Hello and welcome to the Sound On Sound Recording & Mixing podcast channel. I am Eddie Bazil. In this podcast I'm going to explore the joys of using tubes, or valves and tapes in mix and sound design projects.
Before I attempt to destroy any vestige of respect you might have held for me, let me bend your ear as to why we use valves and tape in today's music production projects. Putting aside the history of the emergence of tubes and tapes and how they were and are designed, let me lament a little as to the days of warmth and fuzzy audio cuddles.
Back in the day when bell bottom jeans and flares were mistakenly regarded as cool and did not prompt a batting from the public, we did not have the bat sonar clarity of digital coupled with today's stratospheric headroom. Nope, it would be fair to say that everything sounded thick, warm, fuzzy and in some cases, bordering on the unintelligible. This state was not born out of choice, but was thrust upon us as a direct result of the signal path and recording medium available to us bell bottom folk at the time. Distortion was the bane of our existence and to a certain degree, our saving grace. More on that later. It would be fair to claim that almost everything exhibited, or created, some form of distortion. Even our coffees. Yes, our cappuccinos were warmer in those days.
Leaving device noise and unintended distortion as a side effect of design limitations out of the warmth equation, certain hardware devices were intentionally designed to generate harmonic distortion. In layman's terms and keeping it broad stroked and simple, distortion is the result of any process that alters a waveform by adding new frequency components that are harmonically related to the source sound. Loosely, this is where the term harmonic distortion springs from. Almost every modern day common distortion term like overdrive, saturation, excitation and so on, falls under the distortion umbrella. The type of distortion and how it is generated is integral in understanding what colour or flavour will be achieved using the various types of distortion processes we have available today and the two topologies I use above all others are tubes or valves and tape.
Nowadays we have clinical clarity at every stage of the recording and mixing process. This is primarily down to the fact that we record to a digital medium using a signal path that is digitally governed. Once the analogue signal is captured, the processes governing and processing it are, in the main, digital. We didn't have that luxury with hardware mixers, dynamics and effects driving signals to a non-linear magnetic medium. At every stage we were faced with sonic compromises that we had to navigate. The earlier sound cards were sonically archaic compared to the budget sound cards of today and to achieve a similar quality you had to part with a lung to move up the quality converter ladder.
Nowadays the analogue to digital and digital to analogue processes are so good that hardly anything is lost in the processes. Technology has exponentially improved every stage of the signal source to destination journey. Using algorithmic based systems we have full control over the integrity of the signal and its manipulation. We do not need to embrace religion and pray for things to work out anymore. Clarity, extended bandwidth and headroom suddenly changed how we heard things. In the bell bottom days mistakes would often get swallowed up in the myriad of distortion artifacts, good and bad. But nowadays, the digital domain is unforgiving. Your vocal recording has a tiny click, the digital system announces it as if it's the Super Bowl halftime 80,000 watt artist performance.
But what was our Achilles heel has now become our saviour. Distortion and in particular, harmonic distortion, has become such an important aspect of the sound and mix colouring processes that without it, you feel almost incomplete. And there's a reason for that. The signal clarity we have today can come across as clinically staid. We need harmonic distortion to infer a hairs on the neck sonic response. We actually miss the imperfections of yesteryear's sound, much like the emotive musical experience of listening to vinyl over CD. I remember when CDs entered our engineering lives. No more losing fingers splicing tape. No more life sucking hours spent calibrating tape machines while spinning rosary beads. No more press, record and hope scenarios. With CD we suddenly had massive dynamic range to explore, pristine detail and an overall representation of sounds that we could hear in their full glory.
We were joyous and then bang! Quality, flexibility and the manipulation of audio leapfrogged a few light years thanks to advances in technology and we were now faced with clarity and expression of sound to such a high degree that you could hear a gnat moonwalk on concrete. In some instances that is a wonderful situation to be in. Classical and acoustical music benefits greatly from detailed clarity and the extended dynamic range. But does it address the emotion conundrum attributed to the listening experience? That's not for me to judge. What is relevant, in the context of this podcast, is that distortion provides fairy dust for both individual sounds and entire productions. It is the modern day vinyl experience, albeit in pristine quality.
