This is likely yet another controversial post… Up front, it’s in no way meant to dismiss the skills of professional Mastering Engineers. Without a doubt, their services are an absolutely essential element in a professional recording studio environment.
That said, this post begs the question:
Can Mastering be performed economically in a home studio?
My answer is yes, at least the basics. Best of all, you can accomplish it for free!
Before continuing, I should note that REAPER (by Cockos) is my primary DAW. In my opinion it’s one of the very best, with a ridiculous number of powerful features. But, enough of that, since these techniques can be applied to just about any modern DAW out there.
If you don’t happen to be using REAPER, all of its plugins referenced here and many more are included in the ReaPlugs VST FX Suite. I’ve also provided links to where the remaining plugins can be downloaded, in each of their respective sections below.
All are absolutely free.
Let’s start with a common misconception:
Is Mastering always going to make a mix sound good?
As an analogy, think of it like a car: You can take an old, rusty one and polish it up, but will that magically transform it into a Lamborghini? Of course not, it’s still a rusty old car.
If the mix is lousy to begin with, no amount of Mastering in the world is going to come to its rescue.
Following are what I consider the basic tasks of Mastering in a home studio, and how I address them once I have a good mix completed. These techniques may seem overwhelming at first, but I assure you they’re quite easy to perform once you become familiar with them.
Mastering Individual Tracks
- Noise Reduction
Naturally, things like clicks, pops, hiss, and between-phrase breaths by vocalists are undesirable to hear on any given track. Worse yet, noise on individual tracks will of course be passed to the master bus, accumulating there and increasing by orders of magnitude in the fully rendered song.
I accomplish Noise Reduction by splitting each track into musical “phrases”, then deleting the garbage that’s in-between (where noise is actually taking place). Then, I’ll often add short fades to the beginning and/or end of each phrase, so they sound smooth all on their own.
If there’s a lot of noise occurring during musical phrases in the mix, it wasn’t recorded correctly in the first place. Time to go back, fix whatever is causing it, and re-record.
- Correcting minor performance flubs
I don’t care how proficient you are in playing your chosen instrument, sooner or later you’re going to make a few screw-ups. Re-recording the entire track, overdubbing and comps are obviously all remedies for this, but can become a major pain, especially in long songs.
A common scenario is an out-of-time snare or kick drum hit. Call me crazy, but I’m certainly not going to re-record the 9 microphone tracks that cover my entire kit because of a few small mistakes like this. Instead, I simply use my DAW editor, making a nudge or two to move that hit where it belongs.
Another example is, I’m not above using a pitch correction plugin to fix slightly out of tune notes, especially on vocal tracks.
Some people frown on these practices, using the slightly derogatory term, “fixing it in the mix”. I respectfully disagree, since it’s one of the major reasons DAW editors exist in the first place.
Mastering using only the Master Bus
This will certainly be blasphemous to professional Mastering Engineers: I perform what I consider the remaining essentials of Mastering directly on the master bus, in the same DAW project file as the mix itself.
Using this approach, there’s no need to bounce back and forth between loading separate mix and Mastering project files into your DAW, which is super-convenient when any adjustments to individual tracks are required. In the latter case, the prerequisite of having to re-render the completed mix for loading back into a Mastering DAW project is also eliminated.
The plugins I normally use follow, ordered from first to last in the master bus effects chain. Alternately, you can of course use your favorites for each purpose, if you’d prefer.
Also note that all referenced plugin settings are examples, and depending on the mix or your particular tastes may naturally require some adjustments.
This will likely be a real shocker, but I very rarely use an Equalizer plugin on the master bus. It’s my feeling that if EQ is required there, I’ve failed to apply it properly to one or more tracks in the mix.
Don’t place an EQ plugin on the master bus just because that’s what you usually do on individual tracks. If the mix already sounds good to you frequency-wise, doing so serves no purpose whatsoever and can easily foul it up.
However there will be instances where the mix is not balanced and clear as a whole. When I feel a tweak or two of EQ is required by using my ears, my choice is REAPER’s ReaEQ plugin (also a component of the previously mentioned Cockos ReaPlugs package).
