Education > Digital Audio Workstation

 

Training Waves

Digital Audio Workstation -
An introduction

DAW - The title seems to point at a huge big black box with nice little red lights, a serious row of pro-audio connectors on the back, an ergonomic design with motorized faders and knobs on top, and an integrated monitor that can present the audio waveform, making editing a snap comparable to using scissors & sticky tape…

Not to mention a truly fantastic logo and a price tag that will seriously temper your usual gear acquisition syndrome. Definitely cool. Oh so very cool.

Of course this scenario is possible, the Pro-Audio world with credit to blow does indeed have these boxes and strolls happily along with them, but a close look at the same boxes reveals a striking similarity with an ordinary personal computer. You’ll find the same hard disks, the same sort of operating system, the same way audio is being recorded and mixed. Only the price difference is enormous. Making a studio out of a powerful PC costs a fraction of buying the Black Box, which is the same computer, only with the proper hardware and software already installed. The fact is that the current evolution in Digital Audio Engineering travels so fast that an investment in a dedicated hardware box seems to be a luxurious one, but not wise. Compare it with the first generation DASH recorders, some of them still work with 16bit linear filters, no oversampling or dithering. In other words: During a period of evolution it’s better to invest in a system that allows the user to grow with the flow.

Waves makes tools for these people, tools that keep on growing and are easily updateable via the Internet. Tools that work on Pro-based audio PCI cards and on host CPUs with no need for extra (costly) DSP power. The Waves toolbox is available to everybody, and sounds as good as the high-end analog equivalents.

As a matter of fact, that’s exactly where it takes off.

Analog gear comes in all kinds of flavors: cheap, extremely costly, going for purity, aiming on “sound”... Drums seem to like the “punch” of analog tape saturation; equalizer designs carry names of countries (“British” EQ) and so on. All this difference in price and quality depends and originates from one simple fact: Every piece of analog gear starts as a mathematical design. From that design certain values of components are being calculated, put onto PCB’s and voila! a piece of gear.

Well, it’s not that simple. If an equalizer would have stayed on the drawing board it would actually “perform” very well, but when it enters real (analog) life, it has to overcome a whole series of traumas: poor PCB design with all kinds of cross-talks and leakages which cause the components to interact, component values which are most of the time far from the exact drawing-board calculation. Put this in the perspective that a well-designed EQ can perform as well as the drawing board promises only when the value of the chosen components has a tolerance of less then a tenth of a percent.

Here one can see the overall difference between cheap and high-end audio: cheap uses mass-produced components with values that come somewhat near the needed amounts while pro-audio gear uses hand-picked and ear-tested components. Not only are the last-mentioned much more costly, they also need more attention during the manufacturing stage.

Of course it is possible to buy a very decent small mixing desk for a small amount of money, but when you need a large one your budget has to grow equally.

The great advantage of a digital piece of audio gear is obvious - it doesn’t need to be transformed into parts and PCB’s, it can be an exact reproduction of that drawing board formula. The whole stage of going from a calculation to a board of transistors, resistors, and capacitors disappears since the audio has also become a large stream of numbers. Once properly digitized, all major audio production can be done in a digital environment, with the top of the analog quality as a starting point.

Analog also has some advantages. It can have a far greater frequency response throughout the whole design where digital is limited to a 22 kHz bandwidth. This forces designers to add filters to their designs, filters that are transparent for the passing audio and limit any high frequencies exceeding the 22 kHz limit. But this is true for digital in general and has proven to be no serious drawback. In any case, most microphones don’t produce frequencies above 20 kHz either.

Another advantage of analog is soft clipping and the distortion behavior during very soft passages. When a digital system has used all its 16 bits, clipping sets in and it sounds horrible. When an analog system reaches its limits, the distortion comes in more gradually and can even (in case of tube-driven microphones and tape-saturation) add “warmth”, thanks to the character of analog distortion.

Waves has taken care of both digital drawbacks (soft-clipping and distortion behavior) with the introduction of the L1 limiter. This plug-in has already been on the market for a few years and does two very important things:

  1. It functions as a brickwall limiter, making sure that no clipping takes place. This can be done thanks to “lookahead” imiting, a function which cannot be performed in the analog world since an analog limiter cannot predict its input. L1 can. It does this by reading a randomly accessible datastream utilizing the lookahead function to ensure that no transient distortion takes place and that the attack portion of the sound does not suffer from the usual VCA latency.
  2. On the silent side of digital music, there is the problem of bit depth. When the sound reaches zero-level, only two or three bits are being used to represent the waveform. This resolution is far too low for a good audio reproduction and creates the possibility for the occurrence of serious distortion. Waves overcame this problem with the IDR function inside the L1. IDR is an “intelligent” dithering function - it adds a little noise at a very low level, mirroring the sensitivity of the human hearing curve. By doing so there are always enough bits used, minimizing audio degradation.

Nick Mulder
Department of Registration
Royal Academy, Holland


All content copyright © 1999 by Waves LTD. All rights reserved