Shuttling Through the DAWs

Fast forward a trifling four years to 1998, AudioTechnology Magazine had hit the streets and hard disk recording had advanced exponentially — literally. Issue One of ‘AT’ included reviews of various digital products: a digital console from Yamaha, a digital processor from TC Electronic, a digital multitracker from Sony using the short lived and now utterly extinct MiniDisc platform, along with a bunch of more traditional analogue devices. Remarkably, Issue One featured reviews of no less than four DAW systems,  with the big news being for each that they could tackle 24-bit recording. At the time, 16-bit recording was dying a quick death, it simply wasn’t precise enough for capturing audio in a professional sense any longer. In fact, at the time it was common practice to record to high maintenance two-inch tape, then transfer the tracks to a DAW for robust editing and mixing. 24-bit digital recording changed that paradigm completely, and is still the standard today. Without question, 24-bit recording was one of the biggest ‘game-changers’ for the digital recording field.

So what were those four DAWs? Four of the big names in DAW history; Digidesign’s ProTools 24, Emagic’s Logic Audio (now part of the Apple behemoth), Steinberg’s Cubase VST, and the progressive/unique, and now extinct, Paris system — a cooperative effort by Ensoniq and the late Stephen St. Croix (not to be confused with the vigorous Steven St. Croix). With the exception of the Ensoniq Paris, these three DAWs are still the big names of the field to this day, however in the ensuing 16 years there’s been dozens of newcomers to the landscape. We’ll look at some of the more prominent manufacturers and their contributions to DAW development soon, but firstly I should point out the two primary paradigms behind DAW design.

TWO OF EVERY KIND

Perhaps obviously, the reasoning behind a DAW is to record multiple instruments, and to facilitate overdubbing. As the DAW has evolved, there has been two paths of development, one being from a tape machine and mixer model, the other from the electronic instrument and sequencer model — I’ll explain the difference. With the emergence of the multitrack tape recorder, tape machines gradually evolved to record up to 24 monaural tracks across a two-inch tape width. For decades the general studio hardware regime included a mixing console and a multitrack tape recorder. As magnetic tape technology evolved, more tracks were able to be crammed into the available tape width. First two tracks, then four, eight, 16, and then 24 tracks. Then, of course, the recording can be edited by physically cutting and splicing back together pieces of tape to rearrange and possibly repair sections of a performance. Some DAWs follow this model. The major distinction of DAWs following this model is that musical sections are measured in time — typically a SMPTE-based timeline — just like linear tape, and edits are conducted according to hours, minutes, and seconds. ProTools is no doubt the most notable DAW to follow this design, and for many years, ProTools was an audio recorder, with mixing and editing facilities only. This approach has stood ProTools in great favour with professional recording studios over the years as it more closely replicated this traditional recording methodology.

There’s of course another angle of approach to DAW design. During the 1980s, electronic music production gained traction due to the groundbreaking rise of MIDI — the Musical Instrument Digital Interface. With MIDI, musicians could sync multitudes of electronic keyboards and rhythm machines together. Instead of recording each and every sound source to multitrack tape, hardware sequencers were used to record the note events of each instrument. Alongside relatively inexpensive cassette and reel-based multitrack recorders, sequencers could provide additional tracks of audio performance. This was a boon for home recordists and composers as they could piece together multitudes of audio tracks using electronic sound sources such as synthesisers and samplers, and their ‘live’ sources such as guitar and vocals could be recorded to an affordable multitrack recorder.

The inevitable next step was software-based sequencers. Two of the bigger names at the time were Cubase from Steinberg, and Notator from C-Lab. Both applications (or ‘programs’ as was the parlance of the day) ‘recorded’ MIDI events to be re-triggered in sequence. As time rolled on Notator evolved into Logic (as C-Lab became Emagic), and by the mid 1990s, when hard drive speeds had suitably fast read and write speeds, both manufacturers began to integrate audio recording into their flagship sequencing packages. What makes this an alternate design approach to that of the linear tape concept is that ‘recorded’ musical sections adhere to ‘bars and beats’, with editing adhering to the bars and beats of a musical piece.

These days, the lines between these two paradigms have blurred immeasurably, with almost all DAWs able to operate in both bars and beats or divisions of hours/minutes/seconds. There’s no right or wrong between either — it’s simply a matter of where your background and training lies and what one prefers — musically trained people tend to favour bars and beats, whereas studio-trained ‘engineering’ folk will tend toward time-based editing. Though the distinction is pretty blurry at this point in time.