While making any required adjustments, ReaEQ displays not only the processed signal (yellow line) but the phase as well (red line), both in real time. You can also add as many frequency band tabs as you like, and there are a wide variety of different types available (low/high shelf, normal band, notch, etc).
Using a Compressor is one method of raising relative (perceived) volume level, but during Mastering I use it for its other primary purpose: To attenuate louder signals while bringing up quieter ones. Leave that final volume boost to a Limiter, which will be discussed later in this post.
As my Mastering Compressor, I use another Cockos plugin called ReaComp. There is also a more sophisticated version of it available named ReaXcomp which allows compression to be applied to specific frequency ranges. To keep things simple I very rarely use the latter, as ReaComp simplifies things by taking care of all ranges in one shot, yet still manages to perform the task of master bus compression extremely well.
A word of caution: Compression must be done with great care on the master bus, to avoid the introduction of “pumping” (discussed below) or other undesirable effects on the entire song. I normally begin with the following primary settings:
- A very low compression Ratio (2:1).
- For openers, very fast Attack and Release times. This is where some experimentation is in order to avoid the “pumping” effect, which is caused by one or more dominent tracks cutting through the rest. As a result, those other tracks will noticeably increase and decrease in level. Simply use your ears, it’s an easy thing to spot.
- A Threshold setting that provides just a few dB’s of compression.
- Volume Level
My absolute favorite Classical Music composer of all-time once said…
“The music is not in the notes, but in the silence between.”– Wolfgang Amadeus Mozart
A good point indeed regarding music in general, and one that also applies to studio work today. What if there is no silence (or more appropriately for this discussion, no dynamics) between those notes? The short answer is, bad things happen!
If you aren’t aware, there’s been what’s termed as “Loudness Wars” going on for several decades now. It’s sort of a sonic arms race where studio engineers feel the need to crush their tracks to the highest possible level for fear of the production not being “competitive” with songs produced by others, especially for radio air play. As a result, the song’s dynamics can be nearly (or in some cases completely) eliminated, which can totally ruin the finished product.
To better understand this, some visualization is helpful. Below are stereo waveforms of songs from four Rush albums released between 1985 and 2002.
As you can clearly see, as time went on, “headroom” (the available dynamic range) was reduced more and more with each album, until it was nearly non-existent.
The original release of their Vapor Trails album is well-known to be a distorted mess. In fact, it was so bad that a “re-mastered” version was released shortly after the original to correct the issue. Fortunately, that was possible since the distortion was introduced during the Mastering process and not during recording, which would have made it impossible to fix.
A good rule of thumb is to set recording levels of individual tracks to average about -12dB, which allows for plenty of headroom in preparation for Mastering. Naturally, that results in a corresponding low level output on the master bus, which is totally normal. However, at this stage it will lack the “punchiness” and definition that’s desirable in the completed production.
All that in mind, there’s still nothing wrong with boosting the output level of a mix, as long as it’s not overdone.
This final boost is accomplished using a Limiter, which is essentially a Compressor with an extremely high ratio. Its primary purpose is to make your master bus signal as loud as possible, without introducing clipping or distortion.
I use a very simple yet excellent “brick wall” limiter, called LoudMax. It performs a “look ahead” at the upcoming signal in preparation for increasing as well as limiting its level. There are just two controls to deal with:
- Threshold: Determines the level where limiting begins.
- Output Ceiling: The “brick wall”, specifying the maximum signal level the limiter will allow through.
Meters at the top and bottom display signal levels before/after Limiter processing, respectively. The meter at right indicates the amount of gain reduction in relation to the Threshold that has been set.
I always set an Output Ceiling of -0.1dB, the maximum before clipping will occur by the signal exceeding 0dB. However, this ceiling is also relative to the volume level of the master bus itself, so clipping can indeed still occur if that’s set excessively high.
Adjusting the Threshold control is what facilitates achieving the balance between dynamics (headroom) and the overall “loudness” of a given mix. Note that this balance is very delicate, so easy does it. I try to keep the gain reduction meter at around 6db or less.
Most DAWs display the master bus waveform while rendering. Here’s an example: It’s visually apparent that the output signal has been greatly increased, but not so much that its dynamics have been destroyed.
To reiterate, since all of this depends on the level of the mix, there’s no hard and fast rule for the Threshold setting. Experiment with this control and let both your eyes and more importantly your ears be the judge using these tools.
