Do you guys normalize all your tracks before you load them into your libraries?
I’m finding it quite time consuming, and I don’t know if it’s really worth it. I’ve been using Audacity to do it. Songs don’t seem to vary all too much, and it could just be fixed with a slight turn of the gain knob.
i’ve used this one for years. i use 94db (ean suggests 96)… if you run the whole collection through just analyze, and choose the “track gain” option. you will still need to tweak the gains but no where near as much.
Unfortunately my library is much too large and has tons of songs from other genres (that I will never play. I love me some Mumford and Sons but that’s not exactly a crowd pumper ). Gonna have to let my computer analyze the tracks overnight and then adjust the gains tomorrow morning!
How would this work with tracks that Traktor has already analyzed and set the autogain too? Would I need to delete them from my collection and reimport?
MP3Gain throws a hissy fit over clipping if even one tiny little bit of anything clips, even if its only a millisecond, and there’s no attack/decay/tolerance setting for it (which is really annoying). So I find the best way is to ignore it, then if you notice a particular track distorts badly, take care of it on a case by case basis.
You’ll mainly see this problem with Hip Hop or RnB tracks, which are fairly quiet overall but have huge kick and bass spikes every few beats.
It uses something called RMS to determine how loud a track actually sounds to the human hear, rather than just peak normalisation, to set all your MP3s to a certain level, specified by you. Most people go by about 96dB on the program setting.
If you go to the website, (mp3gain on sourceforge, google it), it explains it better.