Final activities to create the song.

What you're talking about is usually done during mastering, but you can do some of it yourself. Basically, after doing your best to create a good mix, you ruin it by putting compression/limiting on the master bus to reduce peaks and thereby increase the RMS volume. X3 has a few tools, most specific for the task would be Boost11 although not everyone is a fan of the sound. You could also consider getting a few basic "home mastering" plugins with presets if you don't know a lot about it, e.g. Ozone (basic version) by Izotope has a very high instant gratification factor. It goes on sale relatively often, you can check jrrshop or plugindiscounts.com.

Kalle Rantaaho

rove
Thanks a lot.
I tried Boost11, but when I listen the song, for example during my trip by car, the volume is a lot down instead a normal audio cd of music...
 

How did you use it? What did your Master Bus meter RMS level show when you played the project using Boost11?
It's near the beginning of red light.
 
It is, for example, possible, that you have so high peaks in the project that you should first do a volume automation to get the volume in control.
How shall I do it?
 
Creating a loud mix starts from the track level. You have to find the right type of sounds and arrangement, use EQ and compression when necessary etc. NOTE: the volume does not need to be maxed  on track level. -6 - -12 dB is
quite sufficient.
Yes, I'm near this values

If the techies on here will allow me to grossly oversimplify some things I'll try to explain what needs to happen and how you could do it.

Basically in the digital audio domain, there is such a thing as a "maximum volume". Because everything is represented by numbers, at some point you reach the highest agreed upon number and you can't go any louder. This is less apparent in the analog world where you usually get degrees of distortion rather than hard line clipping.

Because of this maximum value, and because everyone wants to have his/her music sound as loud as or louder than the competition (for various reasons), firstly songs are generally mastered so that the highest peak reaches the maximum value. Our DAW master bus meter generally shows this as 0dB and will warn you if a mix goes over. So the first step if you want to do some home mastering is to mix so that the end result gets close to a peak of 0dB. Not each individual track, but the master bus.

If your track still isn't loud enough for your tastes after this, a common method to get increased loudness is what we were talking about before - reducing the peak values so you can push the total up. The absolute simplest way to do this is to put a brickwall limiter as the last effects on your master bus and make sure it's doing plenty of gain reduction. Depending on the quality of your mix and the limiter in question, this can make the end result pretty bad (but definitely louder). Another option is to address what's making the peaks. E.g. snares are common sources of peaks, you could use heavier compression on the snare or just bring it down in the mix, voila, lower peaks. In practice it's often a combination of these things that will yield the best result. I would ask your ProTools friend to come by and show you a thing or two using Sonar Tools. I guarantee you that it's not a magic tool - he just knows which buttons to press ;)

Another factor in "loudness" is how we perceive volume.

Our brains tend to judge volume not by transient, brief peaks of a few milliseconds but by it's average over a longer (though still brief) time. This is why we distinguish between peak volume (the absolute maximum reached) and RMS volume which (put simply) is a kind of average. Pulling up the transient peaks will have less effect on perceived volume than raising the RMS volume.

An example of this is the Fuzz Face pedal. With its' volume and fuzz controls maxed out it sounds deafening compared to the clean amp volume, yet the Fuzz Face design can generally only just reach unity volume with the "clean" sound if you look at what it does using a meter. It sounds louder because it compresses like mad and so raises the RMS level which makes it sound really loud even when its' peak volume is actually lower than the clean signal. Many a guitarist has found that the setting on a fuzz, distortion or compressor that sounds like a huge volume boost at home results in their relative volume dropping when in the context of a band mix.

So to maximise volume we need to both raise the transient peaks as far as possible and also reduce the gap between peak and RMS by raising the RMS level. This is done using compressors or limiters, usually after mixing and at the mastering stage. A good starting point is to set the master bus's meter to read both peak and RMS. then put a limiter last on the master bus (out of Cake's plugins the Concrete Limiter is particularly good for this, though Boost 11 or any compressor capable of a very high compression ratio will do). Set the limiter up with a high ratio and adjust the output so that the maximum volume never goes above 0.3 dB or there about. Now use the threshold and ratio controls to raise the average level.