One particularly dark horse to completely smear this delineation was Ableton’s Live. Birthed in 2001, Live was a return to bars and beats sequencing and the clip-based style of music creation originally seen in C-Lab’s Notator and Creator software on the Atari ST platform, only with the amenity to pull off similar compositional stunts with recorded audio, virtual instruments and plug-ins. Live has a huge following these days and justly so — it filled a gaping hole in the market at the time; a DAW aimed at performance and composition rather than recording and surgical editing and mixing. Another roughy to make its way to the head of the pack was Image-Line’s FL Studio, which began life as Fruity Loops before metamorphosing into a fully-fledged performance-based compositional tool and DAW. Both DAWs were and are game changers, simply because they brought a raft of flexibility and a completely new crowd of composers into the DAW fold.

HANDS ON

By Issue Two of AudioTechnology, the first incarnation of an undeniable DAW game changer had reared its head, and with it came a number of changes to the audio industry that shifted the way composers and engineers work. In the preceding five years or so, a great number of studio outfits and composers began to latch onto the power of computer-based mixing and editing. By this stage it was possible to run a fist-full of natively driven (i.e. powered by host CPU processing) audio plug-ins, with the more affluent and professional waveform-tamers opting for DSP-powered plug-ins such as those offered by Digidesign, Creamware, UAD, and TC Electronic/Works. You see, with all this new-found DSP power, a game changing aspect in itself, people began ditching their mixing consoles completely. Mixing and processing ‘in-the-box’ became an extremely effective option as it offered complete control over a mix, and often, more importantly, complete recall of a mix. It was as easy as loading a session file. No need for photographing the console or shelling out for a console that offered automation and recall. It also gave studio operators and composers more time for audio work rather than the time-soak of maintaining outboard mixers and the kilometres of cabling that went with them. It also provided the luxury of keeping multiple projects on-the-go at once, which again lent itself to more output and further profitability. The in-the-box revolution was a game changer in itself, yet there was an intrinsic snag with in-the-box mixing; unlike mixing with a console where you could access multiple volume levels and EQ settings spontaneously, DAWs offered control over merely one parameter at a time — the limitation of a single computer mouse pointer. For the many who’d developed mixing skills on consoles with immediate access to hundreds of options at a moment’s notice, this was a colossal hinderance. Enter the Mackie HUI.

HUI was the first ‘Human User Interface’, and was a joint effort between budget console manufacturer, Mackie, and Digidesign, the creator of the now well established DAW, ProTools. HUI was, in itself, an engineering triumph, utilising complex MIDI data to allow a traditional set of knobs and motorised faders to control the virtual knobs and faders within a DAW. For the first time a ProTools operator could quickly and simultaneously alter the level and pan of eight channels within a DAW, instigate and edit plug-ins, record channel automation and have access to palpable, physical transport controls, all without resorting to the ever-so-restrictive mouse and pointer.

The HUI made all the difference. Suddenly mixing without a console made a lot of sense, and the bottom fell out of the small footprint mixer market for a good while afterwards. In its wake came countless monitoring systems like Mackie’s own Big Knob — devices to provide attenuation over a DAW’s monitoring output, along with headphone outputs, talkback, and monitor switching — all the things people held onto their analogue consoles for. A procession of summing mixers also proliferated at the time — for those who missed the sound of a console feeding summed audio from various audio interfaces into a two-track recorder — but this fad lasted only until DAW manufacturers got a solid handle on digital summing. These days nobody seems too worried about digital summing.

Not too long after the physical HUI debuted, Mackie’s HUI protocol became a de facto standard for control surfaces, as they became known. Shortly thereafter, Mackie released its own control surface protocol, ‘Mackie Control’, designed for use outside of the Digidesign/ProTools ecosystem. Consequently a stream of control surfaces hit the market — from Mackie, Tascam, Novation, Alesis, SSL, and of course, Behringer. The upshot was a massive move toward in-the-box mixing. It’s around this time a lot of vintage consoles were sold off at crazy you-can-have-it-if-you-move-it prices.

MEAURE TWICE, CUT ONCE

Once outboard, tactile control surfaces began to, erm, surface, DAW manufacturers began focussing on mix automation. Some would recall the pseudo-standard practice of composing in Logic Audio, Cubase VST or Digital Performer, then bouncing and exporting all the tracks as audio files over to ProTools to mix, as the automation facilities of ProTools left the other applications for dead. It was a pain in the proverbial but it provided superb results. Most of the ‘bars-and-beats’ style DAWs offered automation which was merely an extension of MIDI automation kludged to handle integrated audio tracks. With 128 steps to a fader this simply wasn’t cutting it for professional automation. You could hear the steps — ‘zippering’ was the term used. ProTools, on the other hand, offered 10-bit precision (1,024-step) faders — it sounded smoother, reacted faster, and provided the style of automation found on professional consoles by the likes of Neve and SSL, such as read/write, latch, and trim.

