Posts made by join

We had a chance to try the R1 recently. It is very nice to hold and the lights are neat. It is targeted at beginners and occasional players with very low expectations regarding performance and quality, and the sound set is aimed at the asian market.

To me, the NodeBB developers are clearly better at making claims than fulfilling them. I don't believe their moderation tools are unparalleled. Discourse looks like a much better engine overall.

We already do IP and some email pattern blocking. User moderation is more difficult than it sounds because these bots arrive in huge numbers and have become quite sophisticated. Often they sign up with just a name and an inconspicuous email (typical actual users of this forum also have weird emails, which makes the distinction more difficult), and they stay like this doing nothing for a random period of time, only to "reactivate" one day by sometimes putting a photo, a website, and then posting spam.

Anyway, yes, there remains a human element where we haven't put enough effort yet. The best thing would be to approve new topics, but it means we should organize to guarantee a maximum response time in order to avoid stifling new discussions.

@peter-ostry It's not that we don't care much about spam, it's that we're hit really hard by it, despite several (including paid) automated protections. We're currently pondering whether to transition to another forum engine to make this less of an issue.

@zelio This has always been possible under macOS. Most people with this issue are using Windows drivers that present this limitation.

In a future update we could maybe consider having a separate USB-MIDI device port dedicated to the editor (or if not possible at least let you dedicate an existing port to editor communication, which would prevent the port from being used for MIDI routing).
In the meantime you can leave the first USB-MIDI device port to the editor, and use the MIDI DIN ports for your DAW, as @frank suggested.

At this point, we do not plan to release a VST/AU version of the Anyma Phi.

Unless a new Sylphyo appears with 10x the processing power, it's unlikely that it's gonna be able to rival computer-based physical models of winds, so no offense taken :).

As for directions for the future, I guess it'll focus on the controller part (reliability, ease of switching between different configurations, more options) about as much as the sound part.

We do not plan to stop producing, developing and supporting the Sylphyo anytime soon, however new developments will arrive much more slowly than before the pandemics.

Our intention has always been to provide instruments that can evolve over time and continue to work independently of computer OS evolutions and even of our own existence as a company.
Since its introduction six years ago, we published 20 free firmware updates for the Sylphyo, much more than most competitors, and we intend to continue as long as possible.

But you guessed it right, a small team fighting amongst giants is not an easy adventure, even more so when you're doing niche products. And COVID and the semiconductor crisis didn't help.
We've been facing significant supply chain issues and cost increases, hence why the case disappeared temporarily, for instance. Even just making internal prototypes has become much harder and trickier due to this.
Delivering the Anyma Phi took us some time, and maintaining production of both the Anyma and the Sylphyo is a very time-consuming activity in itself.
Recently we've been busy preparing various trade shows (Synthfest, SuperBooth, NAMM) where you will find both the Sylphyo and the Anyma Phi (yes, not just at NAMM).

Even though we're growing in the face of all cosmopolite microbes and Suez canal obstructors, we're just not enough people yet to have a decent capacity of parallel processing, so we have to deal with things one after the other.
The result of all this, for the Sylphyo, is that v1.4.9beta is one year in the making despite being a relatively mundane update, and new developments will arrive slowly.
The best way to significantly speed things up is to get more people to know and buy our products, so we're very grateful for @Clint's work and we encourage anyone to spread the word about us.

I totally understand that some people find this an uncomfortable position to be in. If you depend on a product to make a living, you need it to be able to evolve at your own pace, and it can't be far behind your needs for too long before the situation becomes untenable.
I'm in that position myself regarding computers: I support small open-source hardware and software companies and buy their products, but I can't "wait for them to get their sh act together" because I've got work to do, so I do my work on computers from a behemoth company whose ethics I despise and end up using both. The behemoth doesn't care about my use cases, and the little ones care a lot but is slow to support them.
Either way I'm angry :), but I'm seeing more and more good things trickle down my way.

Now, what's left for the Sylphyo?

Major updates take time, and cooperating with others wouldn't make development shorter because it's often not a problem of not knowing how to do or finding solutions, but rather of finding the time to develop them and ensure they won't cause further issues down the line.
We've been carefully listening to all the feedback we received along the years, and this shapes our roadmap for future major updates. But there are many things that could be introduced much earlier in minor updates, simply because they're easier to do.
It's impossible to tell from outside if a feature request is trivial or not, only we can tell. For instance, the fingering changes Clint submitted were easy to do, but sometimes a seemingly insignificant fingering request could require major rework in some parts of the fingering system. Sometimes a blockage dissolves, and a bunch of requests change from hard-to-do to trivial, but we're not sure there's still interest in it.
That's why it's important to us to constantly get as much feedback and ideas as we can, and we encourage you to do so, keeping in mind that only time will tell which of these ideas will end up in an update and when.

I'll conclude this by giving maybe another perspective on the developments at Aodyo in the past few years. Yes, during this time our major focus was the Anyma Phi, but it would be an error to see the competition in Aodyo resources between the Anyma Phi and the Sylphyo as a zero-sum game. Many developments originating from the Sylphyo's firmware and synth culminated in the Anyma Phi we have today, which is much more than a few Mutable algorithms slapped together. The new knowledge we gained doing and perfecting the Anyma Phi will ultimately come back to benefit the Sylphyo in some form or another.

