Using FX to fix volume was something I was told I shouldn't do. FX (ie: Reverb, Delay, Chorus) = icing, it's not a substitute for turning down volume.

When you say -6dB, is it understood to be 0db (as in go past this point, you see red, meaning you're clipping)?

Again, you want headroom and noting before that you can't trust that the peaks will be constant (in the same spot every time).

As time-consuming as this is, you will have to adjust the tracks' volumes individually or use a volume send while leave the master at -.5 to -.3dB. Essentially avoid recording/mixing down/bouncing @ 0.0. 

Question: Do you have input monitoring on? If you're recording from hardware (guitar/keyboard/drum machine), this need to be on, so you can see how much of the signal is gong into the software. Also, if you are mixing with speakers or headphones, have this on so you can hear yourself.

If you have no such access, get a program such as Audacity which can recording incoming sounds from your computer. It has the Input Monitoring Function which means what it says. It'll monitor all audio coming through your computer speaker (Make sure Stereo Mix is selected).

In fact, this program will save you from mixing down/exporting and importing as you'll be able to see the waveform as it's being created.

Here's a turtorial: http://www.youtube.com/watch?v=QXOOHOzCHHY (voice)
Here's another: http://www.youtube.com/watch?v=21wh0iixiDA

Now, he's talking about recording voice, but in your case, the incoming audio will come from whatever DAW you're using. The second video is about using it as you would your DAW.

Here's another video entitled the "4dB Rule": http://www.youtube.com/watch?v=XlY_Yxe0IVE (This is subjective). And he's referring to the individual tracks, but like I said before, the master needn't be touched once you've set it to where ever you desire.

Turning down the individual tracks turns down the Master Track volume. (ie: every track turned down -4dB, the Master goes down by the same amount) Anything done to the Master Track, should be done last.

"Using FX to fix volume was something I was told I shouldn't do. FX (ie: Reverb, Delay, Chorus) = icing, it's not a substitute for turning down volume. "
 
No I don't use fx to fix the volume. I record peaking at -6db looking at the track meter. Add fx keeping the same level at -6db. Then I use the volume fader to bring the track meter to 0db more or less to do a mix.
 
"When you say -6dB, is it understood to be 0db (as in go past this point, you see red, meaning you're clipping)?"
 
When my track meter shows it peaking at -6db it does not reach the red zone. I believe it only reaches the red zone at 0+db.
 
"Question: Do you have input monitoring on? If you're recording from hardware (guitar/keyboard/drum machine), this need to be on, so you can see how much of the signal is gong into the software. Also, if you are mixing with speakers or headphones, have this on so you can hear yourself."
 
I do turn this on when recording. But is it necessary to turn this on just to see how much signal is going into the software? Does the track meter show this without it on?

^ That's fine. I tend to creep the volume up as well.

@ -6dB you still have headroom, so that is good.

No. Input Monitoring must be enabled for the meter (Recording - not Playback) to work. I would say so because though you can turn it down in the software, why not just turn it down at the source? You can always turn it up later in the software. However, if you record it to where it's clipping (and it's not just volume that causes this, but EQ, Compression/Limiting), then it will still clip as that is how it came in from the source had any of those effects been applied.

Get the sound and levels right from the hardware to go into the software than make adjustments within the software.

It's very subjective when trying to attach numeric values because there will be a gazillion different opinions. Also, keep in mind that no boost or reduction may be needed and as the guy in the "4dB Rule" video said: "Leave the mix to rest your mind/ears and come back with fresh ones."

Rarely are mixes/recording right the first time around, so just experiement, but remember, headroom, headroom, headroom!

sharpdion23


Though how come the level of commercial music CD's are much louder than the ones I burn? Where should my tracks & Master level be at?

This is due to what is done to the processing ...aka ...Mastering.   With limiters and compressors it is possible to have acoustic music extremely loud on a CD. 


This is a topic which you should spend some time reading about and studying both sides of the argument as to why they do what they do. I think the pendulum is swinging back slightly in the direction of more dynamics vs make this sucker loud. 


It was the old bragging rights thing..... our CD is louder than yours.....kinda like TV commercials a few years back and sometimes today are louder than the program. 


I prefer to come down on the side of keep dynamics in the music but with that stipulated, I still want to have the levels up to a respectable level. 


Many folks might disagree with how I do things but here's my mixing in a nut shell. I mix to have the song sound right and when I look at the wave after export I don't want to see to many "overs" in it..... peaks slamming into the ceiling. I run the master and sound card levels up but make sure none are in the red.  I often have an exported wave that doesn't hit the ceiling at all. So I will use 98% normalization on it. That raises the highest peak up to just under the ceiling.... to 98% of full and that has the effect of bumping the final volume slightly. 


This picture illustrates what I look for in an exported wave...... nice consistent peaks but nothing anywhere near the ceiling.  I have a little bit of wiggle room if I decide to normalize this wave and still keep it off the ceiling. The dynamic range of the song is also more consistent and doesn't vary by too much. Most of the energy is lower down in this song.... unfortunately, I do not have this song posted on line.  

If this song is run through a 98% normalization process, you would hear a noticeable bump in the volume, probably by only a few db. Nothing extreme or unpleasant when compared to other songs.  




.


The second example has more dynamics than the first one.... and there are a few places where it hits the ceiling with overs but it still works and sounds fine.....  http://soundclick.com/share.cfm?id=11386873








Hopefully these pics help you to see what I'm speaking about. 

@Rus W

"@ -6dB you still have headroom, so that is good."

What would be considered no headroom? 

"No. Input Monitoring must be enabled for the meter (Recording - not Playback) to work." 


Not sure if I understood you correctly, but this is what appears when I don't have the input monitoring on. It still shows the recording level showing on the track meter.

