cherry audio gx-80 Team Test: Cherry Audio GX-80

[Al Coda]

3 hours ago, GabrielPas64 said:

"This idea of pseudo-poly-AT was implemented for the first time (AFAIK) in Cherry Audio's GX-80 VST plugin."
I stand corrected! I have updated the docs in the new version which I am about to upload.

 

Gabriel

 

GunnarE (Memorymoon) released ME80 december 22, 2009.

The v2.0 version was released  july 29, 2014.

ME80 offers usage of MIDI channel pressure or MIDI pressure w/ last note priority alternately and storable per patch.

I´m the beta tester and this info comes from emails being stored on my office machine´s drive still.

I was also somewhat involved in Memorymoon "Memorymoon" and "Messiah" 32Bit version development in the past, which I did by interest because I wanted these plugins for myself.

 

It´s all no biggie IMO and I appreciate, GX-80 works the same.

But,- Gunnar and his partner coders were faster.

 

 

A.C.

 

 

[Al Coda]

3 hours ago, GabrielPas64 said:

I am glad you were able to install and test the script. A new version is included here (with a pdf manual). I have tested it with ME-80, set according to your suggestion, and it works nicely. There is still the problem of the installation procedure in Reaper, which I would like to clarify. For now, the installation described in the docs is still the one which does not work for you. In the zip file you will find also a new script (midipressuretopolylagdel) which is able to delay the release of the poly-AT for the previous note, thus producing smoother transitions (it is explained in the pdf, with pictures).

 

All the best,

Gabriel

 

 

 

Hello Gabriel !

 

Thank you,- just downloaded !

Will try tomorrow.

 

 

A.C.

 

[chaosmeister]

On 1/10/2023 at 9:02 PM, Robert Saint John said:

Hi folks, just a quick update: we just released a new version update for GX-80, version 1.0.13 (build 147), that adds optimized multithreaded processing for presets that utilize the Dual Layer or Split modes. This should give you a nice performance boost for these presets or your own sound designs!

Maybe it's just me... but the new version seems to be quite a bit heavier on the cpu, esp. with the GX filters enabled.

[MRBarton]

On 1/11/2023 at 1:06 PM, chaosmeister said:

Maybe it's just me... but the new version seems to be quite a bit heavier on the cpu, esp. with the GX filters enabled.

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

[Robert Saint John]

On 1/11/2023 at 4:06 PM, chaosmeister said:

Maybe it's just me... but the new version seems to be quite a bit heavier on the cpu, esp. with the GX filters enabled.

Can you tell me what platform you're using, which OS and the specs of its CPU?

 

ADDED: we just released a build number 149 that may address your issue, let us know! Open up GX-80 and let it update. No download or restart required.

[mate stubb]

Just got around to downloading the demo. First impression is very WOW! The sound is beautiful. Well done team!

[David Emm]

Just a brief comment... after combing through the presets a few times, I'm well-satisfied that the broad range included gives me ample leaping-off points for programming. I'm still trying to get used to its organ-style aspects and filter oddities, but about a third of the presets are the instrument's base VOICE. For example, Mike Martin's harmonica patch is startlingly good and probably the ultimate expression of that renowned sound. I tweaked a parameter or two and it was suddenly perfect, FOR ME. I was easily able to position it in my setup as a result.

 

Its enough of a synth-y synth that it'll do a fair amount of Synth Stuff™, but my task is going to be properly applying it where its natural strengths are the most prominent. I can make a Memorymoog cover a lot of ground, but the GX-80 holds a unique place in the more lush middle range, where nothing else touches it. I can't help but grin and think of it as a 400-pound cello. Keith had big brass ones and no doubt a small army to tour with a GX-1.            

[Adrian Lee]

On 12/9/2022 at 2:35 AM, jerrythek said:

When discussing it with Cherry Audio, when I felt it was really needed, they helped convince me that it's only a likely very small group of people who would find this a big problem.

