Mounds of Sounds

Sep 1, 2003 12:00 PM, By Dave Hill Jr.

Creating music with samples offers you a chance to flaunt your aural tastes. The sounds you buy, make, fake, swipe, sample and re-create showcase your musical sensibilities and ultimately shape your sound. Give 10 producers the same 10 samples, and you'll end up with 10 different tracks. What each producer does with his or her samples depends on the intent, technique, style and dexterity on their sampler. Just what do people mean when they refer to a sampler as a musical instrument? And how can you harness its power? To get a better understanding, settle in and explore a few of sampling's overlooked fundamentals, tips for building your own patches and some other sampler-tweaking tricks.

SAMPLING 2003

A sample is simply any digital recording (usually a short one). A sampler, on the other hand, is more than a digital-audio playback rig: It's a musical instrument that can morph digital audio into wild new textures and apply deep filtering and modulation (motion), as well as repitch, resize and reverse the original sample. Just as guitarists can pluck a string in many different ways and route their signals through stacks of effects and processors, a sampler can take a given sound, enhance it or make it unrecognizable.

Traditional samplers, namely hardware samplers, are capable of digital multilayer recording, as well as resampling (recording a processed sample to itself). Most software samplers don't record outside sounds but opt for importing WAV or AIFF files recorded by another application. If you plan to make your own raw sample material, you need a decent audio editor such as Syntrillium Cool Edit Pro (www.syntrillium.com), i3 DSP-Quattro (www.dsp-quattro.com) or other hard-disk recording software. Anytime you work in these applications, you can listen for (and grab) samples. Even tiny snippets of sounds can be turned into interesting percussion and sound-effects noise with a sampler. With practice, you will begin to imagine what a recorded sound will become when transformed by your sampler's envelope, pitch, modulation and amplitude filtering.

For drum kits, most sample manufacturers place the sounds from lowest to highest starting with C1 as a kick, C#1 maybe another kick, D1 a snare and so on continuing up the keyboard. This technique of placing sounds against MIDI note numbers is called mapping and has no bearing on the original sample's pitch. You can place a sample of a C major chord on any key, just as you could put the lowest bass on your highest octave. Once you map a sound to a given key, you have the option to designate the key range (think transposition range) for that sound. Notes in the key range that are to the left of (lower than) your original sample will be played at gradually slower and lower pitches — one semitone (or half-step) per MIDI note number. Playing notes to the right of (above) your original sample will be sped up, or pitched incrementally higher. For drums, the pitch may be less relevant, but for tonal instruments, getting the original sample in tune is vital if you plan to play melodies, pads and bass lines in the context of a song.

I find that being consistent about mapping allows my patches to be interchangeable, as well as easy to navigate. By creating a workable system, recalling and working with sampler patches will be quicker in performance situations. Once I program a drum part in my sequencer, I audition several different sampler patches made up of different drum sounds. Because my patches are mapped similarly (kick, snare and hi-hat are aligned to the same MIDI note numbers), I can then scroll through them one at a time. By creating custom-made patches (like the previous example), you create a personalized library of sounds that inspires you and works for your music. Try combining and comparing your sample patches with ones that you have bought or heard in online demos. You can learn a lot by examining the work of others. Save your new patch musings often and try to give them descriptive, or at least memory-jogging, titles.

ENTER THE LAYER

One idea for building your own patches is to create a multipatch, in which several different instrument types coexist in the same patch. For instance, drums are placed in the lowest octave, bass in the next two octaves and a synthesizer pad or lead in the highest register. This technique allows you to do more within each “instance” of a software sampler and may eliminate the need for an additional hardware sampler.

A different but related tack is to build a multilayered patch, in which the same instrument is recorded at different dynamic levels, with different inflections for added realism. Usually, the sounds are mapped according to velocity (how hard you strike the key). This can really spice up a drum track when the snare and hi-hat vary according to velocity. One method is to use loud velocities for the backbeat (normally accenting on beats two and four) and then use a softer snare hit for the grace (ghosted) notes that typically land between the accented notes. Hi-hats are also a natural application for this sort of treatment when trying to simulate a real drummer's feel. Dynamic sample ranges can also work with horns, guitars, strings, vocals and many other sounds.

To add realism to the second velocity layer of a patch, try altering the layer's sound by decreasing its volume, sustain and velocity. Imagine how a softer performance might sound. For softer snare and hi-hat samples, lower the sample's pitch a couple of cents to simulate a lighter hit. When a drummer hits hard, the pitch of the drum goes up slightly, so lighter hits sound just a touch lower and often a bit duller. This same concept can work with other instruments, as well. Although the difference may seem to be quite small, the listener's subconscious hears the subtle variety.