…if you live long enough to see it ;).
(just kidding, back to work)

Yup, that's about right!

Hi Thomas,

It wasn't practical explaining the whole thing in the release notes, but you can find them in the Fingerings section of the user guide. I've highlighted the changes in the picture below.

The Link gets state information as well as the current Sylphyo settings, and with those settings the same piece of software turns that state into MIDI in the Sylphyo and Link.

I'm carving a few minutes from my vacation time to type this, so there might be more details later on (and answers to other posts as well).

As @Peter-Ostry said, there will always be a fundamental difference in how pressure builds up in the Sylphyo compared to other closed-ended wind controllers, and this might translate into a slight amount of latency in the attacks. As for how it impacts perception, it looks like it's a very personal appreciation. Some don't like it, but it seems more due to how they prefer the tradeoff offered by the other solutions. Some prefer it that way and even say it's allowed them to feel less latency. For instance, one of our artists, Florian Becquigny, quit using his EWI for very nervous pieces because he much prefers how responsive the Sylphyo is in terms of latency. He doesn't seem to care much about the sensitivity aspect of it, so he always plays at very high pressure and completely plugs the bottom hole.

As for the comparison to a Lyricon setup, the latter should offer objectively lower latency (especially on an all-analog system), but again I'm not sure the perceived difference is that important for everyone.

Here's a brief overview of the pipeline involved in turning your breath into sound on a Sylphyo playing its internal synth over headphones.
First, the Sylphyo gathers samples from the breath sensor at 15 kHz, which it then turns into a filtered 1 kHz signal. Keys and other sensors have a lower rate, especially the keys as it often includes the configurable delays that are there to make the thing playable for a human being (but Florian Becquigny, for example, disables all of them because he needs to play pieces at extreme speeds, much than the EWI allows him, and he has done the practice to be able to play like this).
But let's focus on the breath for now. As soon as a breath sample is available, it is sent to the internal synthcard in a "state packet" that takes a bit more than half a millisecond to travel. Once received, it is made available for the internal synth engine, which operates with a fixed 1ms buffer and transmits 48 kHz audio to its codec.
You should add some time to account for CPU interrupts, context switching, and other high-priority concurrent processes, but it shouldn't add much to the overall latency as individually these things are counted in nanoseconds or microseconds.
The path to the Link receiver is obviously a bit more complex due to the fact the data must go to the radio package, over the air, then from the Link's radio package to its main board, and finally to the Link's internal synth, thus latency increases a bit.
Overall, the pipeline was pretty much designed for low latency, and we had to make significant departures from off-the-shelf peripheral driver designs to get there, so I'm not sure even an Arduino-based controller with CV output would fare better.
(I'm not sure my colleagues would be OK with the amount of info disclosed here, so I might redact some parts of this in the future)

You should also factor in that perceived latency will also depend on what kind of sound you play; i.e., what's the synth doing. Envelopes, delays, internal debouncing mechanisms in percussive oscillators might well be more impactful in terms of latency compared to the software pipeline I've described above.
And you should also note that digital synthesis will always respond a bit "behind" compared to analog, because in the digital world you have to deal with a finite sampling rate and resolution, aliasing and other unwanted effects whose mitigations always incur a bit of latency (for instance, smoothing a filter's cutoff frequency). This can be solved with more processing power, which leads to more expensive instruments, or with a simpler synth architecture and more limited sounds.

But overall, the difference between a Sylphyo played over headphones and a Lyricon shouldn't amount to much compared to an average human's just-noticeable-difference in latency perception.
I think we're already crossing a qualitative threshold between a square wave played through the MIDI-USB-Computer path and the same square wave played through the internal synth, as "it feels more like it's coming directly out of my head".
I bet the full-analog Lyricon experience is even better and more enjoyable, but I don't expect it to be "a whole another step" better for a majority of people.
To me, it depends on individual perceptual experience much more than everything else.

The same story seems to repeat in various different fields. When I was doing research on the accuracy limits of the perceptual-motor system, prior to Aodyo, we found that it was pretty much impossible to come up with a formula that would work for everyone. Even without prior training, we had people who could easily do and perceive the equivalent of surgeon work using a high-resolution mouse, while others weren't able to fully utilize the resolution of a basic 2000's-era office mouse.

Peter's suggestion to look into your acoustic instruments is a good one. Sure, once the air is moving, reaction is pretty much instant and action-perception coupling is ideal, but there are a few things that incur latency even in an acoustic context, such as the time to build up air pressure, the key mechanisms if they exist, etc. And we don't necessarily perceive them, or at least we can adapt to them.
More than a decade ago, I worked on a video-based computer music system where latency varied greatly due to the fact the input device was a noisy webcam. In this case, the solution was to add latency so as to arrive at some outrageous number, but it was much easier to perform with it, because latency was predictable and always the same and the body knows how to adapt and anticipate. That human process of anticipation is even the modeling basis of the best score-following systems (used to accompany the varying tempo and expression of a live musician with a predetermined score).

Digressions apart, we could progress towards lower latency by shuffling the software around and maybe making different tradeoffs. I'm not sure if there have been big regressions since the earlier firmware versions, but if there are then we would have an easy first step: correct them :).

Great idea. It'll be published along with the next minor update.

