Organelle synth patches?

Perhaps I’m looking for the wrong thing, and it’s not what the organelle is for, but i find that there aren’t many synth patches available. By synth patches, i exclude the ones the emulate other gear, like melowtron, or moog or the basic analogue style patches.
I mean some big lead synth patches or some bass patches etc, like the 20+ presets you might find on a hardware synth that you might buy.

I mean there are lots of wierd and wonderful things available that C&G and the community have spent time creating, but sometimes I’d just like a bunch a patches with a few parameters to tweak.

Am i making sense? Perhaps pure-data doesn’t lend itself well for this sort of thing

There is fluidsynth~ . It should make possible to use soundfonts on the organelle patches.

If you only care about high quality sounds and a lot of variety of presets, I think this is the way to go. But keep in mind that the soundfont files are sample based, so you will not wave a synth patch like in the Juno patch where you can dig in and see how the sounds are made from scratch.

Don’t know for sure how and to what extend fluidsynth works on the organelle, because I still didn’t try it myself. I Just bought my organelle one month ago, still playing around with the weirder side of things. Maybe someone else can share his experience with this?


My “Aquarius Patch” is a humble effort to create a decent “Virtual Analog Synth” for the Organelle (granted I have a lot of work in front of me, which btw. I am willing to do, just bear with me). I am currently working on a Revision 3 with graphics and better sound, but VA is not that easy and there are a number of difficulties to overcome…

Regarding your question, I think here are some things to consider:

  1. Sound: Good OSC or filters need CPU and the Organelle is not a high-power computer and most of the PD externals are not aimed towards the limitations of this platform…
  2. PD is mostly an “amateur” environment and we can not expect “professional work” (e.g. MAX) … Yet I have to add that so many wonderful people have created so many amazing things for PD and I can’t express how freakishly awesome their work is… (a big thank you to all of you!!! You are truely amazing people!!!)
  3. Regarding interface: there are major “difficulties” to overcome here. Although I really love my Organelle I have to say that the potentionmeters were such a horrible design choice (i can not express how much i hate these knobs…)

Anyhow, I currently exploring virtual analog techniques and I think with some ingenuity we can overcome the limitation of the interface (using graphics and the aux button as shift mode)…

Can the Organelle be a good VA? Certainly, but it is up to us to do it…


There are many decent implementations of bandwidth limited oscillators that sounds pretty decent.

Dont agree with you on that one. You can make perfectly well professional stuff in PD. It might be a bit harder to make in PD than in Max, but in my opinion you can make stuff to professional standard in PD.

Just curious… What is it, that you don’t find professional?

You just need to make a paging system and maybe a preset system and you got all you need. I think the knobs are decent, but yeah could always be better. I would rather have had endless potentiometers/encoders, that would have been nicer. But the ones we have, I think they work pretty okay.

It has all ready been done several times.

1 Like

That’s not what I said. I said they need a lot of CPU. Try to make 5 voice Polysynth on the Organelle and you’ll push the CPU to its knees. And don’t forget that there are BL-OSC the Organelle can’t handle. I tried several and a number of them will not work well or sound very good, or have other problems…

Ok, here I was a bit unclear. My bad… What i was trying to say is: PD is not a professionally curated platform and for PD you will have to find externals and abstractions for certain things to do. Which boils down: It is harder to do things in PD (and will not necessarily sound equally good) in comparison with MAX

I personally find mult-ipage interfaces extremely annoying and impractical to use. Maybe you’re fan of them, I just don’t like them… Although I do believe you could improve on the design by using the encoder to directly scroll through the pages (with no menu), i feel that by not having endless encoders it’s just a slight improvement and not worth exploring further…

Other issue: Not sure if it’s method regarding how the potentiometers are read or something else, but a full twist of a knob will not get you the consistent results and especially with GUI it becomes a real problem when a fully turned not does not equal max. value (and maybe only 0.7 of max) and values jump when you turn it back again… As I said not a fan and it makes making patches a lot more difficult…