SAMPLE SANDWICH

Although the layering concepts discussed in the previous section can take some time and practice, there are simpler ways to create some interesting sampler patches. Layering sounds upon sounds (so that they play together) can thicken, texturize and complicate sounds in amazing ways. Dance-music producers and programmers frequently layer several drum sounds on top of one another to take advantage of various sounds' strengths or to make altogether new sounds.

By mixing an electronic and acoustic snare, you can create a third snare that will contain the best of both worlds. For a more creative percussion hit, try mixing two found sounds — sounds recorded from your world — such as a car door slam with a hammer striking a nail or a shaker and a hand clap or a tympani and a gong scrape. Have fun with it. Long sounds also work surprisingly well when recombined, though occasional phase problems can occur. Vocal samples with synthesizers, engines and ocean waves or two contrasting synthesizers, such as an airy pad and a driving lead, are just a few of the combination possibilities available when mixing sounds. Software synthesizers are particularly well-made for this kind of resampling and patch-construction work.

CIRCLE OF FIFTHS

Resampling is merely sampling your samples. I like to think of it as submixing, rendering or even bouncing tracks into your sampler so that you can have a fresh go at applying envelopes, filters and effects to sounds that are already heavily processed. Resampling is a powerful tool for making new sounds, capturing an inspiring combination of samples and reinventing a stale patch. Also, because effects (plug-ins) swallow up scads of processor power, it can be a real CPU saver. Here is how it works: Imagine recording several takes of a monophonic synth with slightly different settings and effects. Next, at various intervals, you play three or four samples that are being run through a couple of effects plug-ins (say, a compressor and a delay feedback loop). Then, while this mound of sound is welling up, you record the output to a single WAV or AIFF file. If possible, record this directly to your sampler via S/PDIF or optical digital connection to avoid a generation loss on the signal. Your new sound will contain the long massive mutation of one or more previously made samples. This fresh digital recording can then be chopped up, reimported to your sampler and retweaked.

There is no limit to the amount of times you can resample. You may decide to resample these new resampled sounds by refiltering and retweaking before continuing on with the sound-designing process. Some sound designers are known to record hours of sample-playback experimentation onto a DAT tape for later chopping, sampling and finally resampling. This technique is perfect for creating wild ambient textures, tribal drum ensemble jams and loops and pads of many kinds. It's easy to see how recording and making your own samples go hand in hand. Creating fresh sample patches by resampling is indeed a creative endeavor.

FILTER FRENZY

A sampler's controls bear striking resemblance to those of a synthesizer. You'll find several different lowpass and highpass filters with frequency, resonance and ADSR amplitude/filtering controls, LFOs (low-frequency oscillators) and wave-driven modulation for changing a sample's pitch, panning, dynamics and more. Modulation can add lively motion and detail to your samples and can be synchronized to your music's tempo (in some samplers). Although this article doesn't cover all of the many facets of filtering, it's a logical assumption that LFOs and modulators will add new dimension to your sounds and that, when coupled with layering, both filtering and modulation are powerful patch-shaping tools. Lowpass and highpass filters allow you to quickly remove large frequency ranges of a sample and help balance your overall mix, similar to how a DJ uses EQ skills to feature a specific range of a track. Often, when combining sample layers, you will have redundancy in certain frequencies that can sound muddy or even swallow up the presence of a track. Use filters to pull out conflicting high or low frequencies.

ADSR (attack, decay, sustain and release) is the standard four settings (sliders), or graphic controls, for controlling a sample's loudness (amplitude) or the shape of a filter. The order of the letters logically dictates a signal's journey. Attack rate affects how quickly the sound is heard when fired. A small attack means that the sound will be heard right away whereas a larger attack rate will create a longer linear fade-in each time the sample is struck. Working in tandem with attack is a sample's start time and velocity — two separate controls present on professional-grade samplers. Start-time settings designate the point in the recording that the sample begins when launched; velocity determines the slope of the attack curve, acting as a volume accelerator when cranked. Fast attack and velocity rates make for the loud, punchy and in-your-face sounds, and slow settings take on more ambient, padlike characteristics.