MIDI channels. ⬆️
I imagine that's true... at the point of entry. Within minutes the realization sets in. The GX-80 is not a CS-80 any more than a CS-60 is not a CS-80.  The lovely Cherry Audio promo video sliding the visual of a CS-80 and a GX-1 into one integrated, single manual, recombination (probably more apt than 'hybrid') somewhat obscures the beauty of there being two, entirely independent, synthesizers on board. Independent, save for the pitch bend and MIDI channel. Both of these are candidates for separate addresses.

Firstly, as has been mentioned, to get a GX dual manual experience, you gotta have assignable MIDI channels. To bring the more 'CS' like presentation of the GX-80 into the GX-1 multi tiered world, recreating its towering presence both physically and sonically, is there a more direct method than providing MIDI channel assignment?

Secondly, as was obvious straight out of the starting gate, the GX-80 demands to be played with controllers fully employed. This instrument ('synthesizer' doesn't really do justice to how we interact with such a creation) is a fraction of itself without those paddles, rockers, ribbons and AT being the experience which brings forth that sound.  To gain access to all parameters, and to promote familiarity with what is where, a dead simple approach is to use the same controllers for both the upper and lower ranks, differentiated only by MIDI channel. This doesn't require the user to employ identical controllers, it just allows for such an approach. The same setup can easily be loaded into two controllers and brought into action with the mere flip of a channel assignment.

Somewhat related is the plea for separate control of the ribbon for upper and lower so that, regardless of channel assignment, the lower ranks are not forced to bend in concert with the upper.  And did anyone mention provision of separate audio outs for upper and lower? How else to get the full benefit of the 4 TX-II's every home installation must have? Let me know when the plugin version of those arrives.🤪

 

[Al Coda]

Every midified synth I know, offering single-, split- and dual (layer) modes, offers user selectable MIDI channel selection for the individual "lower"- and "upper" parts.

Even dual timbral hardware synths in MIDI chlldshoes offered at least 2 fixed different RX-MIDI channels once they were midified.

So,- softsynths, offering "dual-engines", should do the same.

 

 

A.C.

[Dave Bryce]

On 1/23/2023 at 3:52 PM, Al Coda said:

Every midified synth I know, offering single-, split- and dual (layer) modes, offers user selectable MIDI channel selection for the individual "lower"- and "upper" parts.

Even dual timbral hardware synths in MIDI chlldshoes offered at least 2 fixed different RX-MIDI channels once they were midified.

So,- softsynths, offering "dual-engines", should do the same.

 

Yeah, but…

 

Hardware synths can’t do multiple instances (unless you own more than one of them, of course).  Why isn’t it a good solution to just run multiple instances of GX80 when you want to use/manipulate different sounds at the same time?  
 

To me, that’s one of the major advantages of software instruments and processors.  I love being able to use the same software compressor as many times as I want on a project with different instances on different sources. 😎

 

If it’s a question of processing power, rendering tracks to audio is a fine solution.

 

dB

[Al Coda]

15 hours ago, Dave Bryce said:

Why isn’t it a good solution to just run multiple instances of GX80 when you want to use/manipulate different sounds at the same time?  

 

Well Dave,- I just only would prefer treating software instruments like hardware instruments WHEN I play.

A (home- or project-) studio situation and when using a DAW is different.

 

I´m not using the latest, greatest and fastest computers anyway,- and from my experience, several instances of a plugin always cost (a bit) more CPU than a single instance.

It might not be crucial in every scenario,- but when using many plugins in a more complex setup, it sums up.

 

It also depends on the DAW application in use,- and I use several,- Studio One Pro 4.6, Reason 11.x.x as also Reaper by occasion.

I like the option using several MIDI channels from a single MIDItrack and w/ just only 1 instance of a plugin, which I consider to be a ideal solution for splits and layers across different MIDI channels.

It also keeps track count visually smaller on a 15" laptop screen.

As an example, this IS possible w/ Arturia Matrix12V which mimiks Oberheim´s "Multipatch" mode perfect.