In the context of mixing and producing music, tubes fall into two distinct types, triodes and pentodes. Triodes generally produce more even ordered harmonics, whereas pentodes produce more odd ordered harmonics. You'll find conflicting generalizations on the internet about generated even ordered harmonics being rich and warm and odd order harmonics being bright and clean and vice versa. I prefer to ignore these generalisations as I believe harmonics are context driven and the source sound and processing applied will determine the colour classification. To delve into the world of harmonics would require a few podcasts and to be honest this podcast is about using harmonic distortion to warm up sounds and mixes.
However, to understand what type of processor to use, it helps to understand what harmonic content a sound has that can then be highlighted or added to, etc. A decent spectrum analyzer like the free Voeango SPAN, configured to display harmonics, can give you a great deal of information about the harmonic content of a sound. Run a sound through it and analyse the harmonic content. You might find a few surprises in there. And when it comes to sounds that use basic waveforms, it helps to understand what harmonics these waveforms have. Use SPAN and a test tone generator and study the various responses. A sine wave has a fundamental frequency, which is also referred to as the first harmonic just to confuse you even further, but has no harmonics. A triangle has odd harmonics, a square has odd harmonics and a saw has odd and even harmonics. This is useful information, it can help you decide what distortion process to use and whether you want to highlight, attenuate, or generate harmonics.
I'll put this information to use in the first example. Rule of thumb with all tubes is the more you drive the tube, the more distortion you get. That's fine for individual sounds, but for mixes you need to be more conservative, as overdriving an entire mix can result in a muddy and congested texture. If you consider what a Roland TR808 sub kick is, you'll understand why we use distortion to highlight the sub. The 808 kick is a low frequency sine wave and therefore contains no harmonics we can highlight. We need to force the sine wave into generating harmonics. In effect, we're changing the waveform's attributes. In example 1, I'm using Softube Summit Audio TLA 100A Tube Solid State Leveling Amplifier. First, the 808 sub with no processing. And now the processed version. I have to say the TLA 100A is a weapon when it comes to processing basses. It does the whole Telectronics LA-2A thing nicely, but to my ears it sounds different and is perfect for very specific sounds, most notably basses, electric or synthetic. I'm using a slow attack to allow the subkick's attack transients to shine through without any peak reduction. I've set the saturation value towards high, but I've been careful to use the wet dry mix, i.e. the parallel mix, to allow for a more gentle mix of saturation and drive. Although there is some hefty peak reduction taking place, I've used very little drive. I don't want an overblown distorted texture. The result is that the sub kick has a nice fuzzy roundness to it and bounces along nicely thanks to the attack and release settings. And coupled with peak reduction, it swells and deflates sweetly.
In the next example I'm going to use IK Multimedias White 2A tube levelling amp to warm up a nice mix. First, the dry version. And now with the White 2A. Out of all the LA-2A emulators, IK Multimedia's White 2A is my favourite. Being an Opto means it reacts slowly and that always helps to let attack transients through before the clamping down takes place. The tube colour is simply gorgeous and I have it set to limit instead of compress as I've placed it on the master bus. It is affecting everything in the mix. The gain, or drive, is set to the same value as the peak reduction, i.e. 40. This gives a well balanced gluing effect to the mix and allows the generated harmonics to shine through. I don't like overblown tube distortion at the master bus. When driving tubes, it pays to be gentle and continually A B between dry and processed to make sure nothing is compromised at the channel stages. The mix sounds bigger and vibrant without coming across as congested and muddy thanks to the conservative settings.
