Why is the volume on my final cd masters so low?

[clonewar]

I'm pretty much a newbie to mixing and mastering. I'm using Sonar 2 and Soundforge to record and master some rock music. The mixes sound good but when I burn them to cd the volume of the songs are very low compared to 'store bought' cd's.

I've been mixing the songs down to .wav files and using Nero to burn them. I've tried to push the volume of the mixes, keeping them around 0 db, but they still come out very low on the final cd's.

Is there anything I'm doing wrong or anything else I should be doing in my mastering process to get the final cd tracks louder?

I'm sorry if this has been covered before, I've searched on the subject but haven't found any clear answers.

Thanks in advance for any help!

Michael N

[wogg]

Even though your peak levels are at 0dBFS on your outputs, the average level of the tracks is going to determine how 'loud' the CD is.

The way to boost the average level is through compresion and peak limiting on the final mix. I'm not sure what software version you're running or what plug-ins you have at your disposal so specifics are a little difficult.

I used to do pseudo mastering using Soundforge 4.5 only, so I'll explain how I did that.

First a quick tools->statistics run will tell you how loud your average level is. For reference you can rip one of your store bought CD's to a wave file and check it's statistics too. I usually shoot for about -13dB average level on my tracks, a little bit quieter than some CD's but not annoyingly so. I've found that any hotter than that starts to sound pretty bad. The tracks end up with no dynamics and sound fatiguing. There's a reason mastering houses use $10,000 pieces of outboard gear to get the hotter than hell level they produce.

I used the graphic dynamics processor in SF4.5 to boost average levels. To do that, draw a smooth knee gain reduction curve on the graph and enable automatic gain compensation. Run the compression, check the statistics, and give it a listen for any nasty artifacts. Undo and modify the compression if not satisfied (compressing again will start to kill your quality).

That was my very ghetto way of doing things when that's all I had. I still do things a little ghetto but now I use a waves R-Comp on the master bus in Sonar to get the average level up without having to re-process the file in 16 bit land after the mix-down. The Cakewalk compressors simply could not get the output hot enough so I wouldn't really try that method without some better EFX.

Keep in mind though that the productions nowadays are in many cases over compressed and peak limited to the point of severe clipping on the dynamics. Good sound should rule over hot levels. Hopefully the industry will figure that out again soon.

[the_lost]

Originally posted by wogg:
Keep in mind though that the productions nowadays are in many cases over compressed and peak limited to the point of severe clipping on the dynamics. Good sound should rule over hot levels. Hopefully the industry will figure that out again soon.

In this day and age?.... I doubt it.

[tripnek]

The problem lies with the damn Radio stations. The louder the CD the more distance they can transmit. The industry has kept boosting the volume more and more to accomidate. The last CD I played on myself was sent to the Masterdisk mastering house(NOT Masterdisc the duplication service) in New York. These guys are some of the best in the biz, but I still think the Raw mix sounds much better. It's loud as hell, but the dynamics are gone. Unfortunately, this trend is not likely to change until the radio stations all start using satilite or some other technology to transmit..

[gregpeters]

I just happened upon this post and I admittedly have the same sort of problem as clonewar, my mixes just sound much too quiet. Now, from what I understood of your post wogg, the the higher the level of each individual track the louder the overall mix. Theoretically speaking, if all of my tracks are nice and "hot" then should the overall output level be that relevant?

Sorry if I sound like a real newbie, but I've really been struggling with this particular problem. Does anyone have any advice on how to do a basic mix (I'm using Logic) and get the best possible sound level. Any advice would be much appreciated.

[strunz0]

ahhhh.....the never ending story. I have personally sent many master mixes to mastering house to be radio ready for major label release. The problem is not with the mastering engineers...it is with radio. Radio has compression to make their station the loudest on the dial. So that qwhen you are flipping through your radio...and you come accross their station, it is loud and in your face. What a mastering engineer is trying to do when squashing your tracks,. is trying to musically get rid of peaks in your tracks. The radio stations compression suck, and if your track is not mastered proerly the radio stations compression will compress your track much more than if your track was already mastered. Get it?

[wogg]

Originally posted by gregpeters:
Now, from what I understood of your post wogg, the the higher the level of each individual track the louder the overall mix. Theoretically speaking, if all of my tracks are nice and "hot" then should the overall output level be that relevant?

