Mastered for iTunes Logo

You’ve probably recently noticed the Mastered for iTunes (tag) on some serious productions on iTunes recently. As Apple is building a high fidelity catalog for its flagship online store in the hope to distribute lossless format only in the future, right now is about the best time to get on board.

Why is that ?

Traditionally, you would get your songs mastered for cd’s, in very high definition. But for the cd format, the mastering engineer has to cut the sound quality about in half, passing from 24bit/96kHz to 16bit/44.1 kHz. Then you would upload it on the iTunes store, where it would be re-encoded in a lossy format called AAC.

The thing here is that you’ve just downgraded the same file twice. This is really bad.

And that’s not all. When you master for cd, the cd is a format that is highly tolerant to clipping and distortion. However, it is not the case of AAC, for which the conversion codec is very sensitive to clipping. Therefore, a mastering for iTunes includes also a good habit of not trying to slam everything in the converters.

For the same reason, a mastering done for iTunes will translate very well into CD, but the opposite is far from being true…

Cool, so why won’t everyone just get your song Mastered for iTunes then ? – you would say.

Here is the trap… Even if Apple mention it nowhere on its MFiT program website, nor in its pdf, Apple is keeping track of the mastering engineers that called themselves MFiT service providers.

So, once you send your files to Apple, if you use traditional aggregators, such as CD baby, Believe Digital, Zimbalam, etc.

1) they ask you to send 16bit/44.1kHz files, and

2) they can’t guarantee you in any way that you can get the MFiT file.

Currently, the only way to get the prestigious MFiT tag on your album is to make business with a label that publish itself its catalog to iTunes and to give the name of your Apple approved MFiT service provider to your Apple contact while submitting. If you submit a 24bit/96kHz file, with a valid aggregator name, you should get the MFiT tag on your product. Still no guarantee.

That’s why, Quantum-Music is doing its best to take care of it for you. We’re currently on the official MFiT service provider list, and we dealt hard to give access to the MFiT program to the independent artists.

To recap, what are the benefit of getting into the MFiT program with us right now:

• You have a headstart compared to other independent artists.
• You get on the Mastered for iTunes page on iTunes, which incredibly boost your album’s visibility.
• At equal budget, you sound better than the competitor.
• You add a quality seal to your name.
• Dude! you’re on iTunes with MFiT tag, that’s not cool enough already ??

So, let’s make the quantum leap together !

Bests,

Even if acoustics is a complicated field, and a proper acoustical design cost a lot, there are some guidelines that can follow in order to tweek your room.

To start with, a great sounding room satisfies mainly two criteria:

• It sounds flat (all frequencies sound about the same level)
• It sounds dead (fast decay time, or as common knowledge would recognize as “echo”).

Hey!

I’ve just responded to a forum post on Audiofanzine.fr where the guy was completely desperate about mixing his rap vocals (http://fr.audiofanzine.com/techniques-du-son/forums/t.537530,au-secours-s-v-p,post.7463643.html) The guy didn’t have bad gear, not at all in fact. He has a Neumann microphone, a very decent preamp, but couldn’t get a good sound out of it.

As I always says, gear by itself is not making things sounding better, sound engineers do.

Having mixed rap vocals for a while now, the answer seems very clear to my mind. But, I have to remember to myself that it is far from being trivial at all. So, since he might not be the first one to struggle with it, and you might not speak french, I decided to post an equivalent answer here:

Here are some guidelines from which you can start, and then tweak for your specific case:

Compression:

First, rap vocals are recorded relatively close to the microphones, in order to get the “proximity effect” that is usually unwanted for vocals.

Rap vocals need strong RMS-based compression. For this, I would go with an opto compressor (such as La-2a or tubetech CL-1B, free plugin: ThrillseekerLA available on Variety of sound http://varietyofsound.wordpress.com/2012/03/02/thrillseekerla-released-today/). Apply a 4-5 db of gain reduction at a ratio of 3:1. Here again, we’re trying to accentuate the feeling of proximity of the voice.

Equalization:
Low cut: 100Hz
-2db at 800Hz
+ 1-2db at 1.2kHz
high shelf + 1-2 db starting at 4kHz.

This equalization will make sure to remove the muddy noise of voice (800, under 100) while bringing some shine to it (1.2kHz, 4kHz).

If your compressor is not an emulation, it can also be interesting to add an exciter to the chain, but make sure it is subtle (enhancing, not degrading).