I'm holding tight to my headphone jack bearing iPhone, so unfortunately I couldn't say how to make that cable splitting trick work with the Lightning port.

What you can do if you haven't already is just disable the wireless transmission. This will reduce power consumption a bit.
As for further reducing the power consumption of the Sylphyo, we'd like to explore options for that in a future update (notably, turning off the internal synth, although not everything can be turned off), but there will always be a slight spike in power usage when turning the Sylphyo on.

Normally, you won't be able to power another device from the Sylphyo's USB port.
A better option would be to look for a powerbank and/or a USB switch that could supply the required power to the adapter and still let it connect to the iPhone and the Sylphyo.

I've been looking at Yamaha SessionCakes a while ago. I don't know if it could work (it would require to have the Sylphyo and the iPhone as separate inputs and mix them to an output jack), but it looks promising. I'm sure there are other companies making portable mixers as well.

Please send us a mail to support@aodyo.com, with the version of the Anyma Phi firmware you're using, as well as the exact version of Windows and as much detail on your hardware as possible.

If you look at the settings of your Sylphyo, you can see that what you are controlling is called "Slider bottom", and sends CC 13 by default, if you didn't change that setting.

Now that we know what you're sending to the Anyma Phi, you just have to check that the Anyma receives it.
In the Anyma Phi settings, go to "CC mapping". If you didn't change any setting, normally you wouldn't find CC 13 anywhere, because it isn't in the default configuration. So now, you will have to select a control (A/B/C/D) depending on which you're not already using, and map CC 13 to it.
Say we're mapping Control D to CC 13.

Once you've done that, you'll be able to use Control D as a source in your patches.
So for controlling portamento, just create a mapping from Control D to Glide, and adjust according to taste.

In addition to a reasonable Reaction time, you'll have to increase Reaction time octaves too.

The pops and beeps you hear are the Sylphyo pick up the slightest action your fingers are doing, because with a short latency, the onus is on you to perfectly synchronize your fingers. And that's pretty hard to do :), because we're not as agile with our thumbs as with our other fingers.
So the two options are training hard, or increasing reaction time.

Now, note that these Reaction time settings only determine the time the Sylphyo takes to take your fingering changes into account, only when you are playing legato (already blowing into the horn, and going from one note to another).
So even a very high reaction time should not increase the time between your breath and the corresponding sound you hear.

Maxence just forwarded the requests to me, so you can just send me a mail directly.

I've sent you one with a link to a beta to help you check your fingering updates, so that you have my address :).

J'espère que les possibilités de l'Anyma Phi vous satisferont :).

Tous les modules que nous avons tiré des modules Mutable Instruments sont des versions "maison" dont nous avons réécrit la majorité du code, de façon à l'adapter aux limites du matériel. Nous n'avons donc pas une copie d'Elements, mais quelque chose de similaire, d'un peu différent, d'où les difficultés que tu as rencontrées.

Le temps de passage d'un son à un autre est principalement lié à des contraintes matérielles qu'il nous serait difficile de contourner. J'ai dit "difficile", pas "impossible" :), mais il s'agirait d'un gros travail, et nous pourrions envisager de nous lancer là-dessus si un assez grand nombre de personnes exprimait un intérêt pour ça.

Il est toutefois possible de faire en sorte que l'Anyma ne demande rien au changement de patch en décochant le réglage "Warn if unsaved".

Sorry, but our desktop software does not support Linux (and even if we provide a Linux build of the latest Anyma Phi editor, we do not offer support for it).
If it doesn't appear to work with the latest version of wine, your best bet would probably be running it on a virtual machine.

The editor should be triggering a save right after importing.

Which exact macOS version are you using, and on which Mac?

Wow, that must have been a bit frightening!

Yes, like all things wireless, as most are more or less restricted by law to the same 2.4 GHz band, when things get very crowded it can quickly turn into a silent fight for bandwidth, and WiFi often wins (hopefully, nowadays WiFi is pushing towards the 5 GHz band and even other higher bands, so in time this will be less of a problem).
The best thing to do to avoid this kind of thing would be to pair the Sylphyo and Link again when beginning the sound check, making sure all the other wireless gear is on-stage too and active, so that the Sylphyo/Link will fail to pair in the more-crowded channels, and will naturally land into a less-crowded channel.

@clint This is a bug related to the former wireless receiver we used before the Link. The Sylphyo's direct MIDI output doesn't have these missing messages.

You have received an answer from the support email.

Please send an email to support@aodyo.com about this and include pictures of the entire Diagnostics screen, which is located at the bottom of the Settings menu.

Have you checked the mapping Amount?
Could you maybe provide the patch you're working on? This way, it'll be easier to determine what's going on.

Thanks for your feedback.
@Paul-Flute @Friedus Any details on your requests are appreciated.

Hey great idea, I wonder why we didn't think of that! Thanks for the support anyway :).
If the Anyma is any indication, it takes some time to make updates. We know Sylphyo users have been waiting for a long time, but let's get real: since last time we've been testing a few new experimental things, we don't have enough new and tested material to release update ready just yet.
Are there specific things you're waiting for?

We had a similar issue pop up a week ago with the latest macOS security updates.
I'll send you a mail from our customer support with instructions on how to solve this.

Yes, Loopop used a Mac for his video. Sorry that it's this way on Windows, but there's nothing we can do about it on our end.

