Mastering CD: Getting same volume for every track

  • Thread starter Thread starter OnTheBlackRock
  • Start date Start date
O

OnTheBlackRock

New member
OK when mastering a track when you have everything mixed accordingly to your liking (or the bands, performers liking) how do you put all the tracks onto cd and have the same volume for every track? is there an outboard unit you can get to do this? or a plug in for protools I could get? or is there a mastering hardware i should invest in... please help me.
 
Firstly, you don't want to have every track's volume the same. Then your ballads are as loud as your rockers and your rockers aren't quite so rockin' anymore. You want them all to be at an appropriate volume in relation to each other. As far as how to do that goes, there are lots of guys here who can explain that better than I, so I will let them have at it.
 
are you sure dude? all the CD's I own and bought in stores seem to have the same volume on every track... I've never heard of a CD where you had to turn the volume up or down every sinlge time you change songs on the disk... that could also cause car wrecks if you had to turn up/down the volume every time a different song came on...
 
I'm not talking about as drastic a change as that, I'm just saying you don't want the average volume of a ballad to be as hot as a rocker, or you lose all the power of the song. Do some reading on the subject and you'll see what I'm talking about.
 
There is no "magic" shortcut. If you don't have exspensive mastering equipment/software it takes patience. I've learned to keep my mixes about the same level by carefully watching my meters and knowing how loud a mix should sound with my mixer volume at different positions. You may be able to use a reference CD to help. Plug a CD player into the mixer and find a song to use for reference. Adjust the CD input gain to match the loudness of one of the songs, then use this to gauge the loudness of other mixes.
 
Bring the output fader of your mix buss on each song up and set it where its just missing clipping, real super close, and start the song over and watch the whole thing. If something clips, bring it down a tad or compress whatever caused the clip or mildy compress the mix buss and listen and watch again.
When you have the levels as hot as you can get it, for that particular mix, without clipping the output, move to the others and do the same.
Once they are done, because of mix variances and song types and whatever, they will seem different volumes.
Now, you need a sperate program like wavelab to export the songs in to.
Now, bring up a good limiter, like Sonic Timeworks Mastering Compressor (my fav) or Waves L1 (or L2 if you are real lucky) and bring the threshold down until the song gets a little louder and punchier, and mess with it tills its right.
Now, unless you are wanting to add an overall EQ to the song (which you would do at this very point, in FRONT of the limiter) you leave all these effects up and process each song through the same setup, and each song will be punchier and closer in level and more flowing as a CD, volumewise. Try not to limit in more than about 6 or 7db... this will bring you real close to standard radio volume.
I noticed that not doing this results in songs that arent anywhere near as loud on the radio by comparison. Its really the only way to compete for volume.
 
I guess if you dont have wavelab or anything, you could set up a mastering compressor in one song, save the settings, and bring the same one up on the other songs so they are all getting the same haircut, but I dont like that idea for some reason.
Dont forget to dither very last after everything else. Apogee HR seems very nice to me, comes with wavelab4.
 
tubedude said:

Now, bring up a good limiter, like Sonic Timeworks Mastering Compressor (my fav) or Waves L1 (or L2 if you are real lucky) and bring the threshold down until the song gets a little louder and punchier, and mess with it tills its right.
Now, unless you are wanting to add an overall EQ to the song (which you would do at this very point, in FRONT of the limiter) you leave all these effects up and process each song through the same setup, and each song will be punchier and closer in level and more flowing as a CD, volumewise. Try not to limit in more than about 6 or 7db... this will bring you real close to standard radio volume.
Click to expand...

I would have to disagree. Your "loudness" is going to have to come mostly from single or multiband compression. Maybe a seires of compressors if you know how to set the attacks and releases. All the L-1 or L-2 (which I love but rarely go past -3db) is going to to is cut off your spikes which sooner or later leads to distortion. Sure it makes your song louder, but without any distortion? I know of one mastering engineer who says he would never use the L-2 for more than -2db. Again, perceived loudness should come more from compression and not so much from a limiter.
 
Good luck getting that radio level then.
Fact is, you're really not going to get the overall volume you want any other way.
I can hit about 9db on the Sonic Timeworks one before I can audibly hear it get ugly, therefore I keep it under 6, and it works great great great. Since I am at maximum level with my mix without clipping the meters anywhere including the output, and considering everything is compressed, EQ'd and processed the way I want it, there really isnt much else to do other than gain a few dbs via L1, L2 or Mastering Compressor. And, it works like a charm. Loud and punchy. Amen.
 
