cherryDan

The Korg PS-3300 is full polyphony. 48 synthesizers per key times 3, for a total of 144 individual envelopes, and filter circuits! The oscillators are indeed divide-down from 12 master oscillators per panel, which is why the Temperament Controls allow you to detune the individual notes. What this means is that every C, C#, D, etc. key will be in perfect tune with all others. Without the divide-down design, the task of tuning 144 individual analog oscillators would likely be a nightmare. And of course the divide-down oscillator design allowed them to save on circuitry for a synthesizer that is already packed full of an insane amount of wires and circuitry. In comparison, the Polymoog also uses divide-down oscillators, but unlike the PS-3300, the envelopes are filters are paraphonic. This compromise led to the Polymoog being a much more 'portable' instrument, but the paraphonic nature pales in comparison to the PS-3300's immense capabilities. Dan @ Cherry Audio

It's not obvious, but there are certain additional controls that change based on what layer you're editing, that are outside of the white and orange framed areas. GX-80's left-hand controls, including Sustain on/off, Porta/Gliss on/off, Sustain mode, Sustain time, etc., are actually per-layer controls. So if you don't want sustain on the left hand bass part, just switch to the layer and turn the SUST switch to off, or set the SUSTAIN short/long time to short. If you want a long sustain on the right-hand sound, switch to that layer and crank up the Sustain Short/Long slider. Also - again, it's not obvious - the Voice Assign and Number of Voices controls are actually per-layer, and not global. Your left-hand bass patch can be a giant unison patch, while your right-hand patch can be a giant polyphonic sound. Or - and this is really cool - in Dual mode, your upper layer can be a gorgeous polyphonic string patch dripping in reverb, while your lower patch is a gliding (or glissing) monophonic lead sound, always gliding up or down to the last note you played. The results can be extremely gorgeous! Dan

