Messages postés par Peter Ostry

I'm afraid it's not that simple. Analysing the sensor data is the core of the Sylphyo, and Aodyo is unlikely to give up this technology so easily. The synthesiser code is perhaps less of a problem, it's not that special. But then there's the radio link to the Link, whose technology is probably licenced to Aodyo. What should be released anyway, the firmware and documentation or what?

I know a similar situation from Eigenharp. A freelance developer took over there, but the instruments needed software that didn't even belong to the manufacturer. The developer could only keep the instruments in operation alive. That was also important, the big Eigenharp Alpha cost more than 4,000 USD. I even had to make deep adjustments to my Mac system for my little Pico to keep it playing for a while. Finally gave up.

I'm not saying it's hopeless to aim for open source, but the Sylphyo is a complex system and an intellectual achievement. Even if Aodyo were to say ‘keep going on your own’, that would be generous, but not necessarily successful ...

However, I would be very happy if it were possible to continue in some form. Perhaps the developers will find a way, perhaps via another company.

@panekharris

You can only do this via an audio connection.

1. small stereo plug into the audio output of the Sylphyo and on the other side of the cable into the audio input of the Mac (but I don't know if this works) or better (and this works) into an audio interface.

2. if you have an Aodyo Link box, go from the two line outputs to line inputs of an audio interface. Or try instrument inputs. For me it sounds best when I use a DI box to go into the mic inputs of the interface. The line outputs of the Link are unbalanced, so use TS (Tip-Sleeve aka "mono") plugs.

@Clint
Well, maybe it depends on the style and/or duration of playing, environment, who knows. The small hole on my Sylphyo is open since we dicussed this. Actually I forgot to glue the ring to the main case after re-assembling the instrument. Occasionally I lift the blue ring and it is always dry, so is the edge of the housing. Now I think, I should rather let the hole open for ventilation, in case some moisture comes from the electronics compartment underneath.

From a physical point of view, it seems logical to seal the hole. At least if you assume that moist air can condense on the metal ring. With my Sylphyo, however, there is no moisture on the ring, so I leave the hole open. But I have to say that I don't play sessions as long as you do and always under the same conditions in my home studio.

Perhaps I shouldn't have presented my method as a recommendation. It's just what works here.

@zikojr
Anyone with a little mechanical skill and electronic knowledge can replace the battery. The blue ring at the top of the Sylphyo is originally attached with a little hot glue. Remove the ring (possibly with the help of hot air), open the Sylphyo and solder in a new battery. Screw the instrument back together and fix the ring with a little hot glue (or not). If you glue, do not cover the small rectangular slot. Two dots of glue on the solid edge will do.

The battery is industry standard:
Lithium-ion Cell
Type 18650 protected (!!!)
3000 mA or more

Thank you for the update!
Is there something else changed, apart from the new sounds?
Regarding the company thing, well, what can I say - Bonne chance à tous !

Lithium-ion Cell
Type 18650 protected (!!!)
3000 mA or more

Ah, you have the labeling on the plugs. Yes, then the shrink tubing can be used without removing the plugs :-)

I have a cheap Brother labeling machine (P-Touch D210, up to 12 mm). The normal tapes come off plugs and cables. I tried a flexi-tape, which has held up so far. But shrink-wrap is certainly better.

Your wiring looks good, and you are taking nice photos!

@Clint said in Controlling the channels that the Link Synth responds to:

and cover most of those labels with clear heat shrink-wrap

What do you use for that? I don't suppose you'll remove the plugs for labeling. But what shrink wrap can you wrap around the cable?

I think 90 minutes is ok for the setup, there are a lot of small parts that need to be unpacked and well positioned, for audio you need gain staging, hum-free cabling, etc.

With the BomeBox it will hardly be any faster, the box offers more technical than practical advantages. There will be fewer DIN cables, but hubs and their power supplies will be added, as well as a MIDI interface. All in all, it may be the same clutter, but the star-shaped structure is more logical and better for the head. We can switch our brains between technology and art, but a structure that is as systematic as possible, helps.

Incidentally, I never have any problems with the Aodyo Link. At first I wanted to use everything it could (perhaps) do, but that wasn't a good idea. Since it's been connected to the USB hub and only the two audio outputs are plugged in, it simply always works.