Not quite what I was going for. I'm referring to processing the final mix as a stereo wave file. That's typically what you supply a mastering house, a stereo file of the highest resolution available.

It isn't the levels of the source tracks or the final mix that are the concern, it's the dynamics. Digital audio only goes so loud, a full scale number can't get any bigger. So if a short peak from a snare drum or something is much higher amplitude than the rest of the recording, the volume must be reduced to fit the peak under the full scale limit to avoid clipping. The only answer is to trim that peak out to bring the average level of everything else up. Hence the compression and peak limiting, a controlled way to trim those dynamic peaks without adversly affecting the sound quality of the source (if not overdone that is).

[clonewar]

Thanks for the responses!

I took one of my tracks to a friend's studio over the weekend and he processed it through the Waves L1 Ultramaximizer+ and that did the trick, I'm going to order that package, its just what I was looking for.

Basically, it does what wogg said he was doing manually in Sound Forge. It boosts the overall level of the mix while limiting at a user defined level. It shows you graphically whenever the mix goes over the limit. We were able to boost the level of my mix by about 7 db, which makes a huge difference.

I agree with what was said about pro mixes now being way too loud. Rip Rowan has a great article on prorec.com called Over The Limit that talks about it. I'm not looking to necessarily be as loud as everyone else, but my mixes were so low that you'd have to crank the volume all the way up and that would introduce a lot of background noise. I'm hoping that as I get more experience in mixing that I wont need to boost the final mixes so much, but for now I'm very happy with the result I got with the Waves plugin.

[TimZ]

Originally posted by clonewar:
I took one of my tracks to a friend's studio over the weekend and he processed it through the Waves L1 Ultramaximizer+ and that did the trick, I'm going to order that package, its just what I was looking for.

Don't get the L1. By today's standards it is not very good and like Soundcracker says can actually hurt a lot of mixes. Go to:
http://www.voxengo.com./

and get the Elephant VST. It is WAY better than the L1, and better than the L2 to a lot of ears, and best of all it is MUCH cheaper.

TZ

[trock]

Tim beat me to it

I 100% agree - the Elephant is amazingly transparent. It even beat the Sony Oxford Inflator on one mix where I needed to tame a drum section.

trock

[Melanhead]

Hmmm...I used the L2 for the most recent masters I've done... I'll have to download the demo and check out the Elephant...

I'd like to see what my ears have to say

[TimZ]

Bob, there's nothing wrong with the L2, but if you don't already have a mastering limiter, the Elephant VST is only $65 and definitely as good as the L2. Since you already have the L2, it may, or may not, be worth the extra expense. I plan to get the Voxengo Elephant, Soniformer & Lampthruster package soon. I do think it would be nice if Voxengo could work on their UI though, but I guess that's partly why they are cheaper. I'll take superior sound quality over UI anyday

TZ

[Melanhead]

Hey Tim...Long time no chat!

more curious than anything ... I don't own the L2 but was thinking of getting it. ( I mastered elsewhere...)The price point is way better for the Elephant so if it's comparable or better then it's less money for me to dish out!... and my wife and I like that!...

P.S... My wife and I had a baby girl on Sept 29th... I've been moving rather slo since

[TimZ]

Originally posted by Melanhead:
My wife and I had a baby girl on Sept 29th... I've been moving rather slo since

CONGRATS MAN!! I hear another tune coming on. Loved the last one about your daughter

Most of my energies have been going into the band doing live gigs lately, and less into the recording end of things. We are still (SLOWLY) working on the Freefall 3 CD. I'll get one to you when it is done. I think you might like a few of the newer tunes.

There is a demo download for the Elephant VST, so you can try it out and see what you think. I found that it was pretty similar to the L2 to my ears, but it was VERY easy to get the levels up with no noticeable artifacts.

Cheers

Tim

[Melanhead]

Thanks Tim!...I'll check out the demo. It'll be a while though 'cause I'm not doing much music lately... Just rebuilt the bathroom/ master bedroom and coverting part of my garage into a studio/get-away to hide from the 3 woman I have living in my house ...

Anxious to hear the new CD...

My wife tells me I have to write a song for the new one as well to be fair...Hmmmm... I guess I'd better continue to work on my vocal chops