Free Yamaha CS80


#1

Hi guys,

Want to own a real Yamaha CS80, that classic everybody drools on about?

Well don't! I own one and I constantly live in mortal fear of it breaking down.

I have however learned an awful lot from the beast and it is still my personal benchmark for what a good analog polyphonic synthesizer should be capable of.

So is it any wonder my main Axoloti project (see Holy Grail) contains quite a few CS 80 genes? And since I started the project with working out how to build an analog version first the similarities became even bigger (see Poly Power User Tips). So much so that creating a functional CS80 copy (sort of) became an almost unavoidable next (side)step.

Here are the screenshots:

And here's the file:

CS80ish.axp (18.9 KB)

I hope you like it.

Some remarks:

Since the Axoloti has not enough internal processor power it can alas not be a totally integral polyphonic CS80 (see points below) but it is close enough for most purposes.

The biggest differences when compared to the original are:

  • 10 paraphonic oscillators (so 2 more voices to play with then on the original. Hurray!)
  • Because of the paraphonic approach there are however some triggering issues. Axoloti's option to provide non-retriggering when playing legato (binding notes) however goes far in compensating this.
  • Only 2 filters in series per channel so no full filter treatment per voice (It's paraphonic you know!).
  • Only 1 oscillator mix per channel (on the CS80 each channel has it's own independent oscillator mix. In stead you get detuning per oscillator / waveform (so you do at least not need a second set for that) and full level-mixing in stead of only switching waveforms on and off.
  • For true CS80 fans: This instrument also provides poly aftertouch on oscillator volume. Since playing the aftertouch will also monophonically open the filters at the same time the difference to real poly aftertouch is remarkably small (thanks to @SmashedTransistors for his help with implementing this!).
  • To avoid the oscillator subpatch surface becoming too cluttered I have left some controls hidden within the subpatch itself. These are:
    • Oscillator detuning
    • Initial pitch (swoop) attack time
    • Pitch bend range
  • I have consciously decided against sending everything through a stereo chorus as on the original because one then still has the option to treat the “channels” individually after they leave the Axoloti (One of the few things the CS70M really did better then the CS80) but even without extra effects it can be used for some interesting true stereo spreading.

The sound that is set up on the downlaod is rather typical for a CS80. There is the “Emersonesque” attack swoop per note but also listen to the more subtle speed changing modulations created by the ring mod when a note is held longer, proving that a ring mod can be used for much more then screaming wails.

By the way: the ring mod envelope is full ADSR in stead of an AR only as on the real thing. In other words: Some you loose. Some you win.

Have fun and feel free to mess around with it as much as you like. Since it 's a quite reasonable polyphonic compromise anyway it should also be a good starting point for other projects. So knock yourself out if you want to use it to build your own Axoloti Prophet 5 on in stead.

Finally: Since the Axoloti's output signals are not that strong because of technical limitations I have added 2 output gain modules. In the download they are set to factor 1.0. Treat these with care if you do not want to end up with bleeding ears!

Marc

(Remark: To keep this post up to date but at the same time avoid unnecesarry confusion for people who see this for the first time I have updated it, including the download, on 08 December 2017 and 13 December 2017)


#2

Really nice to see something like this from you, Marc, and funny enough, It did cross my mind when reading that CS80 write-up of yours that maybe one day you would attempt something like this with your Axo.

Believe it or not, I have actually managed to misplace mine, been moving stuff around and it's in a box somewhere, but I thoroughly intend to dig it out just to use this. I've always wanted a real CS80, who hasn't, but even if I had one there would be that constant fear of it breaking down and needing repair.

Really looking forward to trying this, especially as it's been created by someone who actually owns a CS80, and is passionate about both it, and detail.


#3

Thanks for the kind words, Axoman. Glad to inspire you.


#4

I really look forward to playing with it, I have to ask though, what about the filter?
Did you model a filter on the CS80 or are you using using a supplied one?

Just in case you modelled it, maybe have a go at creating a whole set of modeled filters. As well as the CS80, other modeled filters I'd love to see are the ones used in the Oberheim OB-Xa, Roland JX-8P, and Korg MS20. Would definitely be cool to have a set of filters modeled on those we've come to recognise. I think maybe that might be pushing it a bit though. I asked about having a modeled SSL Buss Compressor and if I recall there wouldn't be enough power.

Not sure if a modeled filter would be any less of a resource hog, but if it isn't, it would definitely be cool if one of the math/filter wizards here made a pet project of it.


#5

I think one can take the whole imitative modeling bit too far. I am looking more for a solution that gives me as much real-time creative modeling control as possible (see post 166 in the Holy Grail thread).

In the case of the CS80 a lot of it's character comes from it's rather unique layout. This aproximation has been built out of existing Axoloti elements but reflects the originals layout specifics. The low and high pass filters in series for instance. In such a double channel layout it provides the chance to shape 4 resonant peaks and thus fixed overtones.