Has it? Look, I’m not trying to challenge you here, but OP is asking for good synths and if you have examples you find good, this would be the place to share. I personally don’t think that the (poly-)synth engines for Organelle are super convincing or very versatile (and I not trying to badmouth here). And to honest my crack at it (the Aquarius 4 patch) does currently not hold up to comparison with a good VA. That’s all…

1 Like

pots vs encoders - no argument there, but in fairness I have the same issue with my Virus TI
(which is multi-part/timbral , so when you switch parts the panel does not match synth engine)
that said, encoders designs tend to have to rely very heavily on a display - though thats not really an issue for the Organelle.

multi-page, I agree , its why I like the original C&G patches so much with just 4 parameters.
but as you know, building a synth and just exposing 4 parameters is quite a limitation.
(hmm, gives me an idea for orac :wink: ) )

the solution is midi mapping - if you want a complex synth on the Organelle, and you don’t want to menu dive, then support midi learn - midi controllers are cheap, and some pretty small and you could even use an iPad if you wanted… theres absolutely no reason to limit yourself to the Organelle UI.

limited CPU, yeah thats a given - but it’s also at a ‘price point’ for a given functionality.
also much of the limitation is because we are using PD, which gives us versatility, but at a price.
e.g. consider things like volca they likely have much less CPU than an organelle, but its specifically coded in (likely) C++, so get more out of their modest resources.
(not that either are going to compare to your desktop/laptop!)

note: there is nothing stopping you writing the oscillator (or indeed entire voice) in C/C++ and exposing as an external to PD - which is whats happening with brds (MI Braids)

max vs pd, hard to compare really - theoretically, being both digital they can sound the same.
main advantage (for non-programmers) of max is gen~ , gives you performance close to c/c++ using externals - but its a moot point, as Max is not available for ARM :wink:

back to OP, I think synths can be written in PD, but don’t expect Serum/Massive :slight_smile:
I’d agree theres not that many, but I think thats partly down to a lot of users liking the ‘experimental’ nature of the Organelle - so it’s where efforts have been concentrated.

(looking at posts here on the forum , I find it amazing how many different uses Organelle has, Id say as many are using it as an FX box, sequencer, sample playback - as are using it as a synth… I think, its just the keyboard makes it look like it should have lots of synths :wink: )


Sorry, my bad :slight_smile:
Anywway, I think there are a few that doesnt use too much CPU.

Yes I agree, I think… I never used Max extensively, but I noticed the obvious differences, like UI and so on.

I agree, its not the most fun thing to work with. One knob pr. function is to prefer. But yeah, I guess we have to work with what we have.

Personally I would love to see some kind of controller extension for the Organelle, something that you could connect and get som more knobs and buttons… But yeah any midi controller could do I guess, but using midi controllers and integrating them into patches makes it a bit harder sharing stuff.

But yeah, working with what we have, which is only 4-5 knobs, a paging system is necessary or else it is going to be a very limited synth. Of course one can do macros, but yeah, 4.5 params is still not a lot.

Sorry, maybe there have not been a good one. To be honest I dont use the com.lib too much, I have just seen a few patches around that could be good. But sorry I dont remember the names. I prefer to make my own patches. Anyway, I have tons of VA synths(Micro Q, Blofeld, Virus, Nord G2), so I dont really use Organelle for that purpose, I used it for granular stuff and sample stuff, cause my other synths dont do that.

I apologise if my comment was a bit judgemental. Didnt mean it that way.

1 Like

I think I was airing certain frustrations in my original post I had with working on my poly-synth patch (and running into tons of issues, which were mostly my own fault, yet not all) and with implementing the knobs in general… So sorry for that as well.

1 Like


If I may give a few suggestions:
In this library, shadylib, I think there are some decent and also cpu friendly band limited oscillators.

Or maybe try this approach here:

(See the last part of the video)

In that approach you sample maybe 12 different waves from a BL’ed osc. And then you change to a new wave each time you reach a new octave. That way, you only need a table to hold the current playing waveform and then a mehcanism to choose the right waveform when it switches octave.