We know some people use third-party drivers like loopMIDI, but I don't know how to set them up so as to copy MIDI to the Anyma.
Maybe you could create virtual loopback MIDI ports with this, then make a MIDI routing application (like MIDI-OX or Bome's MIDI Translator; or maybe your DAW) pass the MIDI data from the controller you're interested in to a cheap USB-MIDI interface that you could plug into the Anyma's DIN MIDI In port.

This is a limitation of the default Windows MIDI drivers: there can't be more than one program using a single MIDI port at the same time. I don't know if there are third-party Windows drivers that do a better job.
Also, it's not recommended to connect the Anyma and the editor through MIDI DIN, due to insufficient bandwidth for the editor to correctly receive all the data the Anyma is sending: always use USB-MIDI for that.

What you can try is :

• connect your MIDI controller to the Anyma using a MIDI DIN cable plugged into the "MIDI IN" port at the back of the Anyma
• or connect your MIDI controller to the USB Host port at the back of the Anyma
  To check whether the Anyma is receiving anything, you can go to the Diagnostics screen (in the settings, it's the penultimate item), and scroll to the "MIDI DIN PORT" or "USB HOST" sections, and then see if the "RCVD#" number increases when you're playing the controller.

These are works in progress and feature demos we sent to Loopop, and we'll release 20-ish new patches soon.

But as they say, sharing is caring…

PolyOrgan.anymaphi-patch
Basic demo of a paraphonic setup, with a Drawbar organ and a Rotary speaker. Matrix and morph can be used.

PiezoSnare.anymaphi-patch
The surface of the Anyma Phi is connected to a snare drum resonator. Hit with a metallic object to add more decay. A few matrix controls and morph can be used.

PiezoBoomTschak.anymaphi-patch
Hit the surface of the Anyma Phi with the pad of the finger to make a kick, and hit the surface using a nail or something metallic to make a snare. Sensitive to hit force and timbre. No matrix controls.

Drowned8BitArp.anymaphi-patch
Simpler arpeggiator setup, with chiptune-style arpeggios. Matrix and morph can be used. The arpeggio runs in the order you enter the notes of the chord. The bass that can be heard with SRC2 high is set to the first note of the chord.

ArpBallString.anymaphi-patch
Bouncing balls falling on a string, with echoes; the arpeggiator is triggered by the bouncing balls. No matrix controls.

@deepsea You can disable the "Large type" option in the setting. As for patch selection, a manual confirm could be added as an option in future updates.

Atmosphere is near the edge of the Anyma Phi's processing power, but it shouldn't normally crackle. If you bypass the reverb, I guess the crackling would go away? Could you please try that, and maybe share your sequence as a MIDI file? Thanks.

Did you encounter the crackling issue before updating to v1.0?

Hi everyone,

We are releasing the v1.0.0 update of our physical modeling synth, the Anyma Phi.
We added tons of new contents for you to play with, for a total of 35 oscillators, 33 effects, and 45 modulators, as well as many new features that will bring you Anyma game to the next level.
This update is the result of months of hard work and testing, and we would like to thank all the people who participated in our beta program and made the product more stable, featureful, and enjoyable.

Get it there:

• Windows package (≥ Windows 7, 32 bit)
• macOS package (≥ 10.9, Universal Intel & Apple Silicon)
• Linux package (64 bit, limited support)
• User manual (PDF)

After updating, it's important to turn your Anyma off and on twice.

Here's a tour of the new features.

Youtube Video

Arpeggiator and MIDI Clock sync
Finally, you'll be able to use that mysterious Arp macro at the lower-right corner of the matrix!
It now enables or disables the internal arpeggiator included in your patch, and a long encoder press gets you directly to all its controls.
The arpeggiator features 15 different patterns, with all the classic options, a latch mode, and even a "trigger" mode where your modulators decide when to advance to the next step.

By default, the arpeggiator is synced to an internal clock, with adjustable tempo, but now that the Anyma Phi supports MIDI Clock sync, you can use that external clock as well.
You can also leverage clock sync elsewhere, with new "sync" versions of the Delay effect and the Advanced LFO, as well as an army of Sync triggers.
Ah, and there's a Sequencer modulator too, with up to 16 steps.

Microtuning support
You can now tune your Anyma Phi to any reference frequency or tuning system, such as just intonation, historic or non-Western tunings, or those of modern microtonal and xenharmonic music.
Up to 8 tunings can be stored, and you can import/export them via MTS SysEx messages (MTS-ESP works quite well too).
You can then select one of them as the global tuning, and assign a specific one to each patch.

The Anyma includes dozens of tuning presets, and you can edit them or create new ones with the onboard tuning editor, which offers a palette of different ways to define a tuning depending on what you need.
Don't miss the Tuning chapter of the manual for more details.

Autofill matrix and Randomize patch
Your Anyma Phi now has a few new commands (in the settings menu, as well as in the PC/Mac editor) that will make your sound design life much easier.
Among them, Autofill matrix analyzes your patch, tries to find the most interesting parameters, and maps them to Matrix macros in the right place, so as to offer a good starting point to further polish your creation.

But the most exciting command is certainly Randomize patch.
Using a revolutionary artificial intelligence algorithm (also known as "random numbers and a bunch of conditionals"), it creates a new playable patch out of nothing, completely different every time.
Some will sound weird, some will sound great, and most will need a bit of polish, but you can be sure it will rekindle your creative spirit and lead you to undiscovered territories!

