Professional Analog Mastering.
Professional Analog Mastering.
226K+ subscribers
I mentioned this in our last video, but I’ll mention it here - you’ll need to create this chain within the context of the full mix.
As soon as you solo the vocal and start adding processing that way, you’ll lose how the vocal and its processing interact with all other signals.
Some of these signals will mask or cover up parts of the vocal, and others will amplify parts of the vocal, so it’s impossible to tell the true sound of your vocal chain if you only ever listen to it soloed.
An exception to this comes with resonances that need to be identified and attenuated - which is easier to do if you have the vocal soloed. But in all other cases, make sure the vocal isn’t soloed.
Let’s listen to a vocal being soloed and then play with the full mix. Notice how the timbre changes significantly.
Watch the video to learn more >
I know this sounds kind of obvious, but it helps immensely to know the style of vocal you want first. Do you want a classic sound, one with a good deal of harmonics and distortion types indicative of classic records?
Or do you want a more modern sound - one with a clean, bright, and upfront sound without the non-linearities of classic gear?
There are more options than this, but these 2 approaches narrow things down a lot. With that in mind, throughout the rest of the video, I’m going to create 2 vocal chains and try to match their sounds for the most part while also embracing what makes them sound unique.
The chains will stay the same, but the processors I pick will vary, and you’ll notice as the video goes on that subtle changes imparted by the processors add up over time.
So, let’s listen to the 2 finished chains, first the classic, then the modern, and then we’ll detail how to achieve these step by step.
Watch the video to learn more >
The first processor to use on a vocal chain is an EQ - there’s a reason why older analog consoles would have channel strips that would start with the mic’s preamp and then go directly to an EQ.
An EQ lets you alter the sound of a signal more than any other processor, setting you up great for subsequent processors.
So, for the vintage sound, I’m going to use this Sweetspot EQ by Pulsar Modular, and for the modern one, I’ll stick with the Pro Q 3.
On the SweetSpot EQ, I’m going to use the section in the bottom left to attenuate the lows before the fundamental - you can also use the HP filter in the top left, but I’m enjoying the sound of the shelf more.
I’ll keep the setting to low shelf, set the frequency a bit higher, which with this plugin is still pretty low, invert the band, and then turn the gain all the way up. It’s a strange setup, but when I invert the gain, I create a cut, then I need to increase the attenuation of the cut by increasing the gain.
In the middle, I’m going to find some lows that need attenuation - after listening carefully, I dipped about 150Hz by 5dB. This plugin doesn’t behave in the way you’d expect, so this cut probably is less than 5dB. This helped lower the fundamental frequency slightly.
Then, in the mids, I’ll boost some of the vocal’s clarifying frequencies around 2kHz with a narrow band. This will help the vocal come forward, but I want to avoid amplifying it too much since my next processor is a compressor.
Moving over to the more modern processor, the Pro Q 3, I found that slightly different settings created a similar sound. Again, I cut the lows, but this time with an HP filter instead of a shelf. Additionally, instead of attenuating the fundamental, I boosted it very slightly. Again, I’ll amplify the clarifying ranges around 2kHz. Strangely, how I equalized the vocal, what I wanted more or less of, changed between the 2 processors.
This is something you’ll probably experience yourself if you recreate these chains and something I noticed throughout the process of mixing the vocals.
Let’s take a listen to the classic and then modern chains with their EQs enabled.
Watch the video to learn more >
Next up, I’m going to compress the vocal - for the classic vocal sound, I’m going to use an 1176 emulation. For the modern sound, I’ll just stick with Fabfilter’s line of plugins and use the Pro C-2.
Starting with the 1176 compressor, a 4:1 ratio will work well - then, I’ll increase the input until I’m getting about 5 to 7dB of attenuation during the vocal’s loudest peaks. From there, I’ll increase the output to compensate.
Now, the important aspect of this compressor is the attack and release - the release will be set to its quickest time of 50ms, whereas the attack can be as fast as 800 microseconds. I’ll start with the fastest time and then back off it slightly until I find the right balance.
Since the attack is so fast, it’ll cut into the transient, causing distortion - the faster the time, the greater the distortion. So blending this attack is almost like determining how distorted I want the vocal to be - the overall effect is subtle but makes a nice addition to the vocal nonetheless.
Moving over to the modern Pro C 2, I’m going to use the vocal algorithm and achieve about the same amount of attenuation. I’ll use a faster attack, but this time, I’ll enable lookahead to capture the transient without distorting it.
The combination of the vocal algorithm, capturing the full transient without distortion, and automatic makeup gain that corresponds to the amount of attenuation creates a forward, cleanly controlled, and detailed sound - indicative of modern vocals.
So let’s take a listen to these 2 compressors, starting with the classic chain and then trying the modern chain.
Watch the video to learn more >
Although saturation can create a wide array of sounds, the 2 main ways to use it are to accentuate clarity or warmth.
I’m going to go with a warmer sound on the classic vocal by using this U-he tape emulation plugin and go for a clarifying sound on the modern vocal with Saturn 2’s clean tube emulation.
Let’s start with the modern vocal this time - so the clean tube setting will add clarity as you gathered, but let’s use a more modern technique and isolate the majority of the saturation to the vocal’s clarifying 1-4kHz range.