As I mentioned, those are guidelines, you might need to tweak them a bit to perfectly suit your needs. Interestingly, those settings are very similar to what I would use on pretty much any kind of vocal, except, of course,  the close miking technique.

Good luck with that, my friends,

Chris

The following question has been asked under the “how to adjust compressor settings (part 1)” post:

I have a question for you if you don’t mind. I seem to fail at squashing peaks with high ratio/high threshold because the compressors i’ve tried won’t have fast enough attack times so the peaks partially go through and they end up eating my headroom (unless I use a limiter and that feels like cheating). Am I doing something wrong?

Hello Rhuobhe’s, thank you very much for your participation. I don’t mind, I encourage it ! It seem to me that the compressor you’re using is either not set properly or simply not the appropriate one for your application. Let me explain:

• Like you have said, yes it’s possible that the attack time is not fast enough. I suggest you to read the blogpost about adjusting attack and release time on compressors that is a bit more recent that the one you posted your comment. If you can set the attack time faster on your compressor, is it solving the problem ?
• If not, it is definitely possible that the compressor you’re using is simply not the good one. There is a subject that I’ve not covered yet on compressor: Peak/RMS detection. Before taking a decision, the compressor analyze the signal. The way the signal is interpreted has a HUGE influence on how the compressor will sound. RMS detection take into account the area under the curve a certain amount of time before  (averaged value) while the peak is simply taking into account the actual position (instantaneous value).  For signal with dangerous dynamic that might cutoff your headroom, peak detection might be the right choice.
• About that, compressors are often based on RMS detection (smoother response) and limiter on peak detection. In my case, I think the best is to have the choice. Compressor RMS-detection-based will sound good on vocals, but not necessarily on high dynamic content. That can be also the reason why we have the impression sometimes that a cheaper compressor is better that an expensive one for some application.
• There is nothing bad using “limiter” if it’s not a maximizer or a hardclipper. A lot of compressor, like VOF Density for example, that let you select between LIM or COMP. In that case, Limiting has nothing to do with maximizer. Those limiters often have a ratio of about 10:1, which  is not clipping. Try to avoid the use of maximizer or hard clipping.
• I think, in your case, the best will be a peak-detection compressor. Set at near-zero attack setting, it will compress.  If you don’t have one, I can program a simple one for you.

I hope this answered to your question. Please, do not hesitate if you have other questions.

Sincerly,

Chris

Like you might noticed, I’m a big fan of digital domain audio. That said, this does not mean that I am not aware of its weaknesses. In fact, it’s quite the opposite. I think, the best way to enjoy digital, it’s by knowing its limits. Welcome to my world !

Let’s go trashy: Discrete versus Continuous

Switching from analog to digital world can be overwhelming. I mean, nothing will be as big, in terms of change, than switching from the tape to the hard drive. Sound engineers (or most often, their assistant!) could spent hours cutting the tape with scissors to put scotch tape on it, in order to replace a snare hit! From that point of view, digital is definitely wonderful. That is one of the main reason why we all work with computer now: We spend less time on stupid things, and more on what matters, which is the sound.

It is true, however, that digital at its beginning was not wonderful at all. The first A-D converters sounded disguistings (literally!) and digital compressors were victims of aliasing since it processed the sampled point only and not the intersampling data !  Why is that ? Let’s go with some theory here…

Analog domain is continuous, every instants you have current passing. It is smooth, there is no aliasing possible. However, in digital domain, the information we got is an instantaneous value at a very specific time, then no info for few milliseconds. So, if a compressor is treating only the information he has, like if it was linear between point, then you’re find yourself with aliasing phenomenon… Why ? Because signal between sampled point can trepass the threshold without ever been treated !

Why the sampling frequency is not higher then?

At the beginning of digital, memory capacities were incredibly low. For this reason, they set the sampling rate standard for compact disc at its minimum. But, at this point, according to there mathematics, it was suppose to be lossless… How do they chose the minimum do you think ???  Check it out… you’re going to find it funny…

Nyquist-Shannon sampling and reconstruction theorem

The actual theorem affirm that it is possible to recovery entirely a signal if the sampling rate is at least two times the highest frequency present in the recording. But the part that most engineers tend to forget is the following: Yes, it is possible to recover entirely the signal if we use the SINC function to recover the sample. Since you’re probably not a DSP engineer, I think it’s not a luxury to specify more clearly what a SINC function is… Sinc function can be described as follow: y =Sin(x) / x. In other word, a very complex function that as to take into account every point that came before and after that actual event. No real time processing could ever do that!