And with the Randomize empty setting, you will get a random patch automatically every time you go to an empty patch in your bank.
Just roam about in an empty bank and save all the interesting patches you'll encounter!
It's a great way to spend an afternoon, and to feel like your Anyma Phi has gone from 200 patches to ∞.

Drum trigger and Timbre follower
You will find an improved Envelope follower that allows you to extract a modulation from an audio signal (like the internal piezo), now with noise gate and gain.
Map the envelope follower to the Patch expression, and now you can trigger and play the oscillators by tapping on the case of the Anyma!

But wait, we added two modules that will bring audio-based modulation to the next level:

• Drum trigger is an envelope follower designed specifically for percussion.
• Timbre follower extracts the "brightness" of the audio signal, so you can discriminate between tapping with your thumb or with a screwdriver (?), or between singing a low C and a high A.

Exciters and Snare drum
The new Exciter group offers three oscillators meant to be used with resonator effects:

• Bow simulates the raw sound of a bow scratching a material.
• Wind simulates a variety of continuous blowing, breathing, or wind noises.
• Strike simulates a variety of impulses and percussive noises: hammers, mallets, sticks, plectrums, or bouncing particles (using the Mallet parameter).

They bring a new dimension to the two existing resonators, but we couldn't resist adding a third one that pairs exceptionally well with Strike: the Snare drum resonator.
This new physical model is derived from an awesome research paper from the nice chaps at the Center for Haptic Audio Interaction Research (CHAIR).

Drawbar organ and Rotary speaker
The new Drawbar organ oscillator offers that classic vintage sound from rock, blues or jazz, carefully replicating the electro-mechanical tonewheels, as well as harmonic foldback, key clicks, and percussion sounds.
You can scrub through all the included presets, or you can adjust the drawbars manually and create your dream sound.

And because not everyone has a Leslie lying around, we also included a Rotary speaker effect, as well as its stereo version that works with both the Main and Aux buses.

Granular processor
If you need to make mind-bending audio textures and soundscapes (or if you just need some clouds to put your rings into), look no further than the new Granular processor effect, which splits your audio into small grains and recombines them in various ways.

Dynamics processors
A new group of dynamics processors joins the effects modules:

• The Noise gate attenuates the input when the signal is below a threshold, with all the classic parameters, as well as optional sidechain and lookahead.
• The Compressor is a run-of-the-mill feedforward design, with optional automatic makeup, sidechain, and lookahead.
• The Dynamics booster is a much simpler compressor for quick and easy sound design, with two parameters: Sensitivity and Amount (tames or boosts the input).

Virtual analog and FM operator
We added a new Virtual analog oscillator that conveniently offers all the basic synth sounds with continuous waveform selection (sine, triangle, sawtooth, square, pulse).

Now, take this harmless little oscillator, modulate its frequency with an audio input, and you get a wicked linear through-zero FM operator effect to make crazy mixed-FM structures.
You can stack up to 5 operators in addition to 3 oscillators, and each FM operator can be oversampled.

Chaos unleashed
If generative music and fun math is your jam, then you'll love the new Chaos modulator group.
Those modulators react to any trigger and generate sequences according to simple mathematic laws that can evolve from predictable states to complete disorder!
Logistic map, Tent map, Circle map, and Discrete chaotic map offer you four different equations for controlled or chaotic modulations.
And if you're looking for something more like Conway's Game of Life, there's a 32-cell elementary Cellular automaton whose rule and initial state you can freely modulate. It can also output three triggers depending on the state changes of specific cells.

I am the law
Some people play by the rules, but aren't they always meant to be bent a little? The new Algorithmic and Physics modulator groups offer you various algorithms and physical models to play with.

The first group offers an Euclidean sequencer to generate evenly-spaced rhythms, as well as an Hexadecimal sequencer that can convert any 32-bit number into a groovy, juicy sequence.

In the second group, the Bouncing ball simulates the movement of a ball thrown into the air and bouncing against a hard floor.
It comes with its little brother, Ball impact, that simulates up to 8 bouncing balls at a time, outputting the velocity of a ball when it hits the floor, which is especially well suited for use as an expression input to a percussive oscillator.
And because there is never enough bounciness in the world, the Spring-damper system allows you to attach a spring to any input value, and outputs the position of the mass at the other end of the spring.

Tools for paraphonic patches
You can now have each oscillator assigned to a different note when playing a chord, making paraphonic patches possible (with up to three simultaneous notes).
There's also many other things you can do, like pegging a specific oscillator to the lowest note in the chord while another follows the arpeggiator.
To make paraphonic patches even easier, there is now a Replicant oscillator that just automatically copies the previous oscillator (with all its modulations), but can be set to use a different expression signal and pitch settings.

Improved stand-alone capabilities
We want to make the Anyma Phi better for everyone, including Linux users and those that despise computers.
You can now rename your patch, copy or swap modules around, and apply complex operations like Autofill Matrix and Randomize directly from your Anyma.

The individual patch files exported from the PC/Mac editor are now formatted as valid MIDI SysEx messages, so you can just grab them from the Internet and send them directly to your Anyma.
You can also dump any existing patch as a MIDI SysEx message sequence directly from your Anyma, without even needing to open the editor.
Finally, you can now reset your entire Anyma, patch bank included.