Craig, I'm so glad you brought this up, because it's such a great illustration of a topic that comes up a lot with Cherry Audio. So, before I tackle the topic of whether or not Cherry Audio substituted a rotary speaker emulator for the tremolo, please indulge me in a bit of behind-the-scenes synthesizer emulation talk. Often during interviews or during one-on-one discussions about Cherry Audio and our vintage synthesizer emulations, people will ask me what the most challenging part of emulating a vintage synthesizer is. It's a great question. Obviously, matching the quality of the oscillators and of the filter are paramount, and a ton of work goes into this aspect of the recreation. But often the most challenging part is not what you might expect. What we've learned is that the character of an instrument is far more than just the raw tone. It's the shape and speed of the envelope generators, the mono-polar or bi-polar nature of the LFO waveforms, the slope, speed, and linear or logarithmic nature of the portamento, the way the sound gets overdriven when you play too many notes or crank the mixer volume up too high. All of these things contribute to the sound and feel of an instrument far more than you might expect. These details are particularly important in the CS-80, which has all sorts of unusual interactions. For example, because the CS-80 ranks have both a highpass and lowpass filter, and because the keyboard CV is always routed to these filters, there is always a bandpass effect on every note, and this is a big part of the CS-80's characteristic sound. So, in this way, it's not just important to accurately emulate the sound of the CS-80's filters, but it's also incredibly important to accurately emulate the external behaviors that control the filters. So here's what I tell people is perhaps the most challenging part of modeling a classic synthesizer. No matter how well you know a synthesizer, no matter how many years you've played a synthesizer, you probably don't know everything there is to know about that synthesizer. And you have to understand everything about an instrument to properly recreate it. My favorite example is the Jupiter-4's arpeggiator. The Internet is full of people saying that the Jupiter-4's arpeggiator is unique, and behaves differently than any other arpeggiator. Everyone says it, everybody knows it, but look around and you'll find no one explaining what's actually unique about it. So when it came time to recreate the arpeggiator for our Mercury-4 instrument, I realized that although I've owned a Jupiter-4 for many years and have played with the arpeggiator extensively, I actually had no idea how it actually behaved. It always seems to produce musical results, but it's strange. Most arpeggiators have a "Number of Octaves" control, but the Jupiter-4 doesn't. However, it always plays multiple octaves, except as you play higher up the keyboard, it players fewer octaves. Very strange. Most arpeggiators, if you play a C major triad (C, E, G) will play these notes in order - C, E, G up, G, E, C down. But the Jupiter-4's arpeggiator doesn't always do that. In short, recreating this arpeggiator meant understanding exactly what's happening under the hood of the Jupiter-4, and that took a lot of study. Mercury-4's arpeggiator works exactly like the Jupiter-4's arpeggiator. If it didn't, it wouldn't really feel like a Jupiter-4, no matter how accurate the oscillators and filters were. Now, all of that said, let's talk about the Tremolo switch on the CS-80. As you can see, the CS-80 has a Chorus and Tremolo on/off switch, plus an effects Off/On switch. So it would be entirely logical to assume that you can have either chorus, tremolo, or both turned on. Personally, I've owned my CS-80 for 7 or 8 years, and I always assumed that the Chorus switch turned on a chorus effect, and the Tremolo switch turned on a Tremolo effect. I'm not certain I ever actually tried the Tremolo switch. I am certain I never understood what the Tremolo switch does. The truth is, until we had the task of perfectly recreating this enigmatic instrument, I'd never really understood the CS-80, even after playing it for many years. The Tremolo switch does not add tremolo. Weird, right? How can that be? Why have a Tremolo switch that doesn't introduce tremolo? First, the CS-80 already has a way to do tremolo. The sub-oscillator (LFO) section has an LFO->VCA depth amount, which modulates the overall volume and introduces tremolo. Second, remember that the CS-80 was built by an organ company. Despite the fact that it's clearly a synthesizer and not an organ, it was still built with many organ-like features, and used lots of organ-like terminology, such as "ranks" and "feet", and it has those weird reversed performance paddles, the ones you pull towards you in order to increase the level of a parameter. They still thought of this thing as a fancy programmable organ. With that context, it starts to become clear why the CS-80 has a chorus effect, and why there are Chorus and Tremolo switches. The "Chorus" effect is actually meant to be a Leslie emulation. After all, all great organs sound even better through Leslie animation. On the CS-80, turning the Chorus switch on and the On/Off switch turns on the Leslie-style chorus animation. Switching the Tremolo switch on speeds up the chorus effect, just like the fast "tremolo" setting on a Leslie speed switch. Turning Tremolo off slows down the chorus, just like the "chorale" speed on a Leslie. In fact, the CS-80's chorus effect actually ramps up when switching from chorale to tremolo, just like a Leslie. Finally, turning off Chorus and switching Tremolo off and on switches between no chorus and fast chorus, emulating a single-speed Leslie speaker, or a two-speed Leslie speaker with the slow motors disconnected. For GX-80, we wanted to emulate this chorus behavior. The Speed control in the Chorus effect emulates the Tremolo on/off behavior of the CS-80. But we also added a proper, modern rotary speaker emulation, so if that's what you want, you don't have to make do with the CS-80's simplistic Leslie emulation. I was really surprised when I discovered that the Tremolo switch is a rotary-style chorus speed control. I always assumed, as Craig did above, that the CS-80 has a built-in tremolo effect. This was one of a hundred little mysteries about the CS-80 that we uncovered while creating the GX-80. Mark Barton and I could write pages and pages about other surprising things we discovered -- including a number of interesting details that our competition gets wrong -- but I've already written enough for now! - Dan @ Cherry Audio

That's a great writeup, Mike. I just want to add that Yamaha doesn't use the word "synthesizer" anywhere in their documentation of the GX-1. In fat, they refer to the analog synthesis technology as "natural sound," a new kind of feature for an organ. Interestingly, the GX-1 isn't particularly great at making traditional organ sounds. You have at most four sine waves per key (if you link the top and bottom keyboards together), so making authentic big 9-drawbar Hammond organ sounds is out of the question. Your Synth Gems 1 book was a great source of information for us at Cherry Audio, as we first explored what made the GX-1 unique. I have to admit that I had all sorts of misconceptions about what a GX-1 was at the start of this project, and I get the feeling that many other people do, too. People expect it to be the most incredible synthesizer they're ever seen. In fact, it's a rather limited preset-based organ, with important performance controls at your fingertips, but absolutely no synthesizer parameter controls. Plus, there are only 10 preset sounds per keyboard. But it unquestionably led to the creation of the CS-80, and that's an incredibly impressive instrument. Dan

Dr. Metlay, I'm not sure if this is exactly common knowledge or not, but the CS-80 actually has two different ways to bend the pitch. One is via the ribbon, and the other is via this pitch wheel with integrated master pitch control. The pitch wheel bends around 7 semitones up and around 4 semitones down, basically giving you the ability to bend up or down to a chord's 5th. Rather than trying to support two different types of pitch bend controls, and in the interests of making GX-80 respond to MIDI pitch bend in a useful way, we removed the pitch bend wheel and replaced it with the ability to switch the ribbon between linear and semitone mode, and we expanded the semitone range from the original +7 semitones to as much as +12 semitones. 24 or 48 might be useful for some, but it's well outside of what a CS-80 can do. As for the current position "jumping around a bit," that's only true with MIDI pitch bend, and it's necessary in order to support the full MIDI pitch bend range up and down. After all, a pitch wheel isn't a ribbon, and we want the entire bend range to be available to anyone using MIDI to bend the pitch. If you click on the ribbon with your mouse, or use a touch screen to control the ribbon, the ribbon will work exactly as the CS-80 hardware does: If your physical ribbon transmits MIDI pitch bend, there's no way via MIDI to know where on the ribbon your first touched it. But it's not true to say the on-touchscreen ribbon doesn't put the initial position marker where you first touch. Try it! It does, and it's fun! - Dan @ Cherry Audio

Hi! We worked with two different CS-80s, and in many cases the calibration of each instrument and the calibration of individual voices differed noticeably. For example, the maximum attack time on the VCA and VCF envelopes ranged anywhere from around 750 ms to 980 ms, depending on which CS-80 we were analyzing, and which voice we were playing. The Yamaha VCA chips specify a maximum attack time of 1000 ms, and since at least some voices came very close to that, we went with a maximum attack time of 1000 ms in the software. Another example is the Sub Oscillator (LFO) speed control. One unit's LFO could go slower than the other unit. So we went with the widest possible range, the slowest available LFO and the fastest available LFO. We're talking about a relatively small difference, but we wanted to make it possible to recreate any sounds that we could create on either CS-80. In pretty much every case the controls are scaled to match the hardware as close as possible. But since each CS-80 is a bit different, and individual voices on any given CS-80 can differ considerably, and since small movements of these controls can cause major audible differences, you'd still have to use your ears to match the hardware to the software even with the controls in identical positions. The CS-80 is very much a synthesizer with major sweet spots. It's really interesting, we noticed quite a bit that using features like the ring modulator can cause all sorts of inharmonic sounds, and then suddenly create something really beautiful and musical, just by careful adjustment of the controls. Once you hit those sweet spots, the results can be incredibly beautiful. - Dan @ Cherry Audio

On the CS-80, wherever you first touch the ribbon controller is the "zero bend" point. Then, bending up or down will cause the pitch to rise or fall in a linear fashion, which is very unusual. If you touch the ribbon at the very left side and slide all the way right, the pitch will rise by 1 octave. If you touch the ribbon at the very right and slide left, the pitch will fall all the way to 0 Hz. There is no way on a CS-80 to do hammer-ons, though you can touch the ribbon with one finger and then use a second finger to hammer-on. Very difficult to do while also playing notes! Dan

Hi everyone, thanks for being a part of our GX-80 launch! I don't think anyone at Cherry Audio got a good night's sleep last night, we're all buzzing with excitement for this release. GX-80 is a product that we've been developing for more than a year, with a single-minded goal of combining everything that's great about the GX-1 and CS-80 together into a single, beautiful instrument. My team went crazy ensuring that every aspect of the software matches Yamaha's insane original vision for the GX-1 and CS-80. To say that these are complex machines would be a massive understatement. These are instruments where cost and technological limitations were not an issue. These were instruments that with velocity, aftertouch, and built-in effects years before other manufacturers even began to dream of these features. The expressiveness and musicality of these instruments is breathtaking, and we have all worked extremely hard to build a software synthesizer that's worthy of this legacy. We'll be here to answer any question you might have about GX-80, it's concept, the development process, or whatever else you'd like to know! Thanks for joining us! Dan Goldstein Chief Technology Officer Cherry Audio