain and Wave Gain

Reply #33 – 2002-07-28 16:19:56

Quote

Originally posted by ahyny

So I thought that normalizing the intermediate wave files (which I delete after a successful burn) is the right choice.

Does this make any sense?

Qualitywise it is better to apply the changes to the mp3 before decoding. When changing the volume of the wav-files you will decrease the quality. Applying the changes to the mp3 itself is done without affecting the quality, and can be undone later.
I can understand your point of not wanting to change your "originals", but it is a better way to do so. You can apply a volume change to them many times without any quality reduction at all. And you can always make a copy before applying the changes...

/Andreas

Mp3 Gain and Wave Gain

Reply #34 – 2002-07-29 11:27:36

ahyny,

Whichever. It's splitting hairs. (i.e. it makes almost no difference).

If I were doing this, and wanted the best possible results, I would copy the mp3s I wanted to burn to audio CD into a new folder, run mp3gain on them, listen to them to check I agreed with what mp3gain had done, and then decode them to .wav using Winamp with the MAD plug-in, and no DSP/EQ. Then I'd burn those .wavs to CD.

But if you decode mp3 to .wav, run wavgain, and the burn the modified .wavs, it'll be fine. The rounding errors are very small. As long as the mp3s weren't clipping (and were NOT very quiet), it probably doesn't matter. If the mp3s are clipping, or contain quiet passages, then there's a chance that you might hear the difference between the two methods.

Now, john33 tell me - does wavgain dither the result? It might be a useful option.

Cheers,
David.

Mp3 Gain and Wave Gain

Reply #35 – 2002-07-29 11:48:41

At the moment, David, no it doesn't. I'm quite happy to look at adding it as an option though.

Mp3 Gain and Wave Gain

Reply #36 – 2002-07-29 16:07:54

This is a very interesting and informative thread, thank you.

Mp3 Gain and Wave Gain

Reply #37 – 2002-07-30 04:53:11

what about implementing replaygain with loseless formats, like monkey's or lpac? apegain... i like it

Mp3 Gain and Wave Gain

Reply #38 – 2002-07-30 13:20:23

Thanks!!!

This was very helpful and educating!

ahyny.

Notice