01. SOLO TRPT (Gary Leuenberger, December 1985)
The only way I can get anything close to a brass sound on the Yamaha DX7 is with a basic FM pair. In this particular case, I chose algorithm 18 because my FM pair happens to be

right in the middle. I've got operators 1 and 3 together, with a one-to-one frequency ratio and feedback set at 7. You cannot get a horn without that feedback. It creates the kind of sawtooth harmonic which we recognize as a horn sound. If you turn off operators 2, 4, 5, and 6, and just listen to 1 and 3, you'll hear a very nice basic brass sound. To make the tone quality brighter or mellower, you would look to the output level of operator 3. In order to stay in the basic harmonic area of the trumpet, the output level should range from 75, for more of a flugelhorn timbre, to about 85 for a very bright trumpet. When someone blows on a trumpet, there's a little bit of flutter, or spit, right at the beginning of the sound. Operator 2 fits in nicely for this because it's coming in on the side of operator1. It is affecting the overall harmonic content, but I only added a little bit. I intuitively chose 2.12 for the frequency. I started at .5 and moved up until I got just the right amount of buzz, or flutter; 2.12 happened to be the right number. You can also change the output level of operator 2 up or

down a bit from 65 to regulate the amount of buzz. It's subtle, but you miss it when you take it away. If you really want to hear the effect of operator 2, and this comes into editing all voices, start out with no velocity sensitivity, because sometimes you won't hit the key hard

enough, and you'll miss the effect. So remove the velocity and play with the output level to

hear the effect of operator 2. You might even want to remove the level scaling on the right hand side, which is at 10, to give you more of a buzz in the upper end of the keyboard. Operators 4, 5, and 6 also contribute to the slight distortion of the trumpet timbre. When you play a trumpet with more force, there seems to be more 'stuff' coming into the sound.

When I made this voice I left the breath control off just in case someone doesn't have one, but I would really prefer to use a breath controller. The breath controller becomes the

envelope generator, and gives you that total control, so you don't have to worry about articulating your fingers too much on the keyboard. If I had lip control, the first thing I would do would be to remove all velocity sensitivity, then decide what inputs I wanted to have control volume and timbre. I want total control of the volume so I would put the AMS (amplitude modulation sensitivity) of operator 1 at 3. I would also like to control the level of the low spit, which is operator 2, so I would put the AMS of operator 2 at 3. Harmonically, you could also put operators 3, 4, 5, and 6, on full AMS, but when you play very softly there's really too much harmonic content, so I bring those down to about 2. This gives you good middle control, and the harmonics come in very nicely if you give a firm little toot on the

breath controller. But also, if you play in a very soft way, you get a real lyrical sound. I don't think you should play a chord with this trumpet patch, so I put it into the mono mode and added fingered portamento, set on 25. When you play a legato line, you get that nice little bend. Performance-wise, you have to be concerned about the vibrato. I've put it into

the voice in the PMD (pitch modulation depth) section, so it always fades in with a delay of 50, which is a nice, natural effect. But when you really hang on the notes, then you want to go to the mod wheel and bring in more vibrato. The after-touch is optional. Sometimes it gets in the way of the voice, because if you have something that's velocity-sensitive, every time you really hit the note you tend to hold on to it and there's too much vibrato. Depending on your playing style, sometimes this works very nicely, so you have a choice here. To set the breath control functions, put the range of the breath controller at 99, which

means it takes over the entire envelope structure. Obviously, you will not be able to hear the

sound until you blow into the breath controller. The EG (envelope generator) bias is the key, so that goes on, but pitch and amplitude are off. If you'd like to control the level of the actual hard spit, which is in the FM stack 4, 5, and 6, change the output level of operator 4 from 65 to 75. This trumpet voice was designed very carefully around the feedback loop in the middle of the FM stack. Switching algorithms is really not going to do you any good in this case, because you're not going to find any other algorithm that uses operators 1 and 3 with the feedback loop in the middle. I tend to choose FM stacks like this when I want pretty complex harmonic stuff, with one carrier and five modulators coming in. If you play around with the output levels and envelope times, you can tweak a great trumpet timbre, or you can stretch the sound into something that no collection of metal tubing could ever produce.

03.      GLASSVOICE  (Jim Aikin, March 1986)
GLASSVOICE, this month’s DX7 patch, was created by our very own Jim Aikin. Jim explains, "This sound has a fairly slow attack, so you will want to play slightly ahead of the beat. When I load a melody that uses this sound into a sequencer, I record the part on the beat with an electric piano sound, auto-correct, and then shift the track four or five MIDI clocks ahead of the beat to drive the Glassvoice patch. This makes it a lot easier to get the sound uniformly on the beat, rather than having to play all the notes early."