Full release notes

• There is now an Arpeggiator module to turn polyphonic note input into arpeggios.
• Added a 16-step Sequencer modulator.
• Added a Delay (sync) effect and a LFO (sync) modulator (in the Change group), which are tempo-synced versions of the same modules.
• Now supports up to 8 tunings (TUN1 to TUN8) that can be imported/exported via MTS SysEx messages. You can select a global tuning, and each patch can select a specific tuning among TUN1 to TUN8. The on-board tuning editor allows you to define a tuning in several ways (reference note, 12-tone octave scale, full tuning table, scale, and n-EDO).
• Added three new matrix commands:
  • Reset sets all the macros of the matrix (or its alternate version) to 0%.
  • Autofill tries to add mappings for the unmapped macros in the matrix, based on your patch.
  • Unmap all removes all the mappings related to the matrix.
• Added three new patch commands:
  • Randomize creates a random patch.
  • Compactify removes the free space in the effects, modulator, and mappings sections.
  • Reload removes unsaved changes.
• Added an option to randomize empty patches.
• Added more options to the Envelope follower modulator (noise floor, gain).
• Added a Drum trigger modulator that generates an expression signal suitable for triggering percussive oscillators.
• Added a Timbre follower modulator that extracts the overall brightness of an audio signal.
• Added a new oscillator group, the Exciters, which are meant to be used with a resonator effect. Three exciters are available:
  • Bow simulates the raw sound of a bow scratching a material.
  • Wind simulates a variety of continuous blowing, breathing, or wind noises.
  • Strike simulates a variety of impulses and percussive noises: hammers, mallets, sticks, plectrums, or bouncing particles (using the Mallet parameter).
• Added a Snare drum effect, in the Resonator group, that simulates the propagation and reflection of the input waveform exciting a snare drum.
• Added a Drawbar organ oscillator, as well as two Rotating speaker effects.
• Added a Granular processor effect that creates audio textures and soundscapes by combining short segments (or grains) of the input.
• Added three dynamics effects: Compressor, Dynamics booster, and Noise gate.
• Added a Virtual analog oscillator with continuous waveform selection.
• Added a FM Operator effect (in the Modulation group) that uses the input as a modulation for its own internal virtual analog oscillator, for making mixed FM sounds.
• Added a Dirty formant filter effect that works pretty much like the Artin VFOF oscillator, but applied to any audio input.
• Added a Ping-pong delay effect, together with a sync variant.
• Added a few new value, gate and trigger sources related to polyphonic note input (Paraphonic group), to assist in making paraphonic patches.
• Added a new Replicant oscillator that copies the previous oscillator in the list but allows for changes in expression input and pitch, which can be useful to build paraphonic patches.
• Added a few new controller triggers (new note played, last note released, note transition).
• You can now rename patches and tunings directly from your Anyma Phi. When renaming, turn kEncoderMenu to select a letter, use kLeft and kRight to move the cursor, and press kEncoderMenu to confirm.
• Added a Dump patch command to send the current patch as a SysEx message chain through the USB device port without needing the editor. A patch can then be dumped back to the Anyma as SysEx to replace the current one.
• Changed the format of exported patches in the PC/Mac editor to allow the files to be directly dumped as SysEx. Patches saved with previous versions can still open without issues, but patches saved with this version cannot be open in previous versions.
• A few modulators can now output useful triggers into up to 4 custom trigger channels (TC1 to TC4), which can then be used a trigger sources for percussive oscillators or other modulators like envelopes. Look for parameters named TC.OUT in modulators like DAHDSR Envelope or Compare.
• Added a Compare modulator that determines whether a signal goes above or below a threshold.
• Added a Logic modulator group with five modulators providing different tools to work with triggers and gates (Gate combinator, Trigger combinator, Gate to trigger, Gate delay, Trigger delay).
• Added a Slow LFO variant of the Advanced LFO modulator (in the Change group) with periods between a second and a full day.
• Added a Change polarity modulator that turns unipolar (0% to 100%) values into bipolar ones (-100% to 100%), and vice-versa.
• Added a new modulator group, Physics, that provides physical models and simulations:
  • Bouncing ball simulates the movement of a ball thrown in the air and bouncing over a hard surface.
  • Ball impact simulates up to 8 independent bouncing balls thrown in the air and bouncing over a hard surface, and outputs their velocity where there are impacts, making it more suitable for modulating the expression of percussive oscillators.
  • Spring-damper system simulates a mass attached to the input with a damped spring, making a "springy" or "bouncy" version of the input.
• Added a new modulator group, Algorithmic, that provides various algorithm-based utilities to generate sequences:
  • The Euclidean sequencer generates a large variety of rhythms whose beats are as evenly spaced as possible.
  • The Hexadecimal sequencer generates a sequence based on the binary representation of a 32-bit hexadecimal number, with each of its 8 digits controllable individually.
• Added a new modulator group, Chaos, that provides various ways to generate chaotic sequences:
  • Logistic map, Tent map, and Circle map, each with controllable parameters.
  • Discrete chaotic map, which allows continuous switching between Duffing, Gingerbreadman, Bernoulli and Hénon maps.
  • Cellular automaton, with 32 cells, selectable rule and initial state, as well as trigger outputs for 3 different cells, whose indices can modulate.
• A few modules now have an Oversampling parameter that allows increases in sound quality at the expense of a higher patch load.
• When selecting a CC in the CC mapping settings, your Anyma can now learn from the MIDI input. You can disable this feature by unchecking the MIDI learn setting at the bottom of the list, or by pressing the Animate matrix button while editing a CC. The same feature exists for tunings, where you can select specific scale degrees or notes using any MIDI input.
• Added four CC Out destinations in the Matrix module, that can be used so that every patch can output different MIDI CCs to an external device. The MIDI CCs, channel, and ports can be setup in the CC mapping settings.
• The volume knob now behaves more linearly.
• MIDI routing now offers a Safe merge option with a different set of tradeoffs.
• Added a Velocity curve setting to change the global velocity response of the synth.
• The Reset settings submenu can now reset the entire bank and settings of the Anyma.
• Improved the display and selection of mapping destination parameters on the Anyma.
• In the module editor screen, modules can now be copied, pasted and swapped.
• In the PC/Mac editor, modules can now be copied and pasted.
• Enabled drag-and-drop for empty modules in the PC/Mac editor.
• Now supports external MIDI clock sync, and provides an internal clock if no external clock is detected.
• Now supports external MIDI hold pedal (by default, CC 64).
• Added a Remote display option in the editor (View menu) allowing to mirror the display of the Anyma, allowing you to record clean videos of the display by capturing the editor window on your computer screen.
• Added a secret Test tone mode (hold any matrix button then press Shift) whose workings are left as an exercise to the reader.
• Fixed the Sylphinet preset of the Windsyo oscillator.
• Fixed the Artin VOSM oscillator.
• The Mac editor is now a Universal application, with native Apple Silicon support.
• Fixed some bugs leading to the Anyma Phi being frozen.
• Fixed some bugs leading to the PC/Mac editor not being able to communicate with the Anyma Phi.
• Added more feedback and confirmation options in the PC/Mac editor for bank operations.
• Added an Expr. detect option to choose the condition used by the synth to switch between expression CC input and velocity envelope.
• Limited Linux support: the editor is now also available for Linux x64, however due to limited resources Aodyo Instruments is not able to extensively test it and provide support for this particular version.
• Changed the order in the oscillator list to lump all pitch-insensitive oscillators at the beginning.o record clean videos of the display by capturing the editor window on your computer screen.
• Added a secret Test tone mode (hold any matrix button then press Shift) whose workings are left as an exercise to the reader.
• Fixed the Sylphinet preset of the Windsyo oscillator.
• Fixed the Artin VOSM oscillator.
• The Mac editor is now a Universal application, with native Apple Silicon support.
• Fixed some bugs leading to the Anyma Phi being frozen.
• Fixed some bugs leading to the PC/Mac editor not being able to communicate with the Anyma Phi.
• Limited Linux support: the editor is now also available for Linux x64, however due to limited resources Aodyo Instruments is not able to extensively test it and provide support for this particular version.
• Changed the order in the oscillator list to lump all pitch-insensitive oscillators at the beginning.

