Techniques for sampling drum
Technology, Digital Sampling
Published by Keyboard/May 1987
Terry Fryer is a synthesistspecializing in sampling and a principal in the Colnot-Fryer music production
company. He is also the owner of Ear Works, an audiophile sampling service.
- TIME TO GET BACK TO THRASHING A drum kit.
- If you have been following the last several months' developments, you should have a well-documented sampling session of a snare drum. But what's next? Drum machines usually contain samples that are made with a single drum or a single cymbal. Since we're trying to be smarter and more flexible than the average drum machine, we're going to sample more than one drum at the same time.
In the realm of "that's the way it's done" listen to what happens when a drummer sitting at a kit is asked to play a crash cymbal. Crash goes the cymbal, boom goes the bass drum. I think there is a gene found in drummers that makes it physically impossible to hit a crash cymbal without playing the bass drum. So why not record it that way? Here's how.
If you're going for a realistic effect, use the first track of your stereo recorder to record the cymbal and the second for the bass drum. This will give you the needed flexibility to later control the balance between the two instruments.
If your motive is to create a not so realistic processed effect, mix the cymbal and kick to one track and record the ambient sound of the room on the second track. Usually, I do not do any processing at this point, because there is an immutable law that says, "Whatever effect you record onto the tape will be almost the right one, but not quite." For those of you who are concerned with tape noise, try some effects such as compression and limiting at this time. This will restrict how much you can change the effect at a later date. On the other hand, if you wait to play with the sound, you will be processing whatever artifacts (such as tape hiss) are added during the recording process as well as the sound you recorded. For flanging and phasing, a little noise heightens the perception of the effect. But with limiting and compression the sample will be made noisier.
- Recording engineers who have to deal with complete drum kits (that means lots of drums and a human to hit them) snarl and gnash their teeth trying to get the sound of the hi-hat out of the snare microphone. One of the benefits that comes with buying a digital sampling keyboard is that we don't care what they think. So give your snare drum a little extra sizzle: Record a hi-hat with the snare drum. In my opinion, the trashier the hi-hat, the better it sounds. Experiment with varying lengths of open/closed hi-hat to simulate different room sizes.
Another thing engineers have to deal with is a ringing tom-tom or bass drum messing up their snare sound. Since we still don't care, use a bass drum or tom-tom to beef up a wimpy snare. A bit of experimenting with microphone placement, mallets, different types of sticks, bass drum beaters, and even equalization can net you a heavy duty snare drum.
- More ideas on the same theme:
- Hit two tom-toms at the same time. Try a combination of tom-toms and kick drum. Or snare, kick, toms, and hi-hats. Or if you're daring, hit the drummer. You get the idea.
Here's a technique that should yield some good Phil Collins/Hugh Padgham gated-type tom-toms: Find a small room with hard surfaces that has a fairly short decay (under 1.5 seconds). You can try a bathroom, a shower stall, or a tiled kitchen. Another solution is to varnish several sheets of 4' x 8' plywood, then put them on the floor and lean them against the walls to provide the reflective surfaces. You can pretend you're 12 again and build a little fort out of them, complete with a ceiling and a floor. That should pay the iguana back for shuffling last month's log sheets.
Tune the drum so it doesn't ring sympathetically with a node in the room, Move the tom-tom and one of the microphones around until you get a nice bright sound lots of stick hitting the head and not a lot of ring from the tom-tom. Take the other microphone and point it away from the tom-tom into a corner of the room.
If you have a boom stand, try moving the mike toward the ceiling. Experiment until you find the best location for the second mike. It should give you a nice, splashy sound.
lnsert a compressor/limiter between the second, ambient microphone and the recorder. Set it to the highest amount of compression,adjust the level of the recorder, and thrash away. The extreme compression will minimize the difference in level between the sound of the drum and the reflected sound.
There are two schools of thought if you are going to record an entire drum kit in this manner. One is to record all the pieces of the kit with similar microphone placement for the near and far mikes. This will ensure that all the resonant frequencies of the room will be the same. The drums should sound as if they were recorded live from one drum kIt if the playback rate is the same as the sampling rate. If you change the playback rate of the samples, transposing them up or down, the resonant frequencies of the room will also change, and the sound of the drumkit will not be as cohesive.
The second method is to change the sound of the room for each of the drums. Instead of tuning the drums, you can literally tune the room. This works best with the varnished plywood scenario. For each drum, move the pieces of plywood around (making the apparent room size larger and smaller) until you hear a resonant and satisfying "sproing." I realize I said a couple of paragraphs ago to tune the drum so it did not ring in the room, but try this technique anyway.
If you retune the drum up or down, you will need to change the plywood until the "room" resonates with the new pitch of the drum. Another tactic is to use absorptive material to change the characteristics of the room. A little Sonex [a sound-absorbing foam material made by Alpha Audio, 2049W. Broad St., Richmond, VA 23220] or some old blankets will go a long way toward changing the sound of a room.
Now that you've decimated your house, stained the carpettrying to varnish some plywood, scared the neighbors into submission, and used all your paper to take proper notesofthe proceedings, takeabreak until next month, when we'll have more on the endless enigma of sampling.
- Gremins Department: Yes, folks, those pesky little buggers have been hard at work in the word processing department at Ear Works. Several months ago, I listed a B&K microphone labeled 4002. Jim Rondinelli of Breul & Kjar pointed out that there is no such thing as a B&K 4002 microphone. It should have been a 4007. Sorry for the confusion.
