Times ago I started a new album project and had the opportunity to check out the new 20-bit ADAT machine. Actually it's not a new machine but an upgrade of the XT. Eventually, (after the initial furor dies down and Alesis puts out the kits) you'll be able to purchase an upgrade kit for around $700. What's changed is the A/D and D/A converters, a logic card and the front display changed just a little bit. For now, unless you purchased a 16-bit XT in December or January past, you'll have to buy the whole upgraded machine to be able to own one. But there's even good news there. The new machines are retailing for LESS than the original XTs.
 Even though Alesis is ahead of the game in the 20-bit race at this time, it seems they just want to capture the market and not gouge the consumer which is admirable.

 My Humble Opinion

 I did two days of tracking on the new machines. I cut live drums, bass, keyboards and vocals and I've got to say that the difference is remarkable.
 My main complaints with ADATs (beside sluggish syncing) is that transient peaks from cymbals, snare drums and percussion sounded harsh and after a day of recording my ears would be worn out. Whereas with analog I could work for hours and not feel fatigued at all. The effect was not induced by my SPL level, I'm very careful about my levels and pace myself during a long session.

 Using the new machines, right from the first playback I could hear the difference. Cymbal crashes sounded more like breathing than a sharp peak and decay. Vocals were smoother and transient peaks of any kind were kinder to the ear. After the first day's session I felt more like I'd done a day of analog tracking than a digital session.

 The Bottom Line

 After working in the 20-bit arena I don't want to go back. I can't tell you how much better these machines sounded. You get more than your money's worth out of the new upgrade. I think it should be FEDERAL LAW!! All studios MUST be 20-bit or more or have to go out of business.
 

 The Battle of the Bits

 Dateline: 5/26/98
 What's in a Bit?

 A bit is short for binary digit and it is the component of digital recording representing the amplitude of the sound wave. Being binary in nature, the bit is either off or on. That is, it's represented by a one or a zero. The more bits you're having represent your amplitude the better the resolution. Every time the wave crosses the bottom of the bit it's assigned to an "on" designation. See the drawing below for a rough representation as to how that happens.

 The sound wave is on the left. You see two different representations of bit-resolution next to the wave. Th bit "ladder" on the left would not be as accurate as the one on the right. This is because as the wave crosses the bottom of the bit level and turns it on, the volume would jump to the bottom of the next level . With more resolution this jump is much smaller and more natural.

 You can see how having more levels would make the loudness curve of your sound more authentic. When you add more bits your levels of resolution jump exponentially. For instance the difference in levels between 16 and 20-bit is remarkable. In 16-bit recording you have 65,536 levels and in 20-bit you have 1,048,576. That is because for every bit you add you double your resolution.

 Noise Floor and Dynamic Range

 As you add bits you represent the amplitude of your signal with more steps. Not only that, you also increase the total dyamic range available.
 The dynamic range (on paper) of 16-bit is 96dB and for 20-bit it's 122dB. Dynamic range is the range from the loudest to the softest sounds recordable on a digital system. This does not increase headroom, which is the maximum amount you can record but increases the level downward that you can represent digitally.

 What's the Effect?

 The outcome of having more bit resolution is that the rises and falls in volume are much more natural because there is less of a jump between levels. To my ears, things like cymbal crashes, pops from an electric bass, drum hits or anything having transient peaks sound much smoother and more natural. The whole recording sounded more "analog" to me. That's a subjective statement but it's one of the best ways to explain it. My ears didn't tire as easily and I found the whole experience to be surprising and pleasing to my ear.

 You may have heard that you should hit a digital machine as hard as you can to use as many bits as you can. What happens here is that at the lower volume levels the resolution is bad and your signal is overcome by quantization noise. The resolution is better up in the higher range. The problem comes when you have a signal with a large dynamic range and you're forced to record near those lower bits. The increased dynamic range of 20-bit or more represents the lower sounds stay out of the digital mud that's at the bottom of the bit ladder.

 What's Next in Digital?

