Synth Programming Secrets from Korg's Jack Hotop

December 18, 2013

In my years working for Korg, I spent innumerable hours listening to sounds, tweaking them, and helping decide which ones would go into a product. The person I sat with the most? Master programmer Jack Hotop, a Korg fixture for over 30 years. Jack has been part of the team of sound designers who worked on synths such as the Poly-800, the M1, the WaveStation, the Triton series, and both the OASYS and it’s step-brethren the Kronos. More recently he completed work on the Krome and Kross workstations. Many things I’ve been teaching come from studying the work of other programmers, and Jack and the whole Korg team have taught me so much. So he has been part of this column from the beginning, and I was long overdue to get him to share his approach to lead synth sound design.

CLICK HERE for examples of Jack's sound designs via his SoundCloud page.


 Smooth Singing Leads

Many programmers build a bit of punch into the attack of the amp envelope by creating a strong attack segment that holds for a little bit, and then comes down to a lower sustain level. I covered this in the December 2011 column and again in October 2013 on Jordan Rudess’ lead sound. Sometimes Jack uses this approach, but he also likes using envelope shapes that are more “even” and sometimes envelopes that decay slowly. When Jack designs leads that will be used for longer held notes and slower playing he prefers to keep a more level volume. 

“The attack spike works well for aggressive playing, but using a smother envelope shape helps make a line sing better,” says Jack. “Using a volume pedal lets you shape your lines like organists who use the pedal for dynamics, or assigning your modulation wheel to MIDI CC 11 (expression) lets you think like a horn player. You can also balance your playing registers better, especially with non- velocity sensitive leads.” Figures 1 through 3 show examples of the EG shapes he uses.

Fig. 1. This spiked amp envelope is the first of three shapes Jack Hotop likes for crafting leads. Fig. 2. This smoother envelope has a more singing quality. Fig. 3. Here’s a decaying amplitude envelope for leads.  

To get a singing lead tone from a distorted electric guitar sound using PCM, he doesn’t always use the original attack from the sample, as it may get in the way when connecting notes within the phrase. Different synths offer ways to avoid this; in the Korg synths he programs, they offer multiple start points for the sample(s). Sometimes it’s only one point moved in from the attack (usually just the looped portion), while other synths offer up to eight different start points that are pre-programmed by the developers, each moving further into the sound. 

Letting Go

Jack feels that the amplitude release stage is critical: too short of a release and the sound gets choppy, too long and it might sound sloppy. In general he will keep the filter release longer than the amp, so if he does choose to add a slight release tail to the sound, the filter won’t be heard shutting back down before the amp fades out.

Sync Suggestions

When creating synced oscillator sounds Jack sometimes puts a small amount of slow LFO-based pitch variation in the modulator’s sound to make it feel “alive.” He likes making both fixed frequency sync sounds (without an automatic sweep of the modulator’s pitch), and sounds that do have the sweep built in. To produce sweeps, he often likes to use an LFO, step sequencer, or loopable envelope, so the sound doesn’t just sweep and stop. This way when you want to hold a longer tone, the sound will keep moving. That slight LFO assigned to pitch helps out as well!

A Distorted View

Jack is very critical of distortion tones, and works hard to keep them from sounding harsh, buzzy, digital, or muddy. Here, he shares his usual workflow: 

“I find that simpler waveforms tend to work the best, although I’ll go beyond what you think of as the usual suspects. I recently made a distorted lead in the Korg Kross using an electric piano waveform (program C-021, “EP Annihilation SW2”), so anything’s eligible to me if it works well! When I first call up a guitar amp/speaker simulation, I’ll just listen to various combinations of amps and cabinet types until I get a general tonality that I like. That involves not only choosing the right models, but also listening to how varying the input gain versus the output changes the sound. Often a speaker simulation cuts too much definition out of the sound, so if it offers a mix control, or a direct signal path I will work with that. I should mention that I usually flatten out any built-in EQ settings within the effect when starting out. These EQ stages are always post distortion/gain, so if I feel something is lacking in the distortion itself I may go to an EQ that occurs before this effect, or add another EQ effect before the guitar sim/overdrive. I might boost bass to give the amp sim some more meaty frequencies to work with. But there are no hard/fast rules.

“Next I’ll work with the post-gain EQ, first sweeping the mid frequencies to find an area that gives me some more body and definition. I often choose a bump up somewhere between 3 and 3.4kHz, but that’s based on the Korg amp models—your settings will be different based on your synth of choice. On the top end I don’t like to just cut highs and harshness with a shelving EQ—that’s too brute-force. I look for the offending frequency by boosting a narrow band parametric to find the problem area, and then give it a cut, trying out varying bandwidths. If there’s a lot of digital harshness I might use a shelving EQ cut, but I’ll put the cutoff pretty high, from 11 to 19kHz depending on the sound. At times, chaining more than one amp sim, or an amp sim going into an overdrive in series can work well, as each has it’s own special character.”

Before or After?

Many programmers like to use a compressor on distorted lead sounds, and Jack has advice on that as well. “If I want the sound to hit the distortion dynamically I’ll put the compressor after the distortion. There are so many nuances of tone and that happen when you vary the input to the effect, and they can get lost due to the actions of the compressor. So I’ll put the compressor after the distortion/amp sim to give the final sound more body and sustain.” What settings does he use? “I have no set formula, for a natural sound I simply experiment, and I suggest you listen carefully to how your effects respond. Keep playing and varying parameters while you listen to what they do. Shut your eyes; stop looking at and thinking about numbers and let your ears and taste take over.” 

Parting Thoughts

Some final advice from Jack: “Always test your sounds through your live rig, and/or against a track. What sounds good by itself often changes through other playback systems or in a live room. If you’re working on a specific song, optimize the sounds for the parts and mix of that tune. When any manufacturer voices a synth, we have to make our sounds all-purpose, but you should hone in on your specific needs. Vary your LFO speeds for vibrato; consider the tempo of the tune and don’t make every sound the same. Last but not least, don’t forget to shut your eyes, make some emotional faces, and get lost in the joy of playing!”

Show Comments

These are my comments.

Reader Poll

Are any of your gigging synths analog?

See results without voting »