Tubedude:
How can a limiter by its self make something more "punchy"? I know absolutely nothing about Sonic Timeworks. Maybe its not just a limiter? Is there compression also going on? Isnt puchiness acheieved by varying the attack times while compressing?

The fact is that I can use either the C-4 or LinMB even without a limiter, and get my song between average -13 to -11 RMS. Your telling me that isn't loud? Then with a L-2 I can bring it right down to -12 to -10dbRMS and it sounds sweet. Most commercial pop songs I check are right between average -12 to -10dbRMS ready for the radio.

Loudness does not come from "peak limiting" because peak limiting does not compress the dynamic range of the music which is where we get our perception of musics actual loudness. Yes, its bringing everything up to be louder but choping off the peaks.

I too, used to crank my lL-1 harder until a mastering engineer sat down with me and pointed out areas of distortioin that I couldnt hear the first time around. He also pointed out artifacts I didnt like from the heavy limiting. And this guy makes some of the loudest stuff around by chaining compressors in series. If you want real loudness....without distortion...compression is the way to achieve it.

Look at any commercial song. Are there hundreds of peaks cut off that makes it look like it was run through a limiter? Or is the body of the wave nice and thick with few cut off peaks? That is one way to tell an amature recording from a commercial release. Both waves sound just as loud, but one sounds definitley more clear and defined. Of course it will vary depending on the style of music.

Try taking 4 compressors and chaining them in seiries. With the first compressor make the release time long. With each succeding compressor make the release time a bit shorter. Eventually the sound becomes so thickened that it doesnt sound like a compressor was even used on the material. Each compressor should do around 1.5db of gain reductioni so by the end of it, you have an increase of 6 db to the music before even getting to the L-2! And by the time the L-2 is done with it there are no sharp transients left in the music.

The trick to using compressors in series is to figure out the settings for your release attack times. Each song requires different times due to different tempos and beats. The only way I know how to figure this out is to set an EXTREME ratio of 10 to 1 and bring the threshold way down to -40db. Now you have extreme squashing.

Next start making the release longer and you will immediately hear the disotortion disappear as you make it longer. Things usually sound good for me starting around 400-500ms. Just leave that release point where the music sounds the smoothest and where the release compliments the beat/tempo of the song. That would enhance the beat by slightly pumping it.

Next start tweaking the attack until you get the desired amount of attack that sounds right for the music. Try to find a spot where again there is less distortion and smoother results.

You have just managed to squeeze the hell out of your music without any distortion artifacts so you can now bring that ratio back to a sensible 2:1. And bring the threshold back up until around 2db of gain reduction. The attack release settings you have just made are compatible with high ratios and deep thresholds so you definitely know that no distortion is going on.

Next repeat the above on the 2nd compressor in the chain and again start with exaggerated settings to hear any distortion. This time you will be able to get away with a faster release time and even faster on the 3rd and 4th compressors. Since you are using 4 compressors I dont aim for more than 2db of gain reduction on any single compressor in the mastering stage.

Try multiple compressors. Certainly it doesnt work for all situations but I have had much success with some songs.
 
Very good advice greggybud!

I don't no that it's essential that you keep lowering the release times with each successive compressor however. For example, some compressors like the Weiss DS1 actually has multiple release times built into one compressor, e.g. release fast, average, and slow.

I would say that each succesive compressor gears itself more toward peak limiting, but that's just my taste I don't know that it's a rule of thumb. Every compressor has it's own "character". Just use it them for what they are good for. For example an L2 is not really what I would use for "average" leveling but is very good at peak reduction. Likewise the Weiss is better at increasing the average level, but tougher to use as a peak limiter than an L2. There's also multi-band compression which is used as a kind of compressor-based EQ, and should preceed either so that the averaging compressor "sees" a balance frequency response instead of lowering the overall level due to a kick or some other frequency being too loud.

As far as levels, it really helps to have good metering or metering software for example SpectraFoo or Waves PAZ.

My general rule for balancing the levels (and other adjustments) is that if I'm listening to a CD in my car, and I reach for the volume or "tone" knob, something's wrong.
 
Weis DS1 is out of my league and I can only dream. My main point is that loudness needs to come from compression and not peak limiting unless you want nasty artifacts.

Once you feed a good thick loud compressed wave into an L-2 then the L-2 can do its magic with superb results.

Cranking up the L-2 just to achieve loudness, and calling it a day, is not mastering.
 
You must log in or register to reply here.
Back
Top