@josie The envelopes are unipolar, meaning that their output signal will be between 0 and 100%. If you just map the envelope output to pitch and don't change anything, the amount of the mapping will be +100%, meaning it will add the envelope to the pitch. So you would need to set the amount to -100% to remove the desired amount to the pitch.

I guess it worked with an LFO because LFOs are bipolar (their output signal varies between -100% and 100%), so even with a +100% mapping amount, the LFO would lower the pitch when in the negative part of its waveform.

I'll add a new global setting, Expr. detect, that allows you to set the condition used by the synth to switch between expression CC and velocity envelope for the synth expression signal, with these options:

• Never means the expression CC will be ignored, and the velocity envelope will always be used instead.
• Values 100ms to 20s mean that at the moment an expression CC is received, it will be used instead of the velocity envelope, but once the specified delay has passed and no further expression CC is received, it will revert to the velocity envelope until another expression CC is received.
• Prg.Chg. means the expression CC will be used from the moment it is received, but when switching to another patch the synth will revert to using the velocity envelope by default, until another expression CC is received.

The Anyma tries to be smart when detecting whether the Expression CC is present or not (because that determines whether it uses or bypasses the Velocity Envelope), but doing so it expects the expression signal to be sent almost continuously. This is fine when using a wind controller, but not in your case.
I think it could be better to allow you to force whether the Expression CC is used or not, for instance with an option in the settings.
Or at least to set the "timeout"; i.e., the time after which the synth considers there isn't an expression signal anymore if it didn't receive any.
Would something like that be OK to you?

@AndyHornBlower The Anyma patch structure is far from ideal for specifying FM sounds, but couldn't the structures you describe be done using VA oscillators and FM operator effects, with judicious mix routing and envelopes that modulate the "expression" and pitch mapping destinations for each of the osc/ops?

@mimj44 Les noms courts des paramètres de la matrice sont parfois "grisés" (et un peu illisibles) pour indiquer qu'ils n'ont aucun effet pour le moment. Ca dépendra donc du patch sur lequel tu es. Pas de panique, le prochain item sur notre liste à la sortie de la v1.0 sont les documents/vidéos tuto/démo !

Didn't know that, thanks!