Okay, so what’s the problem ?

Do you really think that every cheap cd player tries to interpolate between points by applying a complex sin(x)/x form on the entire song before even playing it ?? The answer is : depends on the daw and its settings. Even in professional software, during playback, the sound you hear is sometimes linealy interpolated ! That is why we are so tempted by 96kHz sampling rates ! It’s not because 44.1kHz is bad, it simply because it takes to much resources to interpolate following a sin(x)/x form! Let’s take a look at the following cheap figures:

Do you really think that every cheap cd player tries to interpolate between points by applying a complex sin(x)/x form ?? The answer is : HELL NO!

So, what is the moral of this story ?

1. Never turn your back on digital. In order to use it efficiently, you need to know its strenghts as well as its weaknesses.
2. People who are complaining that digital sounds cold, ironically, they usually the same people that are claiming that is no difference between 44.1kHz and 96kHz… Those people are also often interpolating linearly in order to save CPU.
3. You can achieve very continuous waves in digital domain, without the analog drawbacks simply by either working at very high sampling rate  or by working at high sinc interpolation quality.
4. If you’re working at 44.1kHz, the loss that you get from linear interpolation is in the high frequencies. If you want to hear the difference, that’s were you have to focus your listening.
5. Every time you use a compressor, make sure it oversamples at least by a factor of 4 to make sure to avoid aliasing caused by intersample overshoot.
6. There is no need to buy summing analog gear ! Simply work at smart digital settings, it will do the same, or even better !
7. A-D/D-A conversions are the most degrading part of a studio… if you can avoid it… do it.

I hope you appreciated that article, and that will be useful to your future digital audio projects !

Sincerly

Chris.

Last week, I talked a bit about the intimate relationship between the threshold and the ratio. It seemed pretty straightforward, and it was. However, this week, we are going to investigate more deeply parameters that still seem misunderstood, even for sound engineers. Of course, attack and release times are less obvious to the human ear than distorsion, for example.

Why attack and release time matter ?

Attack and release times play an important role in the quality of compression [Zolzer, 2008]. It is true that attack and release set to zero will distort more easily. Also,  in that case, the  compressor doesn’t care about the feeling transported by the waveform; Mechanically, it will simply chop off everything by half (for ratio 2:1) that will trespass the threshold. Like Bootsie said, “the magic is where the transient happens” [Variety of Sound, 2009], if true, which is, therefore it cannot be systematically cutoff whatever it is supposed to express. For those who don’t know what transients are, let’s say peaks.

What are the challenges one might face while adjusting these parameters ?

First, monitoring and acoustics will play important roles on those adjustment. The fidelity of the speaker will play on the attack time, while the acoustic of a room will play on the perception of release time. Why so that ? Let’s go deeper into that:

The attack felt is directly proportional to the ability of the woofer to reproduce the dynamic. If you’re woofer is made of heavy materials like cardboard, the speaker will respond slowly compared to the one made of kevlar. In other words, if the dynamic of the speaker is slower than your attack setting, you won’t hear any difference.

The release time is hard to hear if you room is very echoic. Why is that ? Because the reverb of your room consists of an amalgam of delayed sound. Therefore, more the reverb level is close to direct sound in terms of power, more you brain will take into account of past events rather than instanteneous ones. It is then a fact that dead environment is better than a live environment for compression release setting.

Theoretical background

First, what are the attack and release time ?

Attack time is the time it takes to the compressor before reaching the gain reduction it should apply once the threshold trespassed. Similarly, the release time is the time the compressor will continue to apply the gain reduction after the signal get back under the threshold.

Where is that coming from ?

The attack and release time are originally coming from the analog domain. Since feedback designs were used, the gain reduction applied by the compression was based on the information that came in few milliseconds before. It’s funny to see in the Altec 436 manual, which has fixed attack time of 50ms was considered as a “fast attack” compressor. Nowaday, it would be considered as slow.

How to adjust them ?