Understanding the beast is much more important then owning a perfect imitation. Not that I would mind to do my own bit of coding. Idea's aplenty (see the same post). At the moment I am however already busy enough exploring all existing possibilities.


#6

P.S. It sounds very much like the original anyway. What helps a lot is that Axoloti's oscillators sound much better then many a VST's. Often they seem to use the same shitty one loop standard waveform samples, probably as provided by a still popular early development platform.


#7

Yes, I've noticed the Axoloti oscillators sound very good and analog. One of my first Axoloti patches was just a couple of sawtooth oscillators and a filter and not much else, but it was really inspirational in terms of sound quality. It's details like that that really make the Axoloti platform shine.

Nice to see a CS80 patch! And done by someone who has access to a real CS80 too (I sold mine many years ago). I need try this out, also I'm a bit curious as to how many voices you'd get if you go for true polyphony. Just getting 4 voices would be useful in many cases.


#8

I know what you mean, some VST use static repeating waveforms whereas Axoloti appears to use a live mathematically produced waveform.

Haven't played with this yet, but will soon.


#9

The factory oscillators use bandlimited transients (sometimes called "BLEPs" in the 2000s or "EdgE" in the 1990s).
They provide signals that are wideband and nicely antialiased.


#10

The term sounds like Star Trek technobabble but it surely works great!

Do I get it in so far that a mathematical function is used to smooth oout aliasing transients which are already in a signal? Or is it a "pre-treatment" filtering out of peaks that wil otherwise lead to aliasing? The term bandlimiting then suggests that it is only done over a certian (higer?) frequency range. Right?


#11

Be it BLEPs or PLEBs, Axoloti does sound good, so hopefully Marc's CS80 will do it justice.
Will let you know my thoughts on it as soon as I find the bloody thing!

Axoman - Living Life on the Cutting-EdgE of Technology :crazy_face:


#12

Next: A Brassé built Axoloti enclosure that is as big as a desk. In other words: I know exactly where mine is! :stuck_out_tongue_winking_eye:


#13

Hi,

I've just updated the initial post to inlcude a few improvements (the poly aftertouch now actually works and I have also added L and R channel output attenuation pots). Those who allready downloaded the original version can now download the newest version by the renewed link in that post.


#14

I need try this out, also I'm a bit curious as to how many voices you'd get if you go for true polyphony. Just getting 4 voices would be useful in many cases.

It should then just about do 4 voices. "All" you have to do is to copy all the stuff in the main patch excpet the subpatch itself and the output stages, press edit on th esubpatch and copy it into the subpatch screen. Then (re)create the right connections. After that you can delete the superflous elements. You for instance only need 1 MPE.

A distinct disadvantge is however that to access every prameter in a subpatch you also have to add every such item on the subpatch list by richtcliking on the pot in the subpatch (bear with me) and choose the top "connect to to parent" option. That however leads to a long list of parameters on the subpatch opbject in the main screen which does not have a graphic relation to the layout of the synth (it actually lists all elements with a top left to low right priority!).

The only other way to inlfuence a subpatch live is to use MIDI CC messages or a hardware controller box connected to the Axoloti's boards analog inputs.

In other words: Beside the controller listing on the subpatches surface going truly polyphonic in a simple setup only really works if you work out an instrument in mono, program a host of sounds (still in mono) and use the polyphonic subpatch option to only drive presets!

And better not complain about it to loudly becuase I recently got into a row with the webhost for doing just that. :expressionless:


#15

So do I, cause I just found my Axo but I need to rig-up my keys again first so I'll leave it until tomorrow now, but looking forward to it. I assume from the "big as a desk" comment that you plan to make a full-size MIDI controlled version with Axoloti inside?

Well I hope you're a good carpenter, Brasse, cause such a thing would look amazing if it were craftsman-built from wood :wink:


#16

Now then, Mr Brasse ... opinion time!

Had a play with your CS80, and honestly I wanted to praise it, but I can't. I hope that doesn't sound negative, cause actually, it's quite the opposite. I've drooled over the CS80 many a time, and had it not been for the price of the thing, I would already have one. But your CS80 has made me aware that I wouldn't enjoy that synth even if I had one. Certainly not as much as I thought I would. It's one thing reading about it, but it's another when the functions are there in front of you, it makes you realise certain things.

I had not noticed the lack of LFO, nor had I realised how cumbersome it feels to go from oscillator to output. CS80 sounds awesome, but if I had a CS80, honestly, I think I would get tired of it's structure or 'architecture' pretty quick.

As for your specific version, I'm surprised you didn't set it to the classic Vangelis Bladerunner sound by default, so that was a bit of a bummer, but all that aside, my opinion after experiencing it is that it's an interesting architecture but not for me, it's not a synth I would get along with if I had one.

So your CS80 didn't inspire me, but it definitely did help me out in other ways, only thing is, I'm not sure whether I like the idea of no longer wanting a CS80, lol, it feels very wrong somehow :open_mouth:


#17

Sweeet, gunna try this out tonight!


