RG Volume Normalisation vs Replay Gain

[a.wright]

Hi,

I am getting to grips nicely with everything now but before I begin my huge operation I need to clear up my confusion around volume normalisation.

Previously I have used a little application called MP3Gain which I now realise, after reading the help files here, only changes the tags and not the actual audio. It now makes sense why all my normalised files still appear to be different volumes depending on which player/hardware I use! :o

I would have thought that if converting to a lossless format it would make sense to change the audio to however you want it because then at least you know everything will always play the same, especially if you then wanted to convert to a lossy format for mp3 players etc? Is there any real disadvantage in doing this given that alot of hardware players seem to ignore the RG tags?

In MP3Gain I set the target volume to 92db. However even after reading the instructions for the volume normalise DSP, I can't quite understand what to do with the settings to do the same thing. And I admit to probably being stupid!

Thanks

[Spoon]

Re: RG Volume Normalisation vs Replay Gain

There is no real need to change the default target volume from the top value as it is adjusted internally to the recommended replaygain standard value.

[a.wright]

Re: RG Volume Normalisation vs Replay Gain

Ok thanks, keeps things simple!

[xoas]

Re: RG Volume Normalisation vs Replay Gain

I would like to add a few comments with respect to this statement:

Previously I have used a little application called MP3Gain which I now realise, after reading the help files here, only changes the tags and not the actual audio. It now makes sense why all my normalised files still appear to be different volumes depending on which player/hardware I use!

Unless Mp3Gain has changed, I believe this statement to be mistaken. Mp3Gain, in my limited experience with it DOES change the actual audio. It also stores information in the tag so that the adjustments applied by Mp3Gain can be undone. This makes it different from ReplayGain (which leaves the original audio intact and carries the information in the tag to allow the audio to be adjusted when the file is played by a player that can handle ReplayGain) and from the dMC Volume Normalization DSP (which changes the actual audio and cannot be undone.

As to why Mp3Gain did not appear to change the files you applied it to probably has more to do with the settings you were using.

As for the disadvantage to using Volume Normalization when ripping, the chief one is that when you do this you are changing the dynamic range of the audio from your source so that your copy is, to that extent, not a true copy of the original.

Personally, part of the reason I use dAP (dBpowerAmp Audio Player) is to make use of its Auto-Boost feature which (after the track is played from start to finish one time) will automatically apply normalization to the track when it is played. I use Volume Normalization primarily when I am burning tracks to a disk (especially an mp3 disk) or copying to my portable mp3 player. With dMC r12+ I find that using the Volume Normalization ReplayGain DSP (which changes the audio, whereas the ReplayGain DSP changes only the tag information but will only adjust the normalization on playback if you are using a player that can read the information) works better for me (at default setting) than the other settings.

Hope I have not made this more confusing but I did want to clear up the inaccuracy re: Mp3Gain.

Best wishes,
Bill

[a.wright]

Re: RG Volume Normalisation vs Replay Gain

Oh right.... well that's very interesting!

The "techie" info with MP3Gain says:

Here's the technical reason why it's lossless, and also why the smallest change possible is 1.5 dB:

The mp3 format stores the sound information in small chunks called "frames". Each frame represents a fraction of a second of sound. In each frame there is a "global gain" field. This field is an 8-bit integer (so its value can be a whole number from 0 to 255).

When an mp3 player decodes the sound in the frame, it uses the global gain field to multiply the decoded sound samples by 2(gain / 4).

So if you add 1 to this gain field in all the frames in the mp3, you effectively multiply the amplitude of the whole file by 2(1/4) = 119% = +1.5 dB.

Likewise, if you subtract 1 from the global gain, you multiply the amplitude by 2(-1/4) = 84% = -1.5 dB.

To me this implies that the audio itself is not being changed, just a field which tells the decoder what to play. However I am not any kind of expert on this so can't really argue it either way!

The RG in the Volume Normalisation DSP seems to work very nicely but I can see the point about changing the dynamic of the original sound. However I have to ask myself, does that really matter to me? It often gets played through an AV amp that adds some kind of effect to it anyway :teufel8:

[bhoar]

Re: RG Volume Normalisation vs Replay Gain

I would argue that MP3Gain isn't lossless, as the *original* gain/multiplier in the field is lost forever when it is replaced by the MP3Gain multiplier. Unless they store the original gain values in a metadata tag so that it can be reversed, then it is lossy, even if the other contents of the frame are untouched.

-brendan

[xoas]

Re: RG Volume Normalisation vs Replay Gain

Brendan-
With respect to your statement:

Unless they store the original gain values in a metadata tag so that it can be reversed, then it is lossy, even if the other contents of the frame are untouched.

I believe that Mp3Gain does exactly this. The Help file I have (my version is not the latest, I have r 1.2.4 and latest non-beta version is r 1.2.5) says:

MP3Gain can store "Analysis" and "Undo" information in special tags inside the mp3 file itself. If tag writing is turned on, then MP3Gain only has to spend time analyzing an mp3 file once. If you load the same file into MP3Gain at a later time, then the program will see the tag and instantly display the volume and recommended gain changes for the mp3.
With the tag information, MP3Gain is also able to automatically undo any changes you make to an mp3 file. All you have to do is load the mp3 file into MP3Gain and select "Modify Gain - Undo Gain changes" from the menu.

Andrew-
The text you cite refers to Mp3Gain adjusting each frame of audio information within an audio file and how it adjusts only that portion of the audio file that has the gain information for each frame. Still, it is changing the audio portion of the file when it does this. This allows it to be read by any player that reads mp3 files. It also stores tag information to allow the changes made by Mp3Gain to a file to be undone and to store the analysis data made by Mp3Gain of the file. This tag information is only meant to be read by Mp3Gain, so it doesn't matter if other programs or players cannot read the data.

I do agree that the issue about normalization changing the dynamic range may not necessarily make a difference for you and that the convenience of using normalization on ripping may be a better option for you. I only tossed that bit in because you did ask why it would make a difference (even though the question may have been intended to be rhetorical in nature).

Best wishes,
Bill

[a.wright]

Re: RG Volume Normalisation vs Replay Gain

This makes a lot more sense now! Thanks for the very useful comments.

[bhoar]

Re: RG Volume Normalisation vs Replay Gain

Ah, ok. It's lossless *if* you store the right tags (is that optional?) and ensure that you know the procedure to undo them. That makes sense.

-brendan

[xoas]

Re: RG Volume Normalisation vs Replay Gain

Yes, you can turn the option for tagging the analysis and undo data on or off.