 If you've been watching the ads and articles in the trade magazines you've seen a big push for 20 and 24-bit converters in digital recording. Pro Tools offers a 24-bit system and consoles like the Yamaha ProMix 02 will handle and even record 24-bit information onto an ADAT machine. This is a good development and will continue on until most of the equipment you buy will be competing on this level.

 The next step will be an upgraded sampling rate. We deal with 44.1k/16-bit sampling in digital recording now. That is the CD standard. No matter what you do in 20-bit land you must eventually get down to the standard. Once DVD-Audio gets sorted out you'll see projects going 20-bit 96k or better from multi-track to finished product.

 Understanding Digital Part 1

 Dateline: 6/2/98

 What's a Sample?

 A sample is a digital representaion of the frequency of a sound. To represent a frequency you need to sample it at least two times a cycle.
 Since our range of hearing is 20Hz to 20KHz we need to be able to represent that digitally. So the CD sampling rate is 44.1k. This means that samples are being taken 44,100 times per second.
      The sampling rate has to be twice the frequency that you want to  record.

 This is called the Nyquist frequency. Nyquist theory states that you can only get half of your sampling rate back in frequency. So the maximum frequency you can record on a CD is 22.05k. You might say, that's plenty since I only hear up to 20,000Hz anyway. Right? Wrong.

 The Reality of Sound

 In reality there is a lot going on above our hearing that effects what we hear below. But this is only one of the reasons that 44.1k is limiting and soon to be a dinosaur.

      When the frequencies above half the sampling rate are introduced to a digital converter it creates a false tone that is audible.

 This is called aliasing. How we get around that is to place a steep filter that cuts off high frequency starting at 20k. This keeps these higher frequencies out of the converters and avoids aliasing. This fix is not perfect. The filters used to cutoff the higher frequencies can handle simple sine waves well but because of the steepness of their slope, they have a problem with complex musical waves. They still create some distortion and odd effects in the cutoff of transients.

 Why Higher Sampling Rates?

 Higher sampling rates solves a number of problems. Put simply it better represents the wave because we have more of it. This of course takes up more memory but as you can see with storage mediums like DVD audio disks, RAM and Hard Drives becoming cheaper and more plentiful, storage is not going to be a problem in the future. Right now there is a major debate going on about where we're going with digital. A lot of people want a standard we can grow into while others seem to have the short range in mind. Keep your eyes here on the latest news on that front.
 
 

 Understanding Digital Part 2

 Dateline: 6/9/98

 The Limitations of the Current Sampling Rates

 Last week we established how any frequency you record digitally can only be half the sampling rate. For instance at 44.1k (current CD sampling rate) the highest frequency we can record is 22.05k. And actually because of anti-aliasing low-pass filters the figure is more like 20k. Analog recorders will record up to 45k. Even though we can't hear above 20k at our upper range, there are cues for stereo placement in these upper ranges that effect our ability to localize sound.

 How we determine the location of sustained sound is by the difference in volume between our ears. However for transient peaks, (sounds with a fast attack and release) we use delay to give us the cues necessary to place sounds in the stereo field. The difference between two sources of sound is called "just noticeable difference" and it is around 5-7 microseconds. When the timing of sounds around us are presented to our ears, we can exactly place where their source is based on this minute timing difference. Our ears are quite the intstruments!

 The problem with our current sampling rate of 44.1k is that the timing difference between samples is 20-28 microseconds. This is not nearly enough to represent how we hear things in the real world. In this case, analog is superior. However with a higher sampling rate we could reproduce sound much more accurately. A sampling rate double our current standard would go a long way in fixing the above mentioned problem. Indeed if it was 4x our current rate it would certainly do the job as well as analog. However there is a storage consideration and this is where DVD-Audio will come into play. Single-sided, double layer DVD disks will store 4 gigs of info compared to CDs now that will only handle 650 megs. We are about to jump into the second generation of CD technology.