#18

Sorry, Axoman but you really do not get it.

This is no perfect replica of a CS80. I did not call it CS80ish for nothing. It is a model that can help you to create typical CS80 sounds on an Axoloti. I could go on trying to get even closer to the real thing and clutter up the user interface and deminish the voice count in th emean time but like I said in my introduciton: This is more of a sidepath then the main stream.

Some specific remarks though:
- The CS80 has a dedicated LFO. A very flexible one actually. So intall one if you want one.
- More imporant though: Typical for the CS80 is that it actully already has a dedicated LFO for about every worthwhile function. So there is a LFO for PWM, and there is one for the Ring modulator in this model so yu can already modulate the hell out of this thing without an extra routable LFO. Crank open the PWM LFO and the thning even moves into FM, something the real thing never could! Understand the Ring MOD's LFO/Env. generator combi and you wil already modulate to Nirvanna without ever feelling the need for anyhting else.
- The layout of a CS80 is very logical but in this Axoloti "recreation" other things where more important. If you do get the whole Paraphonic approach I talked about you might actually understand why the "VCA's" are actually in front of the Filter.
- One of the most important features on the CS80 are its velocity and polyphonic aftertouch facilities. As long as you cannot try out the real thing the best way to get a inkling of the CS80's potential is to first get an understaning of what polyphonic aftertouch can do for you in general.
- More importantly: Even when owning the real thing one could still circle around it for years without getting to grips with it's real potential. Playing around with synths is more then pressign one single key and tweaking the filter control and LFO speed to death, in th emena time constantly ignoring the ADSR's in the process (I'm basically describing about 80% of all Youtube "demo's" with that, Grrr!).

I could go on for quite a while only talking about such stuff but I think the CS80's reputation is deformed anyway. When you talk about certain "classic" instruments people immediately start building their own personal myths around them. The fact that you ask for THAT BLADE RUNNER SOUND is already a very bad sign. people running after THAT sound has already created an Internet vortex in itself.

You know what. THAT sound is basically a very simple sawtooth construction. The special thing is how expressive something like that can get when you play it well on a CS80. Scoop: If you know what you are doing you actually do not need a CS80 for it!

Again: I could go on like this for hours;

Here's some dedicated information though.
- Understanding my paraphonic approach: https://sebiik.github.io/community.axoloti.com.backup/t/polyphony-power-user-tips/3957
- To get a better impression of the real things layout: https://www.manualslib.com/manual/778093/Yamaha-Cs-80.html
- Taking a deeper look at what makes the CS80 tick: http://www.brassee.com/writing.html#cs80
- If you still insist on THAT sound: https://www.youtube.com/watch?v=Fne0oIEv-WI

Please work through that first and then come back again, providing you are stil interested. :slight_smile:


#19

I thought it would be more interesting, inherently, as well. But then again, we lowly troglodytes just don't get it.

There was a VST emulation I had at one point that seemed to produce a huge variety of different sounds, I think the problem may be that the layout/knob/toggle access is what makes a huge difference?


#20

As far as I am concerned there is no right or wrong here. This has been discussed before and wil be discussed many times more again: It's not a matter of being troglodytes or not. Sorry of that is the impression one might get from my explanations.

Everybody has to do music his own way. Getting into the deeper side of Axoloti however means that one starts to conceptualize what a synth does and getting the best out of these insights. Some people here go a lot deeper then I do (for instance into coding and the arithmatic behind sound). They in their turn humble me.

I do however dare to say that if you are not prepared to go rather deep there is not that much sense in "going Axoloti". You might find loosing interest quite quickly after the novelty aspect wears away. Better buy yourself the next VST, a hardware synth or even a "real" modular then and start turning knobs and stuff at random. Many a fine sound has been created that way.

What I have merely been trying (again: as a sidepath) is to set up something that can make convincing CS80 sounds. Using the name of one of the synths with the most overblown reputations (even if rightly so) might however have been a mistake. I probably should have called this post "Free Quite A Bit Yamaha CS-family Sounding Axoloti Patch If You Are Not Too Particular About It" but people nowadays simply need hyperboles to become interested at all.

As far as a comparison to VST's is concerned. You can take it from me that my thingy sounds much more like the real deal then any VST I've heard. I wil however not make a whole set of presets to prove it. That would actually be defeating the whole point. Troglodyte or not, in one respect Axoman is right. If you do not get this on your own strengths you'll probably also not get the original.

What about something like this however?: "Hey, you forgot the modulation generator but I just added one myself by linking a thingemajig module with a whatzitcalled invertor. Here's my download in return!" or "Man, I just found a way to add polyphonic portamento (Oops, just gave away another "glaring" ommision!)".

Now that sort of stuff would really warm my heart. As far as the "It's not the real thang, dude!" part of the discussion is concerned I would however like to quote the big bad computer in Tron:

END OF LINE

:sweat: