The MIDI Page: Samplers: You Still N-N-N-Need One!

by Todd Albertson

MIDI synthesizers in your production studio are useful and fun. Boy, are they fun! Samplers used to fall into that category as well, but recently, they have started to be seen as a basic requirement for CHR, Urban, and Dance format production rooms. Furthermore, samplers have now begun to creep into nearly all other formats as well. The types and capabilities of these valuable instruments can be somewhat overwhelming. After all, we are talking about everything from Ensoniq's "Mirage" to New England Digital's "Synclavier". (Rick Allen, you lucky dog!) Let's look at what a sampler really is, and some of the ways one can make use of this formidable tool.

In geometry, you learned that a line might be characterized as a series of points. Extending that concept, imagine waveforms as a series of points on a graph. In other words, in the wave's crest we might pick a point that is exactly, say, 32 units above our "zero line". A point in the following trough might be measured at MINUS 32 units below our zero line. Now then, if one were to look only that those two points in the waveform, we would have taken two "samples" of the waveform. If we took one sample every half second, our "sampling rate" would be 2 Hz, or two samples per second.

In samplers and digital recording (such as RDAT), the sampling is handled by an Analog to Digital Converter. This A/D circuit looks at points along a waveform at a given rate. What is the minimum rate needed? The answer is supplied to us through the Nyquist Theorem, which (in short) states that the sampling rate must always be AT LEAST double the highest frequency present in the waveform. Human beings are said to hear frequencies from 20 Hz to 20 kHz. This is not actually the case except for VERY exceptional individuals (usually women), but nevertheless, it is the accepted norm. Remember that to correctly sample sounds including 20 kHz, we must use a sampling rate twice that high. In other words, 2 x 20 kHz = 40 kHz. For this reason CD's are manufactured using a sampling rate of 44.1 kHz. Notice that this sampling rate provides further additional ceiling above the range of human hearing. When we work with samplers however, the unit's filters are designed to remove frequencies that are too high for whatever sampling rate we have chosen. Since the FCC has already selected a high end limit on your station's frequency content, your sampling rates can be lower than full bandwidth. This has the advantage of saving memory, thus allowing longer samples.

What happens to the sound information once the A/D converter changes it into numbers? Obviously, it must be stored in some form. You have no doubt recognized that these numbers are no different than any other computer code, and thus can be saved in a variety of ways, including floppy and hard disk drives. RDAT machines store the digital information on little cassettes that look like miniature VHS cassettes. Samplers in general, store the digital information onto disks of some sort, where it can be reloaded into RAM memory, for playing or editing. Once the sound data is in RAM, the fun begins!

Samplers are the most amazing production tools yet, because they allow you to cut, edit, and process sound in its digital state, thus eliminating all noise, and tremendously increasing the SPEED of editing. For instance, the "E-Mu" products (Emulators, EMAX, etc.) are wonderfully easy to use, and permit the operator to manipulate the sound in all sorts of interesting ways. After the sound has been loaded into RAM, with the touch of two or three buttons, one can reverse the sound, change its pitch (without affecting the length!), splice it to another sound, combine it with another sound, MODULATE another sound, and on and on... The implications for Radio production should be immediately apparent! After processing, the digital information (numbers) can be passed to a Digital to Analog Converter that generates the sound for human hearing once again.

Most samplers today provide multiple methods for "triggering" the output of the finished sound. Nearly all samplers allow for MIDI control. If one links up a sequencer with a multitrack tape machine, and then uses this to trigger the sampled voices in perfect time to whatever music is in the underlying bed, the results can be captivating. Imagine the fun you could have with your station's calls! The stuttered effect is now so old, and passé' that most listeners are bored to tears with it (rightfully so!), but what about a call sign that starts backwards, and then inverts to normal right in time to the music? Better yet, how about using the call sign to modulate the underlying bed?!?! Now THAT would be different!

All this and much more is possible if you choose the right sampler for your station. Make sure you thoroughly understand what you are buying. Many samplers that serve musicians well, are inappropriate for production use. But *DO* get a sampler for YOUR station. If you don't have one, your GM has misunderstood its POWER to dramatically change the sound of your station. Beat him senseless until he gives in!

Add comment


  • No comments found