The D (in ADSR) is for decay rate and represents the sound's rate of volume reduction. Think of small decay values as “quick,” as in quick to get out of the way. Large decay values take their time backing off to the sustain level. For example, a snare-drum hit will usually have a quick (small) decay value that dissipates quickly whereas a Hammond organ will play at relatively the same volume as its initial attack — an example of a larger rate of decay. The S is not a rate, but the sustain level: how loud the sound remains when you hold the note down on your keyboard (or sustain pedal). The bigger the sustain level, the louder the sound remains. A brass horn section might sting a note and then quickly back off yet remain holding the tone. That would be a sustain level beneath the volume of the note's attack. Relating to sustain, R is the rate of release, or the length and speed of the sample's fade-out after the sound is no longer triggered (or you let go of the key). Quick release times can be effective with sharp staccato-driven dance music; long release rates are great for gradually tapering tones, like an echo in a canyon.

You can also apply ADSR to filters for some charming aftereffects: The original sound fires, and then the filter attacks and subsides. As you become more comfortable with each control, you'll find that ADSR can provide tremendous shaping and power to the way that your sounds evolve.

By creating custom patches and sounds, you are taking an important step in becoming a musician and artist. Knowing what your filters can accomplish and how to work with ADSR envelopes will also help you to develop your own sound and inspire you to make music. Although there are thousands of great sample CDs on the market, many find that using what comes out of the box doesn't necessarily reflect their personal tastes. However, most sample makers intend these sounds to be a safe starting point and not the final word. Who would dare to tell you how your music should sound? Your artistic task, then, is to make these sounds your own by customizing, resampling and redefining the originals. As with any instrument, technique and experience all play a part in mastering a sampler. So now that you have a little know-how, take some of these ideas and go build a beautiful patch.

AVOIDING SAMPLE QUICKSAND

One day, while searching through gaggles of sample-packed gigabytes, I realized that there must be a better way to keep your samples organized than to become mired in the muck of sloppy file management, an unkempt hard drive and multiple (and often incompatible) sample formats. The amount of loops, hits, sounds, banks, patches and sample formats — as well as sound-making software, synthesizers, drum machines and keyboards — is mind-boggling. When creativity strikes, it is hard to remain calm (move back to the other side of your brain), create a proper folder, logically name each sample and then press on to make new right-brain-influenced sounds.

Even more difficult is to come back to a pile of poorly labeled digital scraps full of wonderful ideas and not know where to begin. I have often recorded long inspired performances, labeled them something akin to “cool drum stuff to edit later” and then wholly forgotten about it for, gulp, years. In talking with other producers, sound designers and musicians, I have found that I am not alone. The solution? A system. If you can get in the habit of naming your sounds with titles that work for you, you can alleviate the guesswork and skewed file-naming antics. For loops, I like to include the tempo right up front as in “120dryfunkdrums” or “135tranceaction.” For pitched instruments, you must always include the original note or key: “C#tympani” or “Gb3#5piano.”

Try to go back within a day of the sample-making session and organize your samples within folders whose names provide solid hints as to what kind of sound is contained. In doing this, you will notice that samples take up plenty of room and often require a dedicated sample-storage hard drive (or two). Inexpensive and portable storage options such as FireWire and USB 2 drives are available nearly anywhere computers are sold. At the time of this writing, you can easily find 120 GB of FireWire storage for less than $200. That's enough storage for 170 sample CDs! On that note, I recommend that you pick up two and designate one as a backup. Recently, I've been using an Apple iPod as a portable 20GB FireWire drive. Because it is durably built, it is perfect for porting samples to a gig. At the end of the night (or the next morning), I back up any changes to my “mother ship” 120GB FireWire drive, a consumer-grade but effective EZQuest Cobra.

CHOOSE THY WEAPON

The number of samplers on the market is enough to induce option paralysis. Software samplers such as Native Instruments Kontakt, Steinberg HALion 2.0 and Emagic EXS24 are solid choices with cross-format importing options. And they all ship with more than a gigabyte's worth of decent factory sounds. Also near the top are MOTU's new Mach 5 sampler, Tascam's GigaStudio (GigaSampler) and IK's rebuffed Sampletank. But that's not all. Laughing off extinction, hardware samplers have gone through a major retooling with Akai's Z-series (now capable of 24-bit, 96kHz sample rates) and E-mu's multifeatured E4, among others. Also, Akai's MPC-series sampling drum machine and Roland's SP-808 phrase sampler are faves among the hip-hop crowd. Spend a little time “sampling” soft-sampler demos, as well as their factory-included sounds and specs, to see what works best with your style of working. Keep in mind that there will always be a learning curve with such a multifaceted instrument. What is most important is not which sampler you own, but what you can do with it. How much (and often) do you enjoy working with it? And will it jive with your studio setup? Do you own several sample CDs in another sampler's format? Can they be easily converted? These questions will undoubtedly steer you toward the perfect purchase.

Acceptable Use Policy blog comments powered by Disqus

Want to use this article? Click here for options!