Also if osc’s are not used, you could turn them off using switch~ to save som DSP.

For filter:
Yeah , that is one of the hurdles I have not been able to get over yet. I think most, if not all PD filters sounds too digital to my taste.

Oh , I found one, called muug~, (NOT the moog~ filter)… Which is a ladder filter. The muug~ has a different sound to me and sounds pretty good, but sometimes it behaves weird/unstable. And also not sure of it even works for Organelle.

I actually like the bob~ filter. It is decent.

But yeah, both of these are only lo pass filter, if one wants a state variable filter, I dont find any really good ones for PD. And SVF filter is pretty important for VA synth, imo.

If you know how to use biquad coeffs you could probably whip something nice together :slight_smile:

But yeah as I think I mentioned above, I use Organelle for almost any task which is not VA synth. But yeah could be fun to make one :slight_smile:

Hey, thank so much for the input!

I did try shadylib, but if I recall correctly I had issues with aliasing when using the square-wave and pwm did not work very satisfactory (although I think they were a bit more CPU friendly). The MMB library has all OSCs as wavetable look-ups and it does sound really good, but the organelle got thoroughly crushed by them.

I do use the switch~ object to turn on&off effects and oscs in my patch (which got me in a whole lot of trouble, as I didn’t realize i had to rebang the transistion function so my oscs wouldn’t alias… it took me weeks to figure that one out)…

I think as well that bob~ is the currently the best sounding filter, sadly bob~ does use up a lot of cpu-power. There are some filters in the “else” library which could be very promising and which I’m planning on trying out, although its LPF is 12 db…

1 Like

Wanted to second this – I finally set up your mutable patches through a Midi Fighter Twister yesterday and it was a whole new world!

I’m not sure I understand all the references to VA (virtual analog?).

I am very interested in digital synths on the Organalle and was wondering why there weren’t that many.


Yes, VA = Virtual analog.

Although Virtual Analog is mostly marketing term I was referring to DSP-synths that do classic subtractive synthesis in a convincing manner and did interpret your original request as asking for such a (subtractive) synth with many tweak-able parameters. My first answers was just concerned with pointing out certain “limitations” and “challenges”, that we face when designing such a patch for the Organelle… (although I feel could have formulated certain things better… so my bad)

If you ask about “digital” synths I’m assuming you may be referring to FM, phase-distortion and wavetable synths. And I feel that are many great patches out there for the Organelle (incomplete list):

Not to forget: CZZ & Zone patches from C&G… From my POV I feel that’s really decent number of patches…

1 Like

By digital, I mean not analog :slight_smile:
For example, I’d call my MS-20 analog, and my Novation Circuit, I’d call it digital.

Is it possible to create a bunch of pure-data synth patches and load them onto the Organelle, that sound like this or am I completely barking up the wrong tree here.

I think the discussion of capability here overshadows the more prescient point of narrowing down what you want here.

Yes you could conceivably create a bunch of synth patches and load them in (relative to cpu/dsp limits as noted above), but it does sound kinda like you are barking up the wrong tree insofar as it seems you are prioritizing the potential utility of the thing over the pieces it is made of, and the very nature of the organelle is kinda the opposite: its all about the pieces and the loose ends and just… providing a concise/reliable interface for the rawness that is pure data itself (or sc…).

The c&g patches, orac, other community patches do tend to put emphasis on capabilities and what you can do with this little computer with knobs and buttons, but in truth these are all exercises of pd programming within a template interface (that we are all now lucky enough to take and use).

People will still make patches, and some will hit what you want and some not.

Like… it sounds kinda like you want to make some bread, and you can either buy a mill and some wheat, or you can just get a bag of flour. With the mill and wheat (organelle), you can grind and process flour to your exact speficications, but its complicated and it takes some work! But you also can find lots of good flour ready made, in different specifications and lots of options (like the novation thingy), and thats cool too.

“capability” is complicated here