What's up with the arbitrary levels in MP3 encoders?

Kelly Holdridge

Kelly Holdridge

New member
Anybody notice this? You d/l a song, run it through a player (WinAmp, here), and it's either off the meter completely or doesn't show up at all? I'm not talking about the WAV issue of WinAmp, but rather the somewhat "arbitrary" level that encoders seem to use (or the arbitrary level folks use creating the WAVs the encoder processes).

I'm ALWAYS having to adjust the gain on WinAmp's "preamp." Any thoughts?

Mine: Different Normalization levels (why not 100%?). Different encoders. uhm... The MP3 fairy.

I dunno. Just ranting, I guess. (does anyone know if WinAmp's EQ is "off?")
 
There's a button on the applet window that's lit when the EQ is ON and not lit when the EQ is OFF.
Same with the "Pre-amp" button that controls the gain/cut slider to the left of the EQ section.

Not sure I understand completely what the problem is that you're talking about, but when in doubt I load the .mp3 in Vegas and watch the meters there.
 
Well, I guess it's more of a gripe... I just figured everyone was normalizing their stuff to 100% (well, 99% is better, folks), but there's a TREMENDOUS difference between songs picked at random from our very own MP3 Mixing Clinic.

It just happened again today... I'm starting to wonder if sharing this computer up here at work has anything to do with it... (i SWEAR this happens at home, too, so I doubt this explains it...)

A CD or tape is continuous in regard to the person who created it, whereas MP3 players mix and match differently produced music... the continuity is TOTALLY fucked with the levels all over the place, and I'm starting to think it's because of the different encoders being used...

btw, Doc, THANK YOU. "Ellis" (LS for Little Shit) says thank you, too! (i was pretty freaked at first, like, WHOA!) :D
 
I don't think what you are thinking is really the case Kelly.

I can send you 10 different mp3's all encoded on the same encoder, and there is a pretty good possibility that all 10 will have at least slight volume differences unless I mastered all 10 of them to match in volume.

I don't "normalize" my mp3's, usually just use the "mastered" version of the mix.

I think what you are hearing is the varying skills of the people doing the production. Some can get pretty hot levels, other can't. This will be reflected when you start playing random mp3's from the forum.

Yes, I notice it too, but it is usually NOT the encoder. I can still hear crappy encoders, but usually crappy encoders don't lower the RMS level, they just mess up the frequency response really bad.

Ed
 
Ok, you're both right, guys, but I'm not really talking about that...

Maybe this should be my question: Why would you need to turn the gain DOWN on WinAmp if a full-blown blast...

Hold on. Ok. You never get above 0 dB digitally, right? So how can an encoder make signals that would need attenuation in the MP3 players? That's HALF of my question...

The other part, the "difference in encoders" part, comes from the faith that most people should know to normalize their master, right? So why, then, would we be getting different final levels across so many MP3s here in the Clinic?

(and Doc, I see the on/off thing, didn't mean to ignore you there, my language just kinda drifted towards this conundrum...)
 
Ok, this is still buggin' me. If you can't get above 0dB digitally (which is what all the shit in the computer is), then how is it I have to keep turning down the fucking gain on WinAmp?

I'm starting to think that the EQ display is not "big" enough, that I think it's topping out when it's not (but dammit, I hear the distortion...).
 
You must log in or register to reply here.
Back
Top