"It's very subjective when trying to attach numeric values because there will be a gazillion different opinions. "

I get that, but I'm just getting a general rule of thumb of what to do and what not to do.
--------------------------------------------------------------------------------------------------------------------------------------------
@GH

"This is due to what is done to the processing ...aka ...Mastering.   With limiters and compressors it is possible to have acoustic music extremely loud on a CD. " 

Would you know what db most commercial CD's peak at?  "This is a topic which you should spend some time reading about and studying both sides of the argument as to why they do what they do. I think the pendulum is swinging back slightly in the direction of more dynamics vs make this sucker loud. " 
So are most commercial CD's fighting the loudness war instead of music dynamics? What do they do? Do they heavily compress the music and raise the overall?
"I prefer to come down on the side of keep dynamics in the music but with that stipulated, I still want to have the levels up to a respectable level. " 

When you finish up a project, where does the  master meter peak at ready to burn on to CD? "Many folks might disagree with how I do things but here's my mixing in a nut shell. I mix to have the song sound right and when I look at the wave after export I don't want to see to many "overs" in it..... peaks slamming into the ceiling. I run the master and sound card levels up but make sure none are in the red.  I often have an exported wave that doesn't hit the ceiling at all. So I will use 98% normalization on it. That raises the highest peak up to just under the ceiling.... to 98% of full and that has the effect of bumping the final volume slightly."


I've been reading about normalization but can't get a clear understanding of what it does. Is it part of the mastering stage? From what I read, normalization is bringing the average and peak volume to a target level.  you said if your project is hit's the ceiling you would normalize it to 98%. Question is, how would that bump the final volume? Doesn't it lower the volume to 98%? 
------------------------------------------------------------------------------------------------------------------------------------------------

Lots of questions there..... 

I don't pay attention to commercial CD levels, I simply use the process described to polish the songs before calling them done.

Some are refusing to participate in the loudness wars.... some are not. That's up to them.

When I finish a CD in Sonar the master meter is out of the red. I don't worry about the peaks too much. Like I said, some folks may not agree with me doing it that way... maybe it's like driving without looking at the speedometer..... IDK.  I look at it's "picture" in my wave editor.

Normalization is a process that raises the levels while keeping the same perspective ratio between high and low points...... but only allows the levels to be raised until the highest peak hits the predetermined maximum point. Technically, unless you have compressed it to the point where there are massive numbers of clipped peaks..... you would only have one peak in a song that hits that point...everything else will fall somewhere below that level. In the second picture above... only about 7 peaks are close to the max level in the entire song after it was normalized. 

I think I'm going to keep my master bus meter below 0db. At the most 1db. Does that sound reasonable?
I've imported CD songs into cakewalk and the highest I read so far was .6 db peaking on the master bus meter.


As for normalizer, If I understand how to use it correctly. After exporting my project I would bring it into my mastering program and normalize it raising it relative to 0db to get the best level prior to mastering.

I'm with GH  on the commercial tracks comparing.

As far as answering your questions aimed to me:

As for what is considered no headroom. Well, what's considered no leg room in a car or on an airplane? When your legs are getting squashed.

If you mix or mix and DIY master or have an engineer do so, they need room to make the necessary adjustments (hopefully, you've sent them as well-mixed track with no mastering since that is what they're there for). It's still subjective as every engineer's ears and tastes are different.

Like I said before, look at the waveform. For me, I try not to have the meat (the thick part) of it reach 3dB (for either channel from a mono perspective or both from a stereo perspective).

Don't mix too hot or too cold. Do mix somewhere in the middle.

Don't throw on a Limiter to bump the volume, simply turn up the fader.

Don't trust your eyes (not all the time) - Do: trust your ears (or more trained ears if yours aren't)

"As for what is considered no headroom. Well, what's considered no leg room in a car or on an airplane? When your legs are getting squashed. "


How would I know if my track has no headroom? Is it by looking at the meters if they reach the red zone, If the level goes past 0db, or if the waveform hits the ceiling?


Thanks

sharpdion23


"As for what is considered no headroom. Well, what's considered no leg room in a car or on an airplane? When your legs are getting squashed. "


How would I know if my track has no headroom? Is it by looking at the meters if they reach the red zone, If the level goes past 0db, or if the waveform hits the ceiling?


Thanks

Use both methods because you can't always trust the meter. See GH's example.

Also, if you can hear clipping (ie: you'll hear it distort).


Two ways this can happen:


Let's say, you're looking at the waveform and you hear a buzz (crank up the low end and you'll get this), the track (or that part of it) has clipped or distorted due to too much low end.


Take a distorted guitar. Something in the signal (volume) has been increased astronomically, that is why it sounds the way it does. (I'm using lamens terms) Note, this is the guitar's signal - not you cranking up the knob on the guitar as the signal is still distorted until you change the effect (ie: distorted to clean)



So, with the distorted guitar example, the waveform doesn't have to touch the ceiling; however, too much of a particular frequency will cause this. So, it isn't just volume, but frequency balancing as well.


Take that same distorted guitar track and take all the low end out of it. You'll notice that the waveform is much thinner because the what is perceived to be the volume being reduced is really the frequency reduction. Reduce the high end of a cymbal (hat/crash) and note the perceived volume decrease - again, this actually being frequency decrease.


So, maybe it's not volume reduction, but a certain frequency reduction. Which brings me to what to do with EQ: To get more high end, reduce low end. To get more low-end, reduce high end. To get more middle, reduce the low and high end. In short, cut rather than boost (bass is a very hard thing to tame)


Now, to the untrained ear, one would lean towards using the eyes; yet, ultimately, it's the ear - despite untrained - that takes over and there's the issue of "being too close to your mix." which is why you need those extra set(s) of ears.


(If anyone more knowlefgeable could answer, thank you)