I’ll create an envelope follower and attach it to the drive dial of the isolated range - causing more distortion to occur during the vocal’s transient.
On the highs, clean tube settings can create a shrill sound, so I found clean tape to be a good, slightly smoother alternative.
Moving over to the classic vocal, this tape emulation introduces a lot of non-linearities associated with older equipment. I’ll drive the signal into the processor to get some tape saturation, then lower the headroom dial to create tape machine circuit saturation.
Now, notice that this plugin can be used to equalize the signal, but I’m going with a more neutral frequency response to focus primarily on the sound of saturation. 2 important functions are crosstalk and wow and flutter. In most digital processing on the left channel, it doesn’t affect processing on the right, and vice versa, but crosstalk imparts this classic imperfection of older equipment.
Again, in modern production, wow and flutter, or frequency and amplitude modulation, don’t exist, but with older gear, primarily tape machines, subtle modulation was introduced.
So, let’s listen to these 2 distinct processors, first observing the modern chain, then the classic chain, and notice how we’re starting to hear bigger differences between the 2.
Watch the video to learn more >
I want to control the sibilance to keep a balanced sound. A de-esser can also be introduced earlier in the chain, but after saturation is a good option as well.
De-essing can either be done gradually or be isolated to a specific range for a precise result.
Gradually de-essing sounds natural and works better for a classic sound, whereas precise de-essing is more audible.
For general de-essing, I’ll use this Weiss de-esser and only use the high shelf with a small range. In the settings window, I’m going to use a softer knee as well as a longer release. This will smooth out the sound, but in a way that doesn’t seem immediately obvious to the listener.
For the modern, precise sound, the Pro-DS is a good option. I’ll use a single vocal, split-band setting and achieve about 4dB of attenuation. Notice that this plugin also utilizes lookahead - this will capture the entire transient of the sibilance, resulting in very accurate attenuation.
I’ll still use a pretty generous frequency range, but notice how these settings still produce a modern sound.
Let’s listen to soft-knee general de-essing on the classic chain and then the precise sound of the more modern processor.
Watch the video to learn more >
After the vocal is dynamically and spectrally controlled, we can introduce temporal or time-based processing like delay and reverb.
Keeping with the classic and modern chain we’re working on - we have a Roland Space Echo tape delay emulation on the classic vocal and Fabfilter’s Timeless delay in the modern chain.
Let’s start with the modern chain first this time - a 1/16th note delay will work well and thicken the vocal, while some mild ducking will keep the vocal dry during the transient before the delay taps blend into the signal.
Additionally, I’ll add a little variation and instability, as well as some crosstalk - all effects that are indicative of classic gear, but we’ll notice in a moment that the sound is still distinctly modern.
Lastly, I’ll set the width of the reflections and blend the effect with the mix slider.
Moving over to the classic chain, I’ll create a little distortion with the input drive function before creating the same 1/16th note delay with a mild rate offset and some additional width.
The bass of the reflections can be cut slightly, and the treble can be boosted to clarify the vocal slightly. Then, I’ll blend the effect to a similar setting as the modern delay plugin.
Notice that even though we’re using similar settings, the delays sound very different and impart unique characteristics to the vocal.
Watch the video to learn more >
In my opinion, the best way to get a classic reverb sound is with an impulse response of older hardware. This plate impulse response will work well, and if you don’t have one immediately available, there are a lot of free impulse responses online if you do some digging - then most DAWs come with a stock impulse response loader or convolution reverb.
By placing the reverb after our delay, the delay reflections will be reverberated, causing a dense and cohesive sound.
Algorithmic reverb always has a more modern sound to my ear, even when emulating classic gear. With the Pro-R, I’ll pick this Vocal Gold Room preset, then adjust the decay time until it works for the vocal.
I noticed that this reverb is a lot more audible than the IR, so I set the mix a little lower to create similar amounts of reverb between the 2 chains.
Let’s listen to both chains, starting with the modern and then the classic. Notice how, by this point in the chain, the reverb really accentuates the differences between the 2.
Watch the video to learn more >
Ending a chain with EQ gives us the control we need to blend the vocal in with the mix. Due to its placement, it adjusts everything we’ve added to the vocal, reverb, and delay included.
I’ll use this Sweetspot EQ on the classic vocal again and attenuate some of the lows that were added with previous forms of processing.
I’ll dip 400Hz, boost a little of 2kHz, and boost around 16kHz. Then, with the shelf on the right, I’ll add a little air to the vocal to give it a brighter sound. Notice that these changes are subtle when compared to the first insert of the plugin.
Whereas the first insert addressed problems, this insert more or less shapes the vocal.
On the modern chain, I’ll insert the Pro-Q 3. I’m going to use very similar settings with the addition of a small dip around 200Hz, but in a moment, we’ll notice how drastically different these 2 chains are.
So, let’s take a listen and notice how, even with similar processors, routing, and settings, we can get entirely unique sounds when creating our vocal chain. Additionally, use these chains and the concepts as a good starting point for your vocal chain, and know you can deviate from them as much as needed.
As I was saying at the beginning of the video - we have a new membership.