In example 3, I'm going to ruin your day. I'm going to use Soundtheory's Crafter. Granted, it's not tubes or tape, but because I care for you, I have to include it. Crafter is a saturator that not only concentrates on managing audio integrity, but generates harmonic distortion that was clearly conceived in the land of deities. First, the unadulterated version. And now with the deity application. I'm a huge fan of sound theory. The Gullfoss plug-in is literally used in every one of my projects and the latest offering Crafter is a continuation of their genius. It might be mistaken as only a mastering tool, but it isn't. It is as effective on channels, groups and the master bus. I'm using a gentle knee and low offset value for the saturation drive response. Using it in a single band mode with very little shift values, the saturation effect is kept to a minimum. The clipper at the output not only controls the output gain, but can be used as another stage of processing. The combination of these simple settings affords a beautifully controlled and expansive texture that does not hinder any frequencies at the channel stages. Most saturators that are combined with clippers give great results, but with some compromises. Crafter navigates all the regular pitfalls like aliasing and oversampling into modulation distortion like a Jedi. The level of control over the integrity of the audio that gets punished with saturation is sublime. The colour, well, listen for yourself. It opens up the whole mix and takes us back to cuddly audio days.
Okay, enough tubes, let's now move on to tape. Ah yes, tape. Most engineers my age will have had to work as a tape operator. My bad, Studio Coffee Maker. When first starting out interning at commercial studios, it was a responsible but arduous job. However, it did teach us a great deal about how tape machines work, how to use tape and more importantly, how best to coax a great recording out of such a compromised and non linear device and medium. Don't be fooled into thinking that we aimed for colour when using tape machines. On the contrary, we did everything we could to achieve a clean and transparent recording. I have no desire to visit that era in my life again and I'm so glad that software developers have furnished us with software tape emulators. All whinging aside, tape and in particular tape saturation does impart a certain characteristic to audio that we seem to like. But it's not just tape saturation that colours, the very nature of the medium also has characteristics that can be good or bad. A gentle rolling off at the top end, or the head bump at the low end, can actually be used to shape and hone specific frequencies. For example, reverbs and delays sound very different when printed and played back to and from tape. The very smearing properties and top end roll off can give a thicker and denser texture to reverbs that digital cannot achieve without some work. The stretching and contraction of tape and the controlled tape heads behaviour can also give a pleasing response to a sound. It's not an accident we have tape delays. And on Inception, they gave rise to new genres. They sound markedly different to their digital counterparts. Even Head Bump can be used to achieve a certain low end colour. And it doesn't end there. The tape machine itself has some hidden wonders that can add depth and width to a sound. And that's apart from loop and playback control to create specific effects. It's interesting that the limitations and flaws of tape machines and the characteristics of the magnetic tape medium have lent themselves to some innovative sound design and mix processes.
In example 4, I'm going to use IK Multimedia's TEAC A6100 MkII tape machine to process the electric bass line from the earlier puppet track. First the dry version. And now with the TEAC. The question as to whether tape emulation plug-ins sound and behave like their hardware counterparts is not a discussion I want to enter. I view tape plug-ins as effects and dynamics and use them accordingly to colour and dynamically shape certain sounds. The settings I've dialed in are specifically for a low end type of sound like an electric bass. Low frequencies and tape are a processing juggling act and can be easy to oversaturate and smear frequencies. So you need to plan each setting in preparation for the next setting. I always start with the bias setting. Bias is used to reduce overall distortion levels. A low bias setting will result in more distortion and a higher bias setting will result in less distortion. But that's not the end of it. Intentionally over biasing is a common technique used to produce a gently saturated sound resulting in a rounder and warmer response. Under biasing can also be used to smooth out high frequencies and add distortion to low frequencies. The effect is very context driven, so play around until you achieve the color you're after. I haven't gone crazy with the bias setting, instead experimenting with different values whilst listening to the bassline, and I settled on a halfway compromise. I've switched on Repro, which passes the signal through the entire analogue recording system. Input to record amplifier, to record head, to tape, to playback head, to playback preamplifier to output stage. You can disable this if you want a simple input to output signal without the tape getting involved. Transport mimics the behaviour of the mechanical transport and the resultant small irregularities add to some interesting textures to the signal. Opting for stereo allows the slight differences between the left and right channels to shine through which equate to a wider stereo field. I've set the playhead to cut a lot of low frequencies out as things can get a tad overblown when saturating low frequencies. Adjust this to taste. Finally, and this is as important as the settings used, I'm using the 499 tape. This allows for less compression and distortion when driven and pronounces higher frequencies, which is perfect for sounds of this nature. The result, though subtle, is a deeper, wider yet unnervingly clean bass texture. Gorgeous.
