When Is It Too Loud?

[Mark Kaufman]

When recording digitally, 0db is the max. Any recording that never even hits 0db is probably much too quiet in comparison to most tracks. But if you maximize everything up to 0db, you get a square wave and free enlistment to the Loudness Wars.My question is, how do you decide when the majority of your piece is loud enough? If you use Maximization, how do you determine how much to crank it? Your ears? The shape of your soundwave? Do the pro studios have any guidelines?I'd love to hear how y'all decide.

[andreh]

Hi Mark-Unless your tracks are destined for commercial CD release or radio play, you don't need to worry about achieving ultimate loudness or matching the levels of other commercial releases. You could just apply a limiter to bring your tracks up to a reasonable level but with no audible limiting artifacts being introduced (perhaps 1-2 dB's of gain reduction at the loudest parts) or even simply normalize the tracks before sending them off to your client.If the tracks are for commercial CD release, then you have to decide whether you want to match commercial levels or preserve the integrity of your mixes by mastering at a quieter level. In either case, you'll likely achieve a better result if you send the tracks out to be mastered by a pro, in which case you won't need to worry about how to get things super loud.If the tracks are for library use, check first to see if the library wants unmastered tracks, which would mean no limiting at all. Some libraries like to master tracks on their own.Tracks destined for TV or film use don't need to be volume-maximized because for one, the music is usually played at a much lower volume level than the dialog/FX...but even when the music is featured and loud, there's 10-20dB of headroom (above normal dialog levels) left in film/TV productions so tracks never need to be played at normal radio or CD levels (speaking in digital terms).One caveat to the above is that if your tracks are not yet approved by a client and are being compared to other tracks, a music sup. might be swayed by the loudness of one track over another (though an experienced one likely won't) - so consider this when making decisions about loudness levels.André

[mazz]

For my TV gig, they specify that we leave 2 or 3 db of headroom. They may want to do some EQ or other types of processing and this way they don't need to do an extra step to reduce the gain so they can EQ without causing overloads.Also remember that peak level is different from average level. What a lot of the loudness wars is about is getting the most average level in the mix. That's why those waveforms look like square waves, because the music in between the peaks has become as loud as the peaks. That's why a lot of music these days sounds and feels (to me) like someone is screaming at you constantly (and I don't mean specifically vocals), because there's no difference between the peaks and what would naturally be softer parts. Music these days achieves dynamics by changes of texture rather than by changes in volume. The thing is, the thinner textures are often made to sound as loud as the much thicker textures so you don't get a volume change, you get a texture change. But the net effect many times, at least to my ears, is that the "louder" parts actually sound smaller and more crushed and diminished, which is the opposite of what the intention is. The loudness wars started because of labels wanting their song to sound louder on the radio than any other song and also to overcome the effects of severe ( and often misadjusted) radio compressors. Analog radio stations have very strict rules against overmodulation because it causes interference in to other radio bands. So raising the average level of a record gave the illusion that it was louder because the difference between the peaks and the valleys was much smaller than it would normally be.Sorry to blather on there but I feel strongly about this whole thing. Mazz

[Mark Kaufman]

Thanks Andreh and Mazz.And so my question remains...although I will modify it to address the average level. How do you decide?Is there a standard average level? Or does it all come down to "feel"?

[billg]

Mark, I had these same questions a few years ago when I was forced to be my own "do it all" guy. There are two things that helped me a lot. First, I sent one song to a pro mastering house to be mastered. This was a person who mastered for all of the Warner Bros. "Indie" imprints and he had mastered stuff for me (when other people were picking up the tab!). The song was one I was very familiar with. I ask him to master it for radio & ask him if he would provide notes on what he specifically did. I listened to his master & read his notes and tried to emulate. Of course I didn't have the great mastering specific gear he had but I could get in the ballpark enough to grasp the whole thing. I compared the waveforms also. This may not be useful or make sense to some folks but for me being able to see how "dense" the compression was and to see the limiting really helped, you can also see exactly where audio level is. The next thing I did was to load up reference tracks of songs I liked into my work station. I keep them there and go back & forth sometimes when I'm doing my poor man's mastering. Giving your song an A/B comparison can also get you closer to eq etc. when your mixing (but don't make the mistake of trying to compare the volume level of your mix to a commercial mastered song!)Most of the commercial stuff I like even though it's squashed still bounces between -3db to just under 0.I guess all I had to say was load up a song you like and try to match it!Mazz is right though, it's gotten insane how squashed music is these days.

[southpaw]

I dont know how anyone could say there is a standard level... I suppose someone with much more experience and and time in this business than myself, could have analyzed the trends and gathered methods from a wide range of people and over time come to draw a conclusion from all the experience and information... I am curious to hear what folks say in regards to your question. Anyhow, all i can say is: for audio, my ears always trump anything else, and i can't say ive ever had two projects that were exactly the same.. I've had projects that sounded right at +0.3.. In contrast, i ve had projects that sounded right at -0.7, which ironically sounded louder to me than the aforementioned. Interesting Q.Jamie

[mazz]

There is a standard level for film mixing but as far as I know, there's none for music mixing.

[Mark Kaufman]

Thanks Bill and Southpaw!Bill, comparing waveforms makes a lot of sense to me...that's probably what I'll do. And Southpaw, I suspect you're right about no particular standard existing. I just find myself always facing the same decision upon final mixdown...should I nudge it up a little more? Should I back it down? Is there an Easy Button?...

[Mark Kaufman]

Ah, thanks again, Mazz.We're on our own!

[andreh]

Jan 11, 2009, 4:34pm, lyle wrote:Bill, comparing waveforms makes a lot of sense to me...that's probably what I'll do.Not trying to butt heads with billg whose music and opinions I respect, but to me this method is the least useful of the ones mentioned and can in fact lead you astray. A waveform can perhaps show you how much compression and limiting is going on...but not each of those separately, and the arrangement of the music in the track will also affect how things sound vs. how they look. Perhaps looking at waveforms is good to see if you're WAY out of whack (which you'll hear anyway), but bottom line, use your ears.André