Attack time:

I usually start by setting the attack time first since in many compressor design, the release time is function of the attack time. When listening to the effect, you have to focus your attention on the beginning of the peaks. While a zero attack will brickwall the peak, a little longer attack will let it pass a bit. Now it’s a question of taste which also depends on the particular situation.  But, a lot of people like to let pass the attack of high dynamic instruments. For drums as an example, letting the attack pass a bit before compression helps to make the compression more transparent, since the hear still feel the punch in the dynamic even if the rest of the curve is compressed. Voice also gains in having long attack time, since the consonants can pass a bit like percussive sounds. On the opposite side, some sounds like slapping bass, with over exagerated slapping noise will gain in being entirely compressed. Same thing when trying to deess (remove harsh “sss” sound in a voice). In other word, if the impact is desirable: go with a longer-than-zero attack time. If not, if the impact is annoying: cut it straight away with near zero attack time.

Release time:

Release time is often used to minimize audible distorsion. The distorsion phenomenon occurs when is squared by the threshold almost like clipping. This occurs since the samples just under the threshold have almost the same values as the one just above and the ear interprets it as if it was a continuous square wave. By adding a release time, we are pushing the data close to the threshold a bit away, so the ear doesn’t hear it at the volume.

So, according to that explanation, when release time is needed ?? Long release time is particularly needed when the overall volume is close to the threshold. Otherwise, if it is an almost instantaneous huge peak and the rest is really quiet, a very quick attack time would do the job without pumping artifacts. Longer release time than required will translate into pumping effect, which is, in most of the cases, undesirable.

To conclude, I hope this article has been  exhaustive enough, please do not hesitate to leave your comments or share your ideas. You can like the www.quantum-music.ca facebook page or subscribe to the RSS flux to get news feeds.

The typical parameters available on a compressor are the following:

• Threshold;
• Ratio;
• Attack;
• Release;
• Make up gain.

To simplify things, we can split this in 2 sections: Gain reduction and time constants. You will usually set the threshold-ratio couple before touching the time constants since the latters are the fine tuning part of a compressor. The make-up is a pretty straight forward setting since its simply the volume compensation you can set to recover from gain reduction.

Threshold and Ratio settings:

Threshold and ratio are working together in order to set your gain reduction. Typically lower your threshold will be, lower will be your compression ratio also. Conversely, higher threshold will allow you to push harder the ratio. How you will set those two parameters will depend on what you’re trying to achieve. Few examples :

• Vocals : You’re trying to make the vocals compete with a bigband… Good luck! You want the average volume to feel “inflated” but also the peak controlled.
• Master mix: You’re trying to inflate the mix without squashing the peaks.
• Drum bus: You’re trying to squash the peaks!!

Every situation will promote different settings, of course, but for those situations, this is what I would be tempted to do:.

• Vocals : Since we have an important amount of gain reduction to achieve I will go with two compressors with different strategies. One that will serve to increase the overall volume to bring up the details, and one squash partially the louder peaks.

1. Inflation: Low threshold with low ratio (1.5:1) achieving 3db gain reduction;
2. Brickwalling: high threshold with high ratio (3 or 4:1) achieving 1-2 db gain reduction;

• Master mix: You’re trying to inflate the mix without squashing the peaks.

• Drum bus: You’re trying to squash the peaks!!

1. Peak squashing: high threshold with high ratio (4 or higher:1) achieving what you need in terms of gain reduction.

See you next week for the following !

The importance of vocals

Except for instrumental music, vocals are the most prominent instrument of a mix. Some engineers say that if you’ve got the vocals right, you’ve got the mix right. Also, the term “song” would be inappropriate if the point wasn’t about “singing”. Interesting fact, the human ear is way more critical in about vocals than any other instrument. The reason is fairly simple, it’s the only instrument that everyone plays everyday. Futhermore, the human has a deeper feeling towards another human rather than any object. That’s the very same reason why they show human faces in product advertisements. Another interesting aspect of vocals is the lyrics. Currently, the vocals is still the only instrument that can put words on a song. This adds an other dimension to a song.

1. Cutting filters
2. Compressor/De-esser
3. Equalizer
4. Exciter
5. Spatial effects (Delays & Reverbs)

2. COMPRESSOR

Why ?

If there is an instrument that absolutely need compression, it’s vocal. In a rock band arrangement, there is no way a voice can compete with a banging drum, two electric distorted guitars and an overcompressed bass. The human voice is simply not dense enough. That’s the reason why we use compression in order to densify it.

How ?

The amount of compression will highly depend on the other instruments, but as a rule of thumb, should never exceed 6dB of gain reduction, preferably 4db. Beyond this point the vocal loses it’s liveness. For some musical genre, it might be appropriate though. A good example could be hip hop where all the other sounds are very loud and the vocals is expected to be upfront. A good free compressor that would do an excellent job is the marvelous Variety of sound Density MkIII reviewed early on this blog. I also like the Kjaerhus classic one. The good thing with compressor is that you can find a bunch of them for free on the internet. Few of them are good, but still.