BRAND NEW SAMPLE BAG

Technological progress is most often about doing more with less equipment. It’s about killing off bespoke hardware and relegating the work to software within the increasingly-influential desktop computer, and nowadays, tablet. Yet another piece of audio gear flung to the roadside in the DAW’s wake was the dedicated hardware sampler. Sampling, the process of taking a snippet-sized digital recording, storing it in RAM and replaying it with a keyboard was a significant milestone in modern music production. During the 1970s, scarily expensive sampling instruments began their march into the mainstream such as the Synclavier and the Fairlight CMI. By the 1980s, Emu, Ensoniq, Akai, and Roland had the concept in hand with affordable machines that went on to change the face of music completely, even sparking genres that relied entirely upon sampling.

There was, however, a major downfall with hardware samplers. They were restricted in the amount of audio they could record and replay because of the amount of RAM they could have installed and access — most were hemmed into 32MB (yes, that’s megabytes) although Kurzweil took its K series workstations through to 128MB. With memory prices decreasing and hard drive space increasing, it was inevitable the hardware sampler would fall prey to the desktop computer.  In late 1998 we saw the NemeSys Gigasampler, a software-based sampling instrument that read samples directly from a hard drive rather than storing them in RAM. Suddenly vast libraries of previously unheard of instrument sizes were commonplace. Entire pianos could be multi-sampled — note-for-note rather than spreading 30-odd samples across an entire MIDI keyboard to save on RAM space. One caveat was the advised minimum tech specs — at least 2GB of hard disk space — a relative aircraft hangar’s worth of storage for the day. It wasn’t long before manufacturers began integrating their own ‘soft-sampler’ plug-in instruments within their respective DAWs, and then the bum fell out of the hardware sampler market completely. I literally threw six grand worth of Akai samplers on the bonfire one afternoon — I really should have sampled the sound of those chips exploding.

Alongside this soft-sampler revolution came some clever sample replacement techniques. Before soft-samplers, the regime was to place MIDI notes as precisely as possible where your recorded kick drums (for example) sat throughout a track (some DAWs offered the ability to place a MIDI note on a waveform transient, some didn’t), then load your replacement sounds into a hardware sampler and trigger those from within the DAW. This process became far easier with soft-samplers such as Drumagog and Digidesign’s (now Avid) TL Drum Rehab.

Equally as game-changing was the combination of soft-samplers and the exponential and rapid growth of hard drive capacities. In true software industry fashion, soft-sampler libraries grew in size as quickly as hard drives could accommodate. It’s now commonplace to have terabytes of sample libraries at our disposal rather than shelves full of sample CDs (remember those?). DAWs now arrive with countless sample-based instruments already installed or a download away. Perhaps one of the biggest advantages for musicians was access to huge orchestral libraries — world class orchestras recorded in world renowned venues — negating the need to hire dozens of musicians and huge studios to create orchestral soundtracks. Nowadays we have thousands upon thousands of samples and software sample players at our disposal — from simple drum replacers through to note-for-note multi-sampled grand pianos, vast synthesis libraries, virtual drummers, phantom guitarists, bouzouki players. More stuff. In-the-box. Ready-to-roll.

SLIP & SLIDE

With magnetic tape and multitracking, there came an age where it became possible to edit a recording. No longer was a recording simply a facsimile of the best take on the day played by a group of musicians in concert. Tape provided not only the option to replace or add single performances to an already established recording, it was also editable — sections of a performance could be cut out and placed elsewhere within the recording’s timeline. By comparison, editing digitally-recorded material is extreme-sports hyper-editing. Not only is it editable to the nth degree, it’s always reversible, un-doable, and completely ‘non-destructive’.

Anyone who’s dabbled even slightly with a DAW can see the propensity to get very lost in this maze of options. It can and does take years of trial and error and/or training to understand how to keep a performance intact, on track, and seemingly realistic during the editing and mixing process.

Where this becomes incredibly complex is when editing across multiple tracks of a single instrument such as a drum kit — each microphone’s recording must occupy a separate track within the DAW. One method now used by many DAWs is to fold all these tracks into a single region or folder. Logic Audio/Pro had been doing this for years with MIDI parts so the progression to this regime with audio files was a natural one. Cubase also introduced folders, a similar approach to Logic Pro whereby tracks could be consolidated into folders. Digidesign’s ProTools took a more tape-oriented approach — that of creating editing groups; assign particular tracks to an edit group, then an edit on any region within the edit group would edit other audio regions assigned to that group at the same points. For all these DAWs there has been constant improvements and baby steps along the way, and many of the later entrants to the market have taken from these cues and arrived at similar methods and combinations, removing a lot of the confusion from digital audio editing.