Though Loudmax is a simple yet effective Limiter plugin that’s easy to configure, I do realize the entire subject of limiting can be confusing. If you’d like to view an in-depth tutorial on its installation and proper use, David Harry has created a detailed one on YouTube.
- Fade Outs
If there’s one thing that drives me crazy, it’s hearing song fades that are cut short.
Fade-outs have nothing to do with plugins, but I’ve included the subject in this section since it is in fact a task which I also perform using the master bus.
I consider applying fades to all individual tracks simultaneously a very bad habit. This can be tricky and cumbersome, especially if there are many tracks making up a given song, which can easily lead to incomplete fade-outs.
Providing your DAW is capable of automation, there is absolutely no need to perform song fade-outs in this manner. Instead, simply use a volume automation envelope on the master bus, which centralizes applying the fade to the entire song.
The shape of the fade is also of utmost importance. Avoid using a “linear” curve, as this brings down the volume level much too abruptly, no matter how far its duration is stretched out.
Different fade shapes available vary by DAW, but I suggest using whatever comes close to a “logarithmic” curve which smoothly tapers down as pictured above, ending at “-inf dB” (absolute silence). Its tail should then last for at least a few more seconds prior to reaching the end point of song rendering, particularly if there are effects such as reverb or delay in use which may sustain beyond that point.
- Spectral Analysis
Lastly, I perform a spectral analysis using Voxengo’s SPAN, which is the final plugin in my master bus effects chain. It’s actually not an “effect” per se, but instead both a visual and auditory tool which will greatly assist in making your final rendered song sound as balanced as possible across the entire frequency range.
In a nutshell, it displays real-time analysis of incoming audio signals on a graph. The horizontal axis reflects frequencies, and the vertical axis their levels.
A “tonally balanced” mix has a reasonably consistent frequency response. Conversely, a mix with poor tonal balance may have very noticeable peaks and valleys in some areas on the graph. This for example could be an overly emphasized low-end, or a diminished mid-range. Lower frequencies should always blend smoothly with higher ones.
Although this plugin may appear a bit intimidating at first, the most critical areas to pay close attention to are:
- Main graph: Don’t expect the graph to be perfectly flat, but look for any significant peaks or valleys. If present, SPAN can easily identify the problematic tracks in the mix, as follows:
- Hold down the CTRL key, then left click. A curve with adjustable bandwidth will be displayed, which is dragged horizontally to isolate those excessive peaks and/or valleys. Then, only that frequency range will be audible during DAW playback.
- One method of correcting these issues is to attenuate or raise those frequencies by applying an EQ or similar plugin to the offending track(s) that you are hearing within the curve. It could also just be a matter of simply lowering or increasing the level of those track(s), so I’d suggest trying that first.
- Clippings/Peak: If you’ve set up your Limiter correctly, the number of Clipping instances should always be zero, and Peak should match its Output Ceiling.
- Correlation Meter: This is the relationship between the left and right (stereo) channels. You should be in the green and as close to “+1.0” as possible, which indicates both channels are in phase with each other.
Any amount of “out-of-phaseness” will push the meter towards its negative (red) half, residing between “0.0” and “-1.0”. If that deviation is constant, it indicates a reduced degree of mono-compatibility. In other words, something will get lost (or attenuated) when listening in mono, which obviously must be totally avoided.
Those are the basics. SPAN also provides a staggering amount of additional function, which you can read about in detail on its download page, if interested.
In summary, is all of this going to have the same result as what a professional Mastering Engineer is able to accomplish?
Of course not… So, if you have one at your disposal, don’t go firing him just yet!
In my case, I certainly can’t afford to spend thousands of dollars on plugins specifically tailored for Mastering music. The techniques I’ve just described cost zero to implement, and in addition are much more simplified. As a home studio enthusiast, in the end they suit my purposes just fine.
If you’re in the same boat, go ahead and give these methods a try… It won’t cost you a dime to do so, and you may be surprised to find them to be as effective as I do.
Here are a couple of my latest productions using these techniques, you be the judge:
Finally, as with most things in life, your mileage may vary. Once more, don’t be afraid to experiment!