Depending on the musical genre, the real challenge will be the selection of the compressor. For some application, it is great to have a rich in harmonic sound in the voice, for other it’s better if it’s clear and transparent. Like most of the decision we make in the studio, it’s better to settle down a bit, think about what we want before trying anything.  An example of very transparent compressor will be the elysia or the waves C1, while a LA-2A or an Altec 436C will be more colored. Now it has been discontinued but, back then, if you wanted to design your compressor by yourself the Kjaerhus Audio Golden Compressor offered all the feature one needed.

And what elssssssse?

De-essing ! I think that it is the appropriate step to add the deesseing. If you have no idea what it is, it’s like a limiter that you apply only on the high frequency range in order the control the annoying “s” sound. Just a little comment: don’t push it to hard, the treatment must be there but transparent. You don’t want to lose the air in the voice with the esses!

Variety of sound – Density MKIII

The importance of vocals

Except for instrumental music, vocals are the most prominent instrument of a mix. Some engineers say that if you’ve got the vocals right, you’ve got the mix right. Also, the term “song” would be inappropriate if the point wasn’t about “singing”. Interesting fact, the human ear is way more critical in about vocals than any other instrument. The reason is fairly simple, it’s the only instrument that everyone plays everyday. Futhermore, the human has a deeper feeling towards another human rather than any object. That’s the very same reason why they show human faces in product advertisements. Another interesting aspect of vocals is the lyrics. Currently, the vocals is still the only instrument that can put words on a song. This adds an other dimension to a song.

Enough talk, more tricks!

First thing first:

1. Have a great song
2. Have great lyrics
3. Record it right: The performance must be flawless first.

1. Cutting filters
2. Compressor/De-esser
3. Equalizer
4. Exciter
5. Spatial effects (Delays & Reverbs)

1. CUTTING FILTERS

Why ?

The best way to start is by removing unwanted frequencies and resonances. This will help to make the vocal track cut through the mix more easily. The idea here is to cut unwanted frequencies before the compressor and boosting others after. The reason why we are doing the equalization in two steps is simply to clean the signal in order to help the compressor doing it’s job.

How ?

First, start by removing everything under ~100 Hz and above ~20kHz (Of course, the cutting frequencies will depend of the singer, use you judgement). There are good chances you can also use a peak filter to cut around 700-800 Hz by few dB in order to remove nasal resonances. A good free equalizer to do this job would be the 1982art Gloria reviewed early on this very same blog.

1982Art – Gloria

That’s enough for this time. See you soon for the following…

« MASTERING IS THE LAST CREATIVE STEP IN THE AUDIO PRODUCTION PROCESS, THE BRIDGE BETWEEN MIXING AND REPLICATION – YOUR LAST CHANCE TO ENHANCE SOUND OR REPAIR PROBLEMS IN AN ACOUSTICALLY-DESIGNED ROOM. »

- BOB KATZ, THE ART OF MASTERING.

The previous definition is coming from the book “The art of mastering” which I consider to be the bible of the mastering engineer. Bob Katz wrote down things in a way that nor I or any other mastering engineer could have. It is right, clear and inspiring. For whom this definition is still not clear enough, mastering can be seen as a multipurpose process… Here are listed the basic reasons to get your mixes mastered by a professional:

Homogeniety:

For whom this definition is still not clear enough, mastering is the step where we try to make fit different mixes in a whole. What I mean by that is fairly simple: Some mixes has been done tired at 2am, others fresh as a flower at 9am; some of them sound crazy loud and others very soft… Now, the question is : How do we put them together on a same record ?

Standardization:

Another good reason to get your mix mastered is to sound good everywhere. The problem with mixes, is usually that they have been mixed in less than perfect acoustical environment, so by definition, they won’t translate well everywhere else. A mastering studio will have a near-perfection acoustically treated room and very flat monitoring setup in order to make a sound-good-everywhere version of this mix.

Loudness:

You want your mix to be competitive in terms of loudness ? Okay, but don’t do it yourself, because it is the best way to ruin in 10 sec an excellent mix. The mastering engineer will use its talent, knowledge and experience to make sure that your songs don’t suffer too much from gain reduction undesirable effects. Moreover, mastering engineers use high end gear that helps to conserve the integrity of the master.