Another more poignant use of multiple tracks is when recording various takes of an instrument — and that includes vocals. Say you have the guts of a song recorded and you’re calling in the guitarist to record a killer lead track. It’s best to get a few good (or hopefully great) recorded takes so you can pick and choose the best sections to jigsaw together when the guitarist downs tools and heads to the pub. The same for vocals; record a few great takes so you have a certain chance of constructing what sounds to be one fantastically contiguous recording — known as a composite. Errors of pitch, timing and diction are then disguised, eradicated, and replaced. It’s how modern records are made.

So how does the budding, or even seasoned recording engineer keep tabs on this shambles in the making? During recent years, many DAW manufacturers have developed comping methods with specialised ‘comping’ tracks. ‘Comping’ (or creating composite tracks) during the early days of the DAW required a bit of lateral thinking as there were usually no dedicated procedures for the process. The early years of ProTools saw users utilising the DAW’s voice stealing allowance feature. Much like playing a mono-voice synthesiser, the operator can set a number of tracks to play from a single voice allocation, then cut, splice, and selectively mute sections on the separate tracks knowing only one track would play at a time. Nowadays ProTools uses a method known as ‘playlisting’.

Cubase and its variants from Steinberg used audio ‘lanes’ for years — an additional hierarchal lane within the audio side of the DAW allowed sections of various tracks to be grouped together and output from a single channel of the DAW’s software mixer. This is similar to how newcomer to the field, Presonus’ Studio One, deals with comping, however the term ‘lane’ is exchanged for ‘layer’. Digital Performer from MOTU uses the term ‘takes’ for comping duties, and offers a way of quickly selecting a range within a take, with that range immediately playing back from the audio output channel assigned to the original record channel. Logic Pro’s methodology is remarkably similar to Digital Performer. A quick range selection with the mouse pointer over a take allows that selection to play from the originally selected record channel. All in all, these systems are much the same — the point is these comping tools are now part and parcel of modern DAWs, making digital comping far easier than it’s ever been.

BEND IT LIKE BECKHAM

Throughout the DAW’s development there have been many ways to alter the length of a recorded waveform to force it to fit into a particular timespan. Originally these methods revolved around the concept first seen in digital sampling instruments — that of ‘time stretching’ — compressing or expanding the length of a recorded waveform. This combined with ‘pitch-shifting’ allowed engineers to force-fit countless perceived errors and mismatches into a musical piece without the need to re-record. In the ’90s this was very much a hit or miss affair, with the resulting audio often being riddled with harsh, telltale audio artefacts — but the concept was a promise of what was to come.

Nowadays, thanks to incredibly fast CPU processing, every DAW has the facility to warp and twist audio in both pitch and time, and shoehorn sub-acceptable audio into listenable and usable takes. ProTools offers ‘Elastic Audio’,  Logic Pro has ‘Flex-Time’ and ‘Flex-Pitch’, basically, all DAWs offer this type of functionality now. The difference from the early days of the DAW and now is these warping algorithms sound extremely persuasive, with little to no audio artefacts present in the altered recording. All this can be done in real-time, non-destructively, while you listen to the resulting alterations alongside the remaining tracks for reference — a game changer of the highest order.

ALL DAWS LEAD TO ROME

In the last five or so years there’s been plenty of newcomers to the DAW landscape. With the array of choices it can be difficult to know which is the best platform to spring for and from. Do you start with something based on bars and beats such as Live or even Propellerheads’ Reason, or does your production roadmap lend itself to a more traditional tape-based ideology such as ProTools. Either way, the resulting landscape means it costs peanuts to get creative with audio now.  The fork in the road here was no doubt Cockos’ Reaper, an extremely capable DAW that was initially free to download. These days it can be owned for as little as $60, and if you’re using it commercially Cockos asks you very politely to pay the full price of $225. It’s all somewhat of a democratisation of the playing field, to the point where even Apple has decimated the price of the extensively developed Logic Pro from $2600 to about 200 bucks. Sure, other DAWs cost more, and if you want the particular features and working methods of those applications you’ve no choice but to pony up. But now anybody can get their hands on an audio production platform for a relative pittance, if not for free. Game changing? Most definitely.

BEYOND

So where does all this end I wonder. In 1998 I couldn’t have predicted the majority of the innovations brought to fruition in the last 16 years. Will every piece of audio-related hardware be perfectly modelled and available at our fingertips via a projected touchscreen? Will a completely professional DAW be available as a download on your ‘smart’ television anytime soon? Will audio content producers eventually bring entire DSD-based sessions with countless tracks into fit-for-purpose, recording rooms on a tablet or their phone? That said, will the need for an acceptable recording space evaporate as DSP algorithms are devised to recreate those spaces. One thing is certain; the game will change, and along with it, the way we play.