 

So,- CA´s GX80, ELKA-X as also Quadra would be good candidates too.

And,- I like their plugins,- owning Miniverse, Memorymode, Polymode, Lowdown, CA2600, Quadra, 8Voice, DCO-106, MG1 Plus, Voltage Modular w/ VM900 modules and Stardust-201.

So, it´s all about more flexibility.

 

 

A.C.

[MRBarton]

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.

[jerrythek]

Hi mb:

 

So another way of saying that (or understanding it), is that there's very little "overhead" load by having the plug-in active, correct? The main load on the system comes from the actual playing... Am I correct?

 

I'm still in the camp of things having a definable MIDI channel, because.... that's the way most of us are used to thinking about MIDI communication. And as Another Scott reminded me much earlier in this thread, it becomes more necessary (easier?) in live performance setups.

 

Thanks as always for weighing in with helpful info.

 

Jerry

 

3 hours ago, MRBarton said:

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.

 

[Al Coda]

4 hours ago, jerrythek said:

I'm still in the camp of things having a definable MIDI channel, because.... that's the way most of us are used to thinking about MIDI communication. And as Another Scott reminded me much earlier in this thread, it becomes more necessary (easier?) in live performance setups.

 

 

A.C.

[MRBarton]

7 hours ago, jerrythek said:

So another way of saying that (or understanding it), is that there's very little "overhead" load by having the plug-in active, correct? The main load on the system comes from the actual playing... Am I correct?

 

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. 

[Adrian Lee]

On 1/31/2023 at 10:05 AM, MRBarton said:

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. 

Hi  Mark, that might be a great reason for offering the incremental voice numbers that we see on every other Cherry Audio synth. The jump in demand using the current 2,4,8,16 offering is huge, especially when considering double racked, dual mode. Between 8 and 16 it's the difference between 32 voices or a whopping 64. Ouch.
Notwithstanding that there are good creative reasons for limiting voice count; to prevent unruly releases, for example. 

 

Quote

the fact that you only have two hands will limit the load for you.

That's true up the point that any kind of Yamaha CS-80 type sustain or release is in use, I would suggest. 

 

Quote

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.

While that is an immediate solution, it does come with more than a little management baggage. The whole shebang living under one roof does take care of recall, splits, duals, naming, and probably a whole host of other conveniences that would now require way more housekeeping,

[nursers]

Just want to say that I've liked the GX-80 so much, that it's featured on 3 of the 4 tracks on my next EP, and I've just needed to write some sort incidental music to segue between 8 songs at an album launch, and five of em involved the GX-80  

[Justin Havu]

Sounds absolutely spectacular.  I just wish it didn't hog my CPU.

[Anderton]

13 hours ago, justin_havu said:

Sounds absolutely spectacular.  I just wish it didn't hog my CPU.

 

Don't you wish those two aspects of virtual instruments weren't always related?

[Dave Bryce]

On 3/5/2023 at 6:47 AM, justin_havu said:

Sounds absolutely spectacular.  I just wish it didn't hog my CPU.

 

Have you tried messing with the oversampling rate?  It's the tab labeled with the Q the bar over the synth.

 

 

dB

[Justin Havu]

Yep.  It's on normal.  I'd be scared to go any higher than that, for fear of my i5 9600K spontaneously combusting.  Lol

[Nathanael_I]

On 3/5/2023 at 6:47 AM, justin_havu said:

Sounds absolutely spectacular.  I just wish it didn't hog my CPU.

 

I can't say that I care one bit.  I've wished for developers to bury the CPU and hog the machine in service of sound quality for years.  I don't think hungry things like U-He's Diva have hurt themselves in any way by being hungry.  CPUs are fast and getting faster.  Every modern DAW can freeze tracks.  The new M1/M2 laptops are most of the power of a modern desktop.  If the sound quality is there, I don't care at all. 