If nothing happens when you click "Connect", it means the editor couldn't even open the MIDI ports. This generally means the ports are already open by some other application or driver, because Windows doesn't allow a MIDI port to be used by more than one application at a time.

I've just uploaded v1.0b276 (links in the top post of this topic), with two new effects and two new modulator groups.

Here's what's new:

• Added a Bitcrusher effect (Timbre group) with continuous bit depth and sample rate reduction.
• Added a Ping-pong delay effect (Delay group), together with a sync variant.
• Added an Algorithmic modulator group, providing algorithm-based utilities to generate sequences:
  • The Euclidean sequencer generates a large variety of rhythms whose beats are as evenly spaced as possible.
  • The Hexadecimal sequencer generates a sequence based on the binary representation of a 32-bit hexadecimal number, with each of its 8 digits controllable individually.
• Added a Physics modulator group, providing physical models and simulations:
  • Bouncing ball simulates the movement of a ball thrown in the air and bouncing over a hard surface.
  • Ball impact simulates up to 8 independent bouncing balls thrown in the air and bouncing over a hard surface, and outputs their velocity where there are impacts, making it more suitable for modulating the expression of percussive oscillators.
  • Spring-damper system simulates a mass attached to the input with a damped spring, making a "springy" or "bouncy" version of the input.

If no new bug surfaces before the end of the week, this will be the last beta version, meaning that the v1.0 release is imminent.
After the release, we will focus mostly on improving the documentation, tutorials and feature demos for a time.

What is the problem with the editor on Windows 8.1? Does the editor stay blank?

Thanks a lot for the great feedback!

I've just uploaded v1.0b270 (links in the top post of this topic), with very minor bugfixes and a new Chaos modulator group.

Here's what's new:

• Added a Change polarity modulator that turns unipolar (0% to 100%) values into bipolar ones (-100% to 100%), and vice-versa.
• Added a new modulator group, Chaos, that provides various ways to generate chaotic sequences:
  • Logistic map, Tent map, and Circle map, each with controllable parameters.
  • Discrete chaotic map, which allows continuous switching between Duffing, Gingerbreadman, Bernoulli and Hénon maps.
  • Cellular automaton, with 32 cells, selectable rule and initial state, as well as trigger outputs for 3 different cells, whose indices can modulate.

Thanks @JLS for the inspiration and advice.

@Clint Thanks for all the very detailed analyses you made. As soon as I get a sizeable chunk of free time, I intend to carefully read them all from the beginning and contribute tests that can help us understand what's going on and get to the bottom of this.

@steviek Nobody else of Aodyo can comment on this particular issue, unfortunately, but we're reading everything posted. It's just that I can only do one thing at a time, and such a question deserves proper treatment and will take time to answer. Sorry that we've not been very good for giving a sense of where we are and where we are heading to. I hope we'll be able to communicate a bit more often from this point on.

I've just uploaded v1.0b266 (links on the top post of this topic), with mainly bug fixes and a new "Dirty formant filter".

Here's what's new:

• fixed the timing of patch imports for the Windows/Linux versions of the editor; patches should now import correctly on these platforms
• added Save and Import commands on patch list context menus in the editor
• fixed the pseudo-random number generator on the Anyma Phi
• mappings now have a +100% amount by default
• lumped all pitch-insensitive oscillators to the beginning of the oscillator list in order to avoid losing the pitch source and mapping setup in paraphonic patches
• added a new "Dirty formant filter" that uses the same old-school FoF vowel filtering technique as the Artin VFOF oscillator, but applied to any input sound

Thanks @markusschloesser, @bibenu, @JLS, @DeepSea, and @AndyHornBlower for reporting.

I've taken some time today to study your list. I'm not sure that all of these would be possible or practical, because in the Anyma Phi modulators can only have a single continuous output.

But I can already see how these could work:

• logistic equation (select the parameter between, say, 2 and 4, trigger to sample&hold a base state, another trigger to advance to the next step)
• elementary cellular automaton (say with 8 cells; select the rule, select the base state, trigger to advance to the next step; however I'm not sure which outputs would be useful here)
• comb filter
• vowel filter (looks like just hijacking the Artin VFOF oscillator could do a pretty useful formant filter with vowel and age parameters)

For the rest, I'm not so sure, so if you have any supplementary info about what these modules would look like (or existing modules or software of similar degree of complexity for inspiration), I'd be grateful.

After testing your patches with the Windows editor, I can see a problem similar to yours: the patch is loaded, its name changed, and the modules seem correct, but I can hear the actual played patch didn't change.
However, after hitting "save" (on the Anyma or on the editor), I can hear the loaded patch playing.
This seems to be a Windows-specific behavior: on Mac the patches load just fine.

@AndyHornBlower Could you confirm that a "save" after import works for you too?

If you look at the stream of SysEx messages, you can see that even when there's nothing going on, the editor and the Anyma exchange messages constantly. As the editor didn't freeze or disconnect from the Anyma, I'm more leaning towards an issue with the Anyma reading the patches, or with Windows (or any high-level intermediate in the USB connection) and its handling of larger SysEx messages, rather than a low-level or electrical cable issue.
I've already heavily reduced the maximum size and throughput of SysEx messages in the beta, because many devices and OS are notoriously bad at handling them (Windows mostly, but CoreMIDI on macOS has its own limits too).