Go too far with this and your music will certainly sound loud, but lose a lot of life and the punch that comes from quiet and loud transients. An RMS level somewhere between -20 and -12 seems to be the popular area to aim for, though some commercial recordings are pretty much a brick-like transient-free block if you look at their waveforms....

And you most certainly do not need Pro Tools to do this. Sonar is quite capable of doing the job as are Cubase, Logic, Soundforge, Audacity, Wave Lab, Reaper, Studio 1 and any other DAWs or sample processors that either have the functions built in or can run plugins.

Oh, and if you're mastering for vinyl forget all the above. Vinyl has technical requirements that mean simply maxing out volume is a very bad idea indeed. Which might be why some people prefer vinyl to CD, what they are hearing in the vinyl (which on paper is a worse medium) is the transients lost in a digital master which focused on nothing but getting it as loud as possible so it grabs people's attention (as every good hi-fi sales person knows, if you turn up the system you particularly want to sell a little more than the one demonstrated before it, the odds are the customer will think the louder one sounds "better").

tlw
Another factor in "loudness" is how we perceive volume.

Our brains tend to judge volume not by transient, brief peaks of a few milliseconds but by it's average over a longer (though still brief) time. This is why we distinguish between peak volume (the absolute maximum reached) and RMS volume which (put simply) is a kind of average. Pulling up the transient peaks will have less effect on perceived volume than raising the RMS volume.

An example of this is the Fuzz Face pedal. With its' volume and fuzz controls maxed out it sounds deafening compared to the clean amp volume, yet the Fuzz Face design can generally only just reach unity volume with the "clean" sound if you look at what it does using a meter. It sounds louder because it compresses like mad and so raises the RMS level which makes it sound really loud even when its' peak volume is actually lower than the clean signal. Many a guitarist has found that the setting on a fuzz, distortion or compressor that sounds like a huge volume boost at home results in their relative volume dropping when in the context of a band mix.

So to maximise volume we need to both raise the transient peaks as far as possible and also reduce the gap between peak and RMS by raising the RMS level. This is done using compressors or limiters, usually after mixing and at the mastering stage. A good starting point is to set the master bus's meter to read both peak and RMS. then put a limiter last on the master bus (out of Cake's plugins the Concrete Limiter is particularly good for this, though Boost 11 or any compressor capable of a very high compression ratio will do). Set the limiter up with a high ratio and adjust the output so that the maximum volume never goes above 0.3 dB or there about. Now use the threshold and ratio controls to raise the average level.

Go too far with this and your music will certainly sound loud, but lose a lot of life and the punch that comes from quiet and loud transients. An RMS level somewhere between -20 and -12 seems to be the popular area to aim for, though some commercial recordings are pretty much a brick-like transient-free block if you look at their waveforms....

And you most certainly do not need Pro Tools to do this. Sonar is quite capable of doing the job as are Cubase, Logic, Soundforge, Audacity, Wave Lab, Reaper, Studio 1 and any other DAWs or sample processors that either have the functions built in or can run plugins.

Oh, and if you're mastering for vinyl forget all the above. Vinyl has technical requirements that mean simply maxing out volume is a very bad idea indeed. Which might be why some people prefer vinyl to CD, what they are hearing in the vinyl (which on paper is a worse medium) is the transients lost in a digital master which focused on nothing but getting it as loud as possible so it grabs people's attention (as every good hi-fi sales person knows, if you turn up the system you particularly want to sell a little more than the one demonstrated before it, the odds are the customer will think the louder one sounds "better").

Thank you a lot for the precius and clean informations.

It's also well worthwhile to get your monitors calibrated to a known level, usually between 78dB & 85dB.
 
This forms part of Bob Katz K-System for metering.
 
There are tons of posts on this subject all over the internet so feel free to have a browse.