 

I think that some things like reverb could have a real-time component and a render time component that upped the resolution significantly.  Give me good and make it stupendous on render.  Kind of like editing video with proxy files - there are some things that if the draft is good enough, you'll take the pleasant surprise on render.

 

I get this approach doesn't work as well for live.  But if it runs in a modern M1/M2 Mac, I'm down for it to be greedy, hungry, and demanding as long as it sounds great. 

[MRBarton]

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

 

[Dave Bryce]

7 hours ago, MRBarton said:

I just want to reiterate.  The oversampling buttons are not "quality" controls.  The instrument will sound no better if you use them. 

 

Wait - I'm confused. 🤔

 

Dr. Sigman's GX80 manual has the following thing to say here about the oversampling feature (bold emphasis mine):

 

Quote

 

Q Oversampling Quality - The Q button sets GX-80's internal oversampling rate; the higher the setting, the better audio fidelity will be, with the caveat that more computer processing power will be required. 

 

Internal processing can be set to 1x (same rate as the current sample rate of the host DAW or in the Settings>Audio/MIDI window for the standalone version) or to 2x, 3x, or 4x the current sample rate. The sample rate is downsampled at the instrument output stage to match the current host sample rate. For example, if the current DAW/instrument sample rate setting is 48 kHz, and oversampling is set to 2x, GX-80's internal processing runs at 96 kHz, and is then reduced back to 48kHz at the output stage. 

 

If I'm reading correctly, Mitch is saying that the higher internal sampling rates should produce "better audio fidelity".  Not the case?  If it's not, can you please clarify what it does do?  

 

I also wanna call this line out from that section of the manual:

 

Quote

If the current DAW/instrument sample rate setting is 192 kHz and oversampling is set to 4x, GX-80's internal processing will run at 768 kHz, and you will very briefly hear the most ameeeezing synthesizer sound quality ever experienced by mankind before your computer instantaneously explodes in a fiery, white-hot supernova blaze. Ok, maybe not.

 

 

 

Mitch Sigman is clearly a deeply funny man.  More manuals should have humanized moments like that... 

 

Thank you again for taking the time to educate us, Mark!  Greatly appreciated. 😎

 

dB

 

 

 

 

[AnotherScott]

14 hours ago, Nathanael_I said:

I get this approach doesn't work as well for live.  

 

Exactly. If you're trying to run 5 VSTs simultaneously for live performance, then overhead matters.

[MRBarton]

7 hours ago, Dave Bryce said:

14 hours ago, MRBarton said:

I just want to reiterate.  The oversampling buttons are not "quality" controls.  The instrument will sound no better if you use them. 

 

Wait - I'm confused. 🤔

 

Dr. Sigman's GX80 manual has the following thing to say here about the oversampling feature (bold emphasis mine):

 

Quote

 

Q Oversampling Quality - The Q button sets GX-80's internal oversampling rate; the higher the setting, the better audio fidelity will be, with the caveat that more computer processing power will be required. 

 

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

[Dave Bryce]

I love to learn. 😎

 

Thank you for laying it out so clearly, brother.  

 

dB

[EricP1954]

Just a quick question to throw out to the GX-80 developers' crew - Is there a simple way to assign a different aftertouch trigger to a given preset? Could the mod wheel for example become the trigger on a preset that uses velocity sensitivity? This also assumes that GX-80 is being used in VST mode inside a DAW.

[jerrythek]

There are dedicated sliders for aftertouch control of some parameters, like Filter Cutoff, Amp Level and the Sub Osc. Since you can map/learn MIDI controllers to these the sliders the answer would be yes.

 

On 3/19/2023 at 2:13 AM, EricP1954 said:

Just a quick question to throw out to the GX-80 developers' crew - Is there a simple way to assign a different aftertouch trigger to a given preset? Could the mod wheel for example become the trigger on a preset that uses velocity sensitivity? This also assumes that GX-80 is being used in VST mode inside a DAW.

 

[Dave Bryce]

We are way pleased to announce that the GX-80 is now available for purchase on the MPN Shop! 🥳

 

dB