Here is an example for a part of your setup in star configuration:

Everything is connected to the BomeBox. The Push 3 is the control center and tells the BomeBox what to do with all the ports and data. It can also send data on it's own.

There is no need to route DIN MIDI through the Aodyo Link or any other synth.

The MIDI Interface for the VL70-m is not critical, a fast and simple one like MIDItech 2x2 or 4x4 would do it.

If one group of devices is close to you and another is further away, simply give the remote devices their own, smaller USB hub that you plug into the main hub on the BomeBox. Cascading is no problem. All active hubs of course, then the power supply for the newer small devices is taken care of at the same time.

The BomeBox has many more options. But compared to your current setup, there is a layer in between. You are no longer directly connected to the devices, but are dependent on your programming of the BomeBox. It may sound strange, but it feels more abstract in operation than the daisy-chain with three times as many visible cables.

You need time. Everything changes. Cabling, power supply, etc. You have to figure out how best to program the BomeBox. Switching presets, converting CCs, MIDI channels.

You can convert the Sylphyo's performance data (roll, elevate, compass) in the BomeBox for each synth in order to trigger meaningful actions there without changing anything in the Sylphyo itself.

… mmh …

1. Channel Filter
   According to the manual, Link (Sylphyo) channels affect outgoing data. An input filter is not mentioned. However, my Link (Sylphyo) is set to channel 9 and when I send MIDI directly to the Link, it responds to Notes and Program Changes on channels 1-9 and is deaf for higher channel numbers. So the input is filtered somehow, but I almost doubt that this is intentional. You may try to set the Link (Sylphyo) to channel 1 and check if higher channels go through without changing anything in the Link. But if this works, will it still work after perhaps a firmware update has arrived at some time?

2. MIDI Thru
   My first setup was mostly daisy-chain and the Link was part of it. It did not work well. Maybe I overloaded the Link with SysEx or too much data or mangled events otherwise. My Eventide devices had similar problems. MIDI Clock was a disaster. The setup was unreliable, almost unusable.

… mmh …

Honestly, I wouldn't use the Link for MIDI Thru, but rather operate it as a single device. Especially with the level of support we are currently receiving for Sylphyo and Link.

I don't know if my method can give you any ideas, but below you see my current setup, consequently in star/parallel configuration. The computer and the BomeBox each have their own worlds and can work independently of each other.

My control center is the LaunchPad and also within the BomeBox domain. This is where I decide which controller goes to which destination. It activates/deactivates presets in the BomeBox that do the routing and channel- and CC conversion.

I no longer have any devices that only have MIDI DIN ports, but I would integrate them in a similar way. MIDI mergers and splitters are not ideal and relatively expensive. I think it's better to add a small MIDI interface to get to USB/BomeBox than to loop MIDI serially through several devices and hope that they don't make nonsense.

According to technical specifications in the website and "Rear Panel" description in the manual, the outputs are "mono jacks".

This usually means type TS = Tip/Sleeve = not balanced.

@Takacsmatto said in Trumpet voice problem:

So something is not ok with the trumpet sound.

The conclusion seems right to me. It could be a software failure in the Link.

What about re-installing the firmware on Sylphyo and Link? The updates usually came with new instruments, so I suspect that the code for each instrument is included in the firmware. Maybe a fresh install will fix the trumpet.

It would be nice if someone from Aodyo could comment on this here.

It sounds to me as if some other controller is sending something to the Sylphyo Link. But why on the first note? Or, on an analog level, a compressor sending into a reverb and not kicking in fast enough, or a residual reverb or an error in an effects unit.

I'm fishing in the dark here. Can you describe your setup?

Just an idea:
What if you store the Sylphyo upside down? If something trickles in somewhere, maybe it can also trickle out ...

(You could also leave something absorbent in the instrument while it is at rest. But I'm not sure if that won't make things worse).

@join said in A Modern Sylphyo Keyboard is Required - The Omega Has All The Keys:

About the Duduk sound, it wasn't meant to be released because it has a few bugs (squeaks) and it still has them, but we could put it back as it is in an upcoming release.

Yes please put it back. We can play with partially broken sounds but cannot play without the sound :-)

@join said in Loom: overlays?:

@Peter-Ostry When you say "the musically relevant features should be refined precisely for this purpose", anything specific comes to mind?

