MRBarton

Wait - I'm confused. 🤔 Dr. Sigman's GX80 manual has the following thing to say here about the oversampling feature (bold emphasis mine): I'm going to add a little to the above stock phrase: "the higher the setting, the better audio fidelity will be" in your mind because you pushed the button. True oversampling will occur, but I will bet anyone that they cannot tell the difference in a blind test. So what's really going on when a plugin is oversampled? It means that for every output sample, the plugin, the whole plugin, ran more than once (say 3x) generating in between samples. Then, a steep lowpass filter is applied and the signal is decimated (take every 3rd sample) to create the output signal back at the project sampling rate. This example would more than triple the CPU requirement. If there are questionable routines in the software that produce aliasing, there might be some benefit in the oversampling process. However, in the GX-80, the aliasing products are vanishingly small. The GX-80 also contains no non-linear processing that may produce aliasing. Now in Miniverse, the filter and the VCA distort and the software has built-in oversampling to take care of this (that means you don't have to the push the button there either). However, only those components are selectively oversampled. Making the whole plugin, envelopes, LFOs, keyboard and knob processing, etc. run at a multiple of the sample rate is a huge waste (unless your computer doubles as a room heater). The steep lowpass and decimation process is imperfect as well, so you might be causing more problems than you're attempting to fix. So if it ain't broke... Some in the know may assume that control signals like envelopes, LFOs, etc. run at a lower rate than audio signals to save computing power. This was a common practice in both hardware and software digital synths in the past. Who knows? Some may still do it. If that were the case, that would justify (maybe) the use of oversampling to produce smoother control signals. However, every signal in Cherry stuff runs at the audio sampling rate, even a 2Hz LFO. Therefore, you can't get any smoother shooting down another justification to oversample. In conclusion, I speak for no one's work other than my own, but I expect that most modern reputable developers are completely on top of this issue and pressing the shiny oversample button on their stuff is also a waste. I have not tested every plugin. --mb

I just want to reiterate. The oversampling buttons are not "quality" controls. The instrument will sound no better if you use them. You should always run the GX-80 (and virtually every other modern plugin) at the native sampling rate of your project. --mb

You are correct, sir. I know there is a specific reason why multiple MIDI channels is not supported. Unfortunately, that reason escapes me at the moment. Perhaps Dan Goldstein will explain it here.

Just want to say regarding the GX-80 CPU load, it's mostly all about how many notes you are playing no matter how many instances you have up. If you have different instances responding to different MIDI channels (physical keyboards), the fact that you only have two hands will limit the load for you. The GX-1 had two CS-80s worth of hardware, one for each keyboard. To play both from one keyboard (which is equivalent to the GX-80 in DUAL mode), you had to turn on a coupler switch. Using two GX-80s both in SINGLE mode controlled by two keyboards would be equivalent to a GX-1 and would use no more CPU than a single GX-80 in DUAL mode. --mb.

I can tell you that the GX filter uses microscopically more CPU than the CS filter, but I'm sure it's nothing you would actually notice. The new, improved version is without a doubt much lighter on CPU. Running it on my doddering, decade-old machine is proof. In addition to the multi-threading improvements implemented by the brilliant Dan Goldstein, I also trimmed CPU usage in core routines down to the bone. I'm an old school programmer to begin with and instinctively write routines with the intention of relieving the computer's burden as much as possible. It's just that the GX-80 is quite the beast with a lot going on. Here's a tip: If you are not using something, turn it off. If a rank is not sounding, mute it. If you are not using a VCO resonator, set its volume to zero. If you are not using an effect, turn it off. There is sensing everywhere in the software to skip execution of code that makes no difference to the sound. Reducing polyphony is obvious, but by far the biggest suggestion I can offer is don't use the oversampling modes!! They are not "quality" settings and will do nothing to improve the sound. Unfortunately, some customers expect those buttons to be there so they're there. Don't waste your precious machine cycles on them. --mb

Great question. You can model at three different levels. 1 ) Component level: This is where you calculate the currents through all the parts in the circuit. This process is very CPU intensive and converges on the final value for a given sample after many iterations. It is my opinion that anyone claiming to be doing this in their simulations is lying, but I could be wrong. 2) Circuit level: This is where the code imitates what the parts are doing. For example, if there is a linear integrator, the code increments a value. It does not simulate an op-amp with a capacitor in the feedback loop. 3) Terminal analog: This is where you write code that replicates the output (terminus) of a given circuit or system. The code couldn't care less how the circuit accomplished the task. The GX-80 uses a combination of the second and third methods. For example, the way the CS-80 handles velocity and pressure is very complicated and had to be circuit modeled in order to capture the behavior. The sub-oscillator uses method 3. Which method to use comes down to experience and knowing what's important and what isn't. CPU resources are precious and it makes no sense to component model a gate signal.

I have also picked up a Hydrasynth Deluxe. They should really thank us for boosting their sales. Anyway, I know folks have gotten it working with keyboard and pressure, but are having problems getting the ribbon to send messages or respond correctly with the GX-80. If anyone has gotten all that squared away, it would be great if you could fill us all in regarding settings, etc. Thanks. --mb

Wow, me too. I wonder where that poster went. If you want one, just get the Voltage Modular version on sale now for a mere $25 (shameless plug -- I wrote those modules too). Sounds the same, works better, can have as much of any of it as you want = no brainer. https://store.cherryaudio.com/bundles/vm2500-collection#description Besides, it's really fun to mix and match it with the Moog 900 stuff. https://store.cherryaudio.com/bundles/vm900-collection. But the smartest move is to get it all right now (with the Oberheim stuff) in the Year Three Collection for only $65. The value for money is insane. I wish I was younger. Sorry for the commercial and going off topic, but we always get requests asking for the Moog and ARP stuff to be made into plugins. I don't think people realize that Voltage Modular is a plugin and making those synth systems into plugins would just put limitations on them. Attached is the poster I had (have). Same as yours? --mb

Oh everybody's got a story like that if you're old enough. Heck, about 30 years ago I traded an ARP2500 (that's 2500) for a DX-7. In my defense, it had a lot of blanks and the matrix switches were in poor shape, but still.. On a happier note, attached here is a hints & kinks (old ham radio term) document that went out to the sound designers before the GX-80 was released to give them a boost when starting to program all our wonderful presets. It doesn't cover everything, so for that you should look at the documentation here: https://docs.cherryaudio.com/cherry-audio/instruments/gx-80/getting-started, but it does cover what is unconventional and weird about the instrument and those things I thought required clarification. AND HAVE YOU GUYS SEEN THIS? https://www.musicradar.com/news/best-new-plugins-2022 It's barely been out a few weeks and we came in 2nd for best plugin of 2022!! I like to think we came in 1st among instrument plugins because the thing that came in 1st is an effects processor. Thanks to everyone who voted. --mb GX-80-Programming.docx

Hi Steve, You might have gotten a tear in your eye playing the Pirates preset, but I also got a tear in my eye reading your wonderful comments. It's feedback like yours that makes this monumental effort we all went through worthwhile. As is my bent, a tiny correction here. The modulation is not accomplished with the RING MODULATOR. It is the SUB OSCILLATOR working the voice filters with TOUCH RESPONSE -- SUB OSC AFTER increasing the SPEED and VCF depth. The Pirates intro that uses this sound is easy enough to figure out and the rest of the intro is just a repeated (and slightly arpeggiated) C F Bb C chord (top note is middle-C). I used to play Pirates all the way through pretty well, but I've forgotten practically all of it and my chops have kinda fallen away too. Too much typing and not enough playing. I can still manage a pretty mean Tarkus though. I've attached all the presets I used to make my demo recording which is among the others at the bottom of https://cherryaudio.com/products/gx-80 Once again, thanks for your great feedback and rock on! --mb MRB.zip

If you're talking about the opening modulation effect, this isn't precisely reproducible on a CS-80, as explained earlier in this thread. If the GX-80 has sufficient extra mojo to do this, I would love to hear it! I've been in touch with a team member who promised to soon have the patch. Seriously looking forward to that! Challenge accepted. Here ya go. I could have worked on it a little longer to get it perfect, but it's close enough, and it is indeed doable on a CS-80. Preset and demo attached. --mb PiratesIntro.mp3 Pirates Intro.gx80preset

Good ear, Mr. Anderton. It is indeed a Crybaby modeled after a vintage unit I have had for many years. We discussed which wah-wah pedal to model and we all instantly agreed. We wanted it to be a pedal that was around when the CS-80 was around to match vintage with vintage. The wah filter in the CS-80 is some kind of 2-pole constant-Q (I think) type which bears no resemblance to a wah-wah pedal's behavior at all and also sounds terrible. It's debatable which sounds worse, the chorus or the wah. Apparently, having built-in effects (no matter how bad) in a synth was a huge novelty at the time, so it's kind of like a dog walking on its hind legs. It doesn't do it well, but you're surprised it does it at all. Our mission statement was sort of like: Absolute authenticity except in the cases where the original sucks (and customers would be sad). Fortunately, this area was the only "except".

Dan left out one important point. We put in the superior effects because the hardware CS-80 chorus/tremelo effect sounds like a**. Probably the reason Dan never switches it on.

No doubt. The CS-80 is anything but dead tuned. Every voice card is different. Every voice card is different every hour of every day. In fact, every time to bring up an instance of the GX-80, you get a different one.

Huh, I guessed right. Who'd a thunk it?