 The Move Towards Higher Rates

 There is great support from those deciding the future of digital audio for higher sampling rates. However, the fight that's going on right now between those deciding the DVD-Audio sampling rate and bit resolution may prevent the new standard from going far enough to take care of all the problems we now have with digital audio. There are many financial factors involved of course and different business factions want to exploit this new and exciting storage medium for their own particular interests. This is understandable but there are a number of recording engineers (George Massenburg for one) who are pioneering the idea that we pick a standard we can grow into and that we not be short-sighted concerning a standard that may be with us for decades to come.

 A View of the Pro Tools Certification Course

 Dateline: 6/16/98

 Intensive Training

 This last weekend I had the pleasure of taking the three-day Avid/Pro Tools course 135 for Pro Tools certification here in Phoenix. It just so happens that the school where I teach (The Conservatory of Recording Arts) is a Pro School and is authorized to train others to be certified. Of course I jumped at the chance to get to know more about what I think is the future of digital audio. Not to say that I think Pro Tools has a lock on the future but they certainly have a lot of systems out there and now that they're owned by Avid I'm sure they'll be a player for some time to come.

 Here at home I'm running version 3.2 on an older Mac. This allows me to do 4 track editing with no TDM capabilities. But for the work I do here that suits me well. The system I trained on was the newest software (4.1.1) run on a hefty Power Mac that I had all to myself. There were six people taking the class and six stations so I had maximum hands on training.

 Day One

 The course is set out in three 8-hour days. Day one we got our books which looked like a small phone book in size. The course is laid out in 14 modules (topics) and 8 exercises. The book is concise and well written and the exercises are peppered throughout the lecture (module) portions of the text. This keeps the course moving along well and also serves to keep it very interesting. Modules 1, 2, and 3 deal with

      Pro Tools overview
      Understanding the software interface and
      Making your first recording.

 After this lecture portion we ran a recording exercise. After lunch we then jumped into basic editing parts one and two with exercises attached to both. I must admit that about 5 hours into day one my brain was starting to drip a bit from the overflow but it wasn't overwhelming. The style of the teacher and the way the text was laid out really kept us hanging in their. After day one I literally had Pro Tools dreams that night.

 Day Two

 Day two's modules dealt with:

      Importing/Exporting Audio
      Punch-in and Punch-out Recording
      Working with SMPTE and
      Fades

 The three exercises applied to spotting a piece of film with audio and effects, SMPTE lock-up and using fades. This day went well and although the course was still thick I had a very nice feeling about how things were going.

 Day Three

 Day three's modules covered:

      Working with Regions and TracksManaging your Session
      TDM basics (plug-ins dealing with reverb, eq, delays etc.)
      Audio Suite basics (time compression/expansion, eq and more)
      Automation, Track bouncing and Mixdown

 WHEW!!

 After the third day (which ran overtime) I'd definitely felt like I'd learned a WHOLE lot. The bottom line is that I'm very much excited about using this resource in the studio. For myself I'm still not comfortable with the idea of running an entire session using a system where my only contact with the audio manipulation is with a mouse. I like the idea of hard faders and dials making my moves as I do them. BUT WAIT!..........Pro Tools, Peavey and JL Cooper all have interfaces that will work with the system. Although this is a bit clunky I think it's the right idea. I think if Avid would alter their I/O to allow digital bussing of Auxiliary sends that would go a long way in making Pro Tools more usable.

 What I'm excited about is the prospect of using the Pro Tools platform as a mixing tool. I've been listening to some great records lately and I'm hearing a lot of cool mix trends that absolutely employ automated EQ and panning. For one check out Natalie Imbruglia's "Torn" CD. The end of track 6 at the fade and the intro of her voice in track 8 contains the effects that I'm talking about. I can see how having a system online at the time of a mix would make these effects very easy to do. Just bounce your tracks in sync into the system and dedicate it to the effect. All the parameters on the desk including reverb parameters are automatable in Pro Tools. The certification course is very well put together and I highly recommend it whether you have a system or not.