Since I do not know the Loom, I can only say what comes to my mind when I see this wooden ribbon:

• Relative and absolute pitch bend mode (Roli Seaboard strips and Y-axis are only relative).

• Behavior at direction-changes (sometimes I hear annoying soundglitches in ContinuuMini demonstrations).

• Configuration of the Y-axis for Aftertouch, Pitch bend and CC (I can imagine that this is already implemented).

• Gliding speed dynamics for bowed instruments (missed in the Touché).

Can’t say much more, would need to play the Loom.

In general, I meant that the inventive and programming capacity should largely go into sophisticated musical playability. The gimmicks with fader groups, strumming etc. seem out of place to me, any mobile phone can do that better.

@join said in flute like Vibrato including pitch:

@Peter-Ostry "Setting several parameters after each sound change is unacceptable. That's why I hardly ever play the internal sounds, but mostly software instruments."
Do you mean that the internal sounds don't remember slider positions?

No, I meant that more generally. If someone wants or needs different Sylphyo settings for different internal sounds, they won't be happy without preset management. Nobody wants to go to the menu hierarchy after a sound change and copy the respective settings from a piece of paper.

At least that's the case for me, I only use a few internal sounds that I like with my basic settings. I do everything else with external synths where I can save presets.

But I don't know if I'm a typical user. For me, the Sylphyo is more of a specialised controller than an instrument on its own.

@dainiak said in Loom: overlays?:

How about <0.7mm rigid plastic film?

You can't play well on plastic film and as far as I understand the Loom is mainly intended as an instrument. I see no reason for overlays. Everything you need to play can be marked above and below the playing surface and how to play certain synths has to be learnt with those synths anyway. If you only play one or two synths, you can add your own markings.

The wooden surface is a good choice for an instrument. The fingers have a pleasant friction with dry and moist skin and if you want it smoother for certain actions, use the top of your fingernails. I use the ExpressiveE Touché this way and it works pretty well.

—

I see the Loom as a cheaper version of the ContinuuMini, without the synth, which is ok. Unfortunately the Z direction is missing. The software control functions are nice and a logical idea, but in my opinion it's counterproductive to focus on them. I realise that the Loom should match the Aodyo Omega, but an MPE instrument controller without a Z-axis, but with prepared virtual faders on a narrow wooden board would be too much diversification for me. I think the Loom should be positioned as a particularly playable and affordable instrument and the musically relevant features should be refined precisely for this purpose. Otherwise it won't find a place between the Linnstrument, Roli, Sensel Morph, Osmose and all the other competitors all of which serve a niche market. The only obvious advantage would be the price.

I'm not familiar with the Eagan Matrix, but the Sylphyo is a strictly monophonic controller which sends only note on/off per note. You may send aftertouch or CCs with breath, but honestly, MPE makes not much sense here. I know MPE from Roli Seaboard and Eigenharp. The Sylphyo does not send such information per note.

If you mean telling the Sylphyo's motion sensors to send on different channels, you can use MIDI software. For example Bome MIDI Translator Pro, possibly with a BomeBox for pure hardware operation on stage. For more complex controls including curves, dynamic mapping, MPE conversion etc. you would probably end up with the Max software from Cycling '74. As far as I know, the Haken editor is also programmed in Max.

—

If you mean something else, perhaps a comment from the website of the Haken Coninuumini will help:

"Can you use another MPE controller to play the ContinuuMini's internal sound engine?"

"All the MPE controllers I know of use the "MPE Strike" (which is the exact same thing as "Midi Key Velocity") to represent the whole of the attack trajectory of a note. Most EaganMatrix presets are designed assuming much more information about the attack trajectory (not just one 7 bit number), so while the presets will play, they won't sound the same as they would with a ContinuuMini.

In Edmund Eagan’s ContinuuMini introduction video, he uses the xKey in one part of the video -- the xKey is not itself an MPE controller, but MPE specifies that channel 1 can have "traditional Midi" polyphonic notes on it. In the example, the ContinuuMini is set up to expect MPE in, so this works since polyphonic-Midi-on-channel-1 is subsumed in the MPE standard."

But you said, that the pitch reaction is too slow, didn’t you? However, what you are complaining about now is a different problem, actually a bunch of technical issues.

