Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[MusicHawk]

I've read many threads here about whether its better to Audio Analyze vs. Normalize, but I'm not sure the matter is settled.

Audio Analyze is a volume analysis that sets a parameter in the music file that can be used by playback software to adjust the volume at play-time.

Normalize is a change to the sound level in the file itself, so it "sticks" no matter how/where the music is played.

Normalize seems more desirable because it is a universal adjustment of the file's volume, no matter what the playback situation. But in MC forums I find cautions that Normalize reduces the quality of the recording.

The issues I've found:

+ When played back via a system that doesn't recognize the "adjust the level" parameter (Netgear MC101 has some trouble with this), the Audio Analyze values don't seem to help every song.

+ When processed via several audio editing/library systems (Sony Sound Forge and others), Normalize is the prominent way to standardize volume levels.

+ Sometimes, Audio Analyze doesn't seem to get a firm grip on a recording, or perhaps the analysis is thrown off by some recording characteristic, so even played through MC or an iPod, some songs can be too soft or loud compared with almost everything else.

Given that MP3 is already a compromise in quality, what's the real impact of Normalizing it too?

Thoughts?

[Higginz]

Didn't have much to add to this topic, besides that I'm a heavy ReplayGain user! .
I love it when playing random songs from entire library. (Using track gain)
I usually disable ReplayGain when playing whole albums. Then it has no perpose.

Higginz

[GHammer]

If you are talking about normalizing at ripping time, you are changing the WAV and then processing it to MP3/whatever. To me, you are starting with a lesser quality source then compressing it to MP3 and losing more quality.

If you are talking about using MP3Gain to write an 'adjustment' value to the MP3 after encoding, the sonic quality of the music is as good as its going to be and now you are simply telling players that this info should have this volume adjustment applied to it.

Better to me.

I always run MP3Gain on newly encoded MP3s. Then no matter the player I get good relative volume.

[benn600]

I remember strictly being told to not use the audio normalizer on the original source material if you are worried about quality and preserving the best original quality.  Are the playback normalize features in DSP studio?

[benn600]

Hey, Higginz, I really like your signature picture!

[Higginz]

Hi benn600, Thanks. 

I haven't used the Normalize feature. I just let MC perform an Audio analyze to my audio files.
Then I activate ReplayGain.
Audio analyzing can be useful not only for the ReplayGain thing. But you could also use it on these smart playlists (smartlists) to only play fast songs, slow songs, or low intensity songs (ballads, classical music etc.)
The audio analyze finds that information too.

I wouldn't use any of the volume levelling gadgets when burning albums.
I would however use it if I would to burn my own compilation CD's ("mixed-CD").

[benn600]

Yes, when outputting media to listen to, it's not nearly as big of an issue if you modify the audio.  I often listen to music with effects turned on.  But at least I didn't add Arena x10 from DSP studio to the original FLAC files!  The original source, as horrible as it may be, is nice to store the way it is.

[jgreen]

hawk--

Your reasoning is sound (sorry), but i would caution against normalizing archival data--this is an alteration that cannot be undone.  However, since ease of use and consistency across platforms are your primary goals, go ahead and shave some sound off your music!

If I was to do this, I would fool a bit with the handheld cache feature--see if I couldn't just create normalized mp3 copies of all my lossless.  I'm not sure if it would prove to be as easy to use as the main library, however.