In the next example I'm going to run a short snippet I created through IK Multimedia's TEAC A3340S 4 track tape recorder and also the Tape Machine 24, an emulation of the MCI JH24, using two different tapes so you can gauge how different they sound. First the dry version. Now with the TEAC. And finally with the MCI. With the TEAC I'm using the 911 tape which is an emulation of the BASF SM911 tape which claims to offer warmth and precision. To be honest, I prefer to let my ears make these textural decisions. Select the tape and listen to how it sounds. Done. I'm applying hefty Bias to keep things a little less muddy and coupling this with a high frequency EQ boost. This is on the record side of things. On playback I've selected to moderately cut low frequencies. Transport modelling and true stereo are also engaged. With the MCI I've duplicated the record and playback settings to offer you a ballpark A B comparison, albeit with different machines. I've selected the GP9 tape. This is modelled on the Quantegy GP9 formula and sounds clean and punchy with less distortion and compression. A perfect compromise between analogue and digital. They both sound excellent but with subtle sonic differences in both colour and control. I'll let you decide which you think sounds better.
In the final example I want to demonstrate how to use harmonic distortion to enrich staid synthetic sounds and in this instance a sample based two dimensional piano. First the dry version. And now the enriched version. The trick with multiple distortion processes run in series is to be very conservative with overdrive values. Otherwise, accumulatively, you could end up with a smeared and muddy response with no definable harmonics. I'm using Waves AFX 204 Exciter to gently entice the processor to generate some nice sheen like harmonics. By leaving input and output at zero and using an exciter mix value of just over halfway, the generated harmonics are very subtle. This processor is then fed into one of the best structured saturators available, the SSL X Saturator. This beast is an emulation of an analogue circuit that introduces either second order valve style or third order transistor style distortion or a blend of the two. I've set the blend value to around 60 percent as I want to generate both second and third order harmonics but leaning more towards third order. Depth which controls the distortion amplitude and the wet and dry mix are both at around 65 percent. I've used the shape function and steered the distortion edge towards a smoother response as I don't want to give a hard edge to the generated harmonics. This is then fed into SSL's FlexVerb. Reverb is one of the primary tools we use to invoke depth, be it on a sound or a mix and I've shaped it to gently smear a broad range of frequencies to allow the distorted response to have a three dimensional quality to it. The result is very subtle but you can hear the piano has more depth and sounds richer. It moves along nicely thanks to the chain of processes applied and gives the impression that it has a front to back texture whilst moving ever so slightly and in tiny arcs from left to right.
Harmonic distortion is not just about generating harmonics to hear. If used sensibly and with other time based effects and dynamics, it can invoke a deeper, wider, emotion rich texture to any sound. I barely touched on the power and use of different types of distortion topologies. I've simply selected a couple of types of distortion and created a few examples to highlight what can be achieved both in mix contexts and as sound design exercises. Try to understand the basics of what they do and how they do it. Once you do, tou will open up a whole new world of goosebump moments. That's it for now, thanks for listening. This has been Eddie Bazil for Sound On Sound.
Thank-you for listening and be sure to check out the show notes page for this episode where you'll find further information along with web links and details of all the other episodes. And just before you go, let me point you to the soundonsound.com/podcasts website page, where you can explore what's playing on our other channels.
Transcript source: Provided by creator in RSS feed: download file