If you want to dive deeper, you may want to look at the threads in this forum linked below. They deal with latency of breath events but for most statements and findings you can exchange "Breath" with "Pitchbend". They are comparable because both are events triggered by human action and sent down their electronical path through various processors, beginning with the internal Sylphyo settings.

Latency Testing (Nov 2021)
How Breath Rate affects Latency (Nov 2021)
Direct vs Link Latency (Jan 2022)
Latency of On-board Sounds (May 2022)

This can help you to find out what you can expect from the Sylphyo, which improvements can be made through settings and sound setup and also what you have no control over because it is system-related.

One thread ends with a comment by Rudy Verpaele (imoxplus, Respiro). He talks about note change, but we learn that reaction time is not just milliseconds. Technical and musical aspects play a fundamental role.

@ahiotis said in Internal Sound different via Headphone and LinkReceiver:

Should I also try combining left and right and send that to the Amp? I think there's like a Y Combinator.

Absolutely not. The outputs may be protected, but you should never combine two outputs without summing them properly. Splitting one line output into two is ok, but simply putting them together is a no-go.

Ah, now I understand you, the sound example was good.

As for the Sylphyo developer, that will probably take some time. We've been waiting here for a year or more and it looks like nothing will happen until the Omega synth is on sale.

But about the Respiro. If the pitch response there is too slow for you, can't you help yourself with the curves so that the pitch responds faster?

Something like that:

Or with a platform in the middle:

The disadvantage of such curves is that they react differently in the outer areas than at the intersection. If this suits you, it could bring an improvement.

Otherwise, talk to imoxplus, maybe he has another idea for a faster response.

Yes, I'm talking about the same wishes. Except that I don't think I can turn the Sylphyo into a perfect flute. Your vibrato/slur example is spot on, there are certain things a device like this can't do. But the Sylphyo is the only wind controller that is able to follow your detailed playing to some extent. Compared to the Sylphyo, a Yamaha WX or Akai EVI sound almost like keyboards.

The Sylphyo is a very good controller and your playing technique paired with sensitive use of the motion sensors for tone colour, texture, volume etc. lets you get sounds that no one else can do.

That's what I mean by utilising the technology. Recognising what the instrument is good at and taking that to new levels in a playful way. Not a replacement, but an enrichment.

For the absolutely authentic flute sounds, there are real flutes and microphones :-)

@wx-is-dead said in flute like Vibrato including pitch:

flutists have their "treined" expression in the airstream.

But this is not a flute. It's a different instrument that should be learnt and treated as such. If you force it towards perfect simulation, you take away its strengths and end up with neither the perfect flute nor the perfect synth.

@wx-is-dead said in flute like Vibrato including pitch:

And the airstream vibrato on the preset sounds is really not expressive (to flutists) because the pitch componente is missing.

I agree with you on that. These are just presets. We would like to make and save our own, including the desired vibrato setting. But the Sylphyo doesn't give us this option.

Setting several parameters after each sound change is unacceptable. That's why I hardly ever play the internal sounds, but mostly software instruments. Would prefer hardware, but I only have two small synths and they don’t perform with the Sylphyo as good as I hoped.

The Sylphyo’s built-in "Shake Vibrato" does not do what you want? For me it’s fine and one of the main advantages of this instrument. You may consult the Sylphyo manual.

In the settings you have three parameters:
Range = amount of pitch bend
Sensitivity = how sensitive to your movement
Threshold = at which shaking force the vibrato starts to work

—

@wx-is-dead said in flute like Vibrato including pitch:

On my sylphio the sounds are all double...

They show all available memory locations and because half of them are empty, they have duplicated the existing presets there.

They are on the list:
Exhibitors

For Mac the download was just an update application, I guess it’s the same for Windows.

Here’s the procedure on Mac:
Remove USB from the Link box (if you have one), connect the Sylphyo to USB and turn it off.
Start the application, it tells you to connect USB.
Turn the Sylphyo on.
Update (you have to wait a while).
The app tells you to connect the Link.
Disconnect Sylphyo, connect the Link box.
Update (you have to wait a while).
Close the app, turn off Sylphyo and Link.
Re-arrange your cables as your setup needs them.

