Hi,
When applying wavegain to audio files I have noticed that all audio files are sampled to a low volume even those that were recorded low too.

My question is:
Is there a way that those low ones could be processed to be at equal volume matching the high ones.
I feel that the process make them sound at a lower db, but they are not at equal volume.
Am I wrong? Am I not using the proper setting?

Any advise about this will be higly appreciated...

Thanks

Gramophone

If you provide the settings you use it is more likely that someone might be able to tell if there is something wrong with them.

It sounds very strange that the files don't have equal loudness after running them trough wavegain. Are you by any chance using foobar2k for playback?

If the files you have are ripped from modern CD:s then it is quite normal that most (if not all) are lowered when running wavegain. This is to assure that samples aren't clipped.

is it then better to run wavegain before encoding or mp3gain after encoding?

That`s fearly up to you, depending on what you will do with the files later on.
If it is for solely to playback true PC, i belive mp3gain is a good method.
If you on the other hand are going to burn theese to audio cd, the corection done by mp3gain will be lost after decoding. There exist a couple of burning applications
that use Wavegain after decoding to temp before burning, (burnatonce and burrrn).



to gain your files before encoding resoults in mainly two things,
RG will lower the overall volume and thus. heavy commpressed music will
be a easyer load for the encoder and maybe a bit faster encoding.
The level of innput will give the encoder better work conditions, and likely
to give you a better resoult. - just my 2c.

has this been proven or is it just an opinion?

This is not true. MP3gain modifies the gain field in each frame, which is part of the MP3 specs and all compliant MP3 decoders should use this when decoding.

I have never heard that, and I find it very hard to believe.

that would the case with replaygain that just applies a tag and doesnt truly modify the file the way mp3gain does...

@magic.

My bad, i wasn`t aware of that. When it regards the speed issue, this sounds
quite logical to me, it`s going to be a bit hard to prove, so you just need to try it.
@holkie. yes this is just my opinion. As i said to magic75, if you need proof,
the only answer i can give, is to try.

I don't know about "faster encoding" but if you apply Wavegain to songs that need a high attenuation, the resulting mp3 will generally have a lower bitrate than the original. The file is using less of the available range, and thus the encoder needs less bits to represent the signal.

There was a thread on this a few months back, but I can't seem to locate it at the moment.

so, thats only for vbr?

maybe here:

http://www.hydrogenaudio.org/forums/index....showtopic=10637

Ok, I'm not using foobar2 to play them, I use any player...windows media etc..
I'm using wavegain frontend.
These are my settigs:

•Radio gain

•Extra gain: 0.0dB
•6 dB hard limiter

•Dither:no dither(default)
•Output: 24 bit signed PCM or 16 bit signed PCM (default)

I don't know, but I thought that if a track was recorded at -3dB
and other was recorded at 0.0dB and I apply wavegain to these two
tracks, wavegain would leave the track at 0.0 untouched and the one
at -3 would be level to 0.0 to match the loudness of the two of them...
Is that the way wagain works?
Because I see that tracks that need to have more loudness, don't get it
after I apply wavegain to them.

Any light about this?

Thanks

Gramophone

How do you see that the files that need more loudness does not get it? Do you run wavegain again on the files and just calculate?

I am not sure but maybe the 6 dB hard limiter could be messing things up.

http://replaygain.hydrogenaudio.org//faq_quiet.html

I have just tested this out, and allthough john33 would be the one who should write
this, (he`s the author). when you have a file who need ex: -2.20 dB,
wavegain do correct this by scaling approx. 0.78. so the signal becames 0.00db.
A file of +2.14 dB wold be corrected by 1.28. I say approx. because i reckon
this is set by the peak amplitude.- This is for 16b, by resample to 24b, the amplitude is increased by a very small number.
Imagine your working on a digital picture, you have cleaned it up so no
"noise" is present. Then you figure out that you need it in higher resolution,
then you notise a lot of noise.-
So the perfect way to "declipp"/lower the volume on digital sound, would be
first resample to 24, alter what you wan`t, then back to 16.

just my 2c, there may be "glitches" in this, factores i have not taken in
consideration.

David's (2Bdecided) FAQ answers this, but if you're not applying any additional gain, I would not use the hard limiter as it's unlikely to be necessary. However, if you want to increase the volume for all your tracks, then apply a manual gain of +3dB and apply the hard limiter (just in case). The manual gain simply adds the required gain to the base level.

Gramophone, when you said that one file is recorded at -3dB, did you mean that this is what wavegain reports, or did you mean that the peak amplitude of the recording is 3 dB below maximum?

Those two are not the same thing. If you mean the latter, then john33 and 2Bdecided has the answer for you...