I don't know about the readme.txt

I don't know about an editor either, maybe Aodyo copied the text from the AnymaPhi update for which there is an editor.

Ah, Max4Live, the ugly little brother of Max :-)
So you got a quick solution, very good!

@Clint
I guess the drum software will insist on note velocity and won't be happy with CC7 or aftertouch. That would be easier, because you could send it along with the breath data from the Sylphyo.

Converting velocity from breath data is not really trivial, you need a certain number of CC2s to calculate the "probably best value" from the surroundings of a recognised note. In Max this can be programmed for tolerable latency and made into a standalone application. I don't offer this because I only work on Mac and have no idea of the conditions under which Max software runs on another computer than mine.

MIDI is not a dense, stable data stream, the calculation is uncertain and takes time, so audio-to-MIDI comes to mind. What if you play a clean synth in parallel with the Sylphyo and use its sound to feed audio-to-MIDI software or hardware? MIDI guitars for example can play drums pretty well and the convertion unit (software or hardware) hears only Audio.

I don't know about Windows software. For hardware, a Sonuus G2M converter may be worth a try. These boxes are made for guitar but take also other sounds. In the video on the linked page, the man tries Theremin and voice (starting around minute 4:00).

No. The Link only gets this information from the Sylphyo. As far as I understand Aodyo's comments (we have discussed such things here before), the communication from the Sylphyo to the Link is not MIDI, so we cannot simulate it by sending MIDI data.

By the way, I don't know of any device where you can remotely set the MIDI channel on the fly by sending MIDI commands.

If you want to change the transmit channel or send on multiple channels, you have to do it the traditional way, after the Link, with hardware or software. The receiving channel is only changeable with the Sylphyo.

@wwoodard8
Thank you for the information.

A balalaika sound for a wind controller, a copper strip trick without instructions and breath presets that no one needs, but not a single bug dealt with. Apparently none of the existing beta functions were dealt with either. Not to mention addressing reasonable feature requests. Looks like we were just discussing here for our own amusement.

By the way, none of these statements on the update page are true:
"The update package contains the update file to transfer to your Sylphyo, the current version of the manual, a readme.txt file with instructions, as well as the current version of the PC/Mac editor for your operating system."
Actually the download is one applikation that updates the Link Box and the Sylphyo. Nothing else and no further information.

As Clint wrote.
There is no Sylphyo or Link based solution.

On a software route you can do whatever you want, on the hardware route you are tied to relatively expensive small devices. The usual boxes are Midi Solutions or the BomeBox.

I myself went with the BomeBox and meanwhile have all my controllers and two synths controlled by this system. Bome software and hardware is not cheap, but very good.

Here in the forum we have already discussed blocking the transmission of program changes and an option to prevent this has been proposed. At least I understood that the programmer recognised the problem. But with Aodyo's current devotion to Sylphyo, I wouldn't wait for a solution from this side.

Hello Frank,
thank you for your comments, but the discussion no longer has anything to do with this forum or this thread and I don't want to continue it here. I am writing this because it would be impolite not to reply after your effort.

@frank
Then you are a lucky man. But there are also sound engineers who route suspect signals through DI boxes on principle.

I have now put the Sylphyo Link and also two synths behind a multiport DI box. The Link box is safe and as an additional benefit there is less noise overall.

You can't convince me with experience alone, since I'm almost 71 we could be about the same age :-)

We won't hear from Aodyo until the Omega synth is on the market or the project is cancelled. Even that is not certain. As we know, there is only one software developer and we will be lucky if he still works for Aodyo after the Phi/Omega adventure.

I predict a similar course for the Sylphyo as for the Eigenharp controller. One of the best products of its time, too few sales, developer lost and goodbye. The Eigenharp fared worse because it doesn't work without computer software.

With the Sylphyo we have the advantage that it is not tied to external software. We can use it as long as the hardware lasts. From this point of view, it is fortunate that there is no editor software necessary for it to function. So be careful with your wish for an editor, it might come true ...

Maybe Aodyo is angry with me now, but I don't even get an answer to a simple question in the forum and by email. After all, I'm just an annoying customer who has already bought. I can live with it, fortunately the product works and we also know how to replace the battery.

@Clint
Yes, of course. Input or output is not the question. It's simply 48 volts of DC hitting some components. If they are not adequately protected, it goes "ssshh..." and the gig is over.

@frank
What is the exact characteristic for an electrical exit, possibly a "no entry" sign? Then my power is not only a phantom, it also can not read. That was certainly the reason why it grilled my old ribbon mic and later burned the output stage of a preamp. I'm for more education: Phantom powers must be able to read!

What happens if the Sylphyo Link line outputs are connected to an interface input with phantom power - is this dangerous for the Link?

@Gromit
Yes, it is spring-loaded and bi-directional. I don't remember exactly, but I think there is a small potentiometer underneath, but quite precise. And yes it is mostly used for pitchbend.

On the Sylphyo I would use it for Pitchbend and ModWheel because the rear touch surface is almost unusable for me, practically out of reach, and I never know where my finger is until I hear it.

@NS-D
I suggest you start with the fingering you are most familiar with, in this case saxophone. If you need it differently, e.g. for fast or comfortable #/b changes, you can try other fingerings or settings.

Except for simple flutes, no fingering matches the original instrument exactly anyway, because the Sylphyo has no additional flaps.

@mathiross
I mean this thing on the WX controllers:

Maybe it is just called Rocker because it is not a switch.

I miss it on the Sylphyo, but I miss mechanical elements anyway. I'm not fond of touch plates. I love the Sylphyo for it's breath sensibility and the transformation to MIDI. Everything else is improvable.

@ark
Well, I can't agree unconditionally. I very rarely use the internal sounds.

But: The blow sensor and gyroscope system from the Sylphyo, key buttons from the Eigenharp, rocker switch, mouthpiece and octave buttons from the WX, and new software with an editor – that would be a wonderful instrument :-)

Don't you miss the rocker switch on the Sylphyo?
Don't you find the WX octave keys better?

@mathiross said in Mouthpiece Mod to reduce Moisture:

I'm playing the sylphyo in the new progressive rock band ODDLEAF. Take a look here: https://oddleaf.net

Wow ... I love this music. Kinda jazzy progrock, if this designation is allowed. I like the overall vibe, the people, the voicing, harmonies, tempo changes, careful selection and usage of instruments, everything. Thank you for showing me this!

I think that works. Painted the inside of the box, sealed it so to speak, and let it dry for a long time. Then put in the sylphyo and desiccant and after three days the silica gel balls are still a little orange and I think that will get better.

I'm not good with woodwork but a second version I would do differently. The volume of the case is about six times the Sylphyo. Difficult to keep that dry. But it is relatively pretty, my Sylphyo shrine :-)

Moving PCBs are a design flaw. Is there no way to keep them in place? With a screw in the right place or something.

I don't have an Anima Phi, but in the photo I see three large copper rings on the PCBs, like the ones usually used for retaining screws. Are they not in the right place, or not in use at all?

Not sure if that scicominc material is the right stuff, but you will find out. Hydrophobic does not like water, Hydrophilic filters water, we don't want either. If it really works, this may be because of changed circumstances for condensation inside the mouthpiece. But then almost any material with a porous surface should work. And what about the airflow?

Regarding the tubes and if the material works – they don’t offer the dimension we need on their website.

—

Have you tried to roll one layer of a thin kitchen sponge into the mouthpiece? I don't know how it is called in your country, we call it 'Wettex' which is a company label.

This is just my first try. I think the "tube" should be a little longer and maybe we have to make sure that we don't blow the whole thing into the Sylphyo’s main tube. An elastic ring that holds it in place? The material definitely absorbs liquid, that's its job, and it costs nearly nothing.

Are you sure that the condensate is disciplined enough to follow gravity and flow into the area where you can comfortably collect it, despite the strong airflow?

Maybe you need an "all-around pocket", an inset second wall, an inner tube with minimal upper gap into which to blow the condensate.

But any pocket in there will give turbulence. If you look into the main tube of the Sylphyo from above, you will see (feel) a very smooth, flat transition where the mouthpiece connects. I imagine they did that for a reason, because strong turbulence could support sensor flutter.

That's why I think your design should be very streamlined. If some droplets enthusiastically take advantage of that to jump over the pocket along with the air, that's fine too.

As a test, perhaps a ring would be suitable, which you insert into the mouthpiece at various points to see if anything at all collects there that is worth to drain off.