Ability to assign CV inputs to destination of choice

[joeman]

The CV outputs are great with loads of flexibility. This is good.

The CV inputs really really need to be assignable.

Is this at all possible?? If so, it should most definitely be implemented!

[PES]

+1, but I'm afraid this is not possible.

Even if I recall early reports stating that the whole matrix thing was quite a technical challenge to realise, I'm 99.9% sure it functions exactly like on any other analog synth with digital tech: all modulators are generated in software, meaning no actual CV's are being mixed in the Matrix. The only difference compared to for example DSI synths is the interface (buttons vs menus) and the number of "slots".

This means there's a DA converter before each input on the analog side; the CV outputs are just duplicates of those voltages, while the inputs mix an external CV with the DA converted signal before it hits the analog input it feeds. This is also why the inputs and outputs are the same.

So there is no AD conversion after the CV inputs, and thus no way to get the external CV back into the digital part of the synth for re-routing.

[DrJustice]

@PES: I can confirm your observations. The CV IO board plugs straight into the analogue voice board with a flex cable, and around the connector there appears to be buffers, then straight off into the voice circuitry. There is one CV input which I assume is sampled though, "LFO 1 AMT", since the LFO code needs that parameter.

While the flexibility of freely assignable CV inputs would be cool, it would also have increased cost and complexity. As it is, I think they struck a reasonable balance there.

There are two upshots of this. One is that the CV IO is pure analogue, so it ticks that box. Then there's the possibility of the CV inputs being full bandwidth - who is going to test feeding some audio rate modulation into the MxB?   

[DrJustice]

Aaand.... Tested! It takes audio rate modulation in 

[PES]

Great!

[joeman]

@DrJustice and Pes thanks for the info. Yes that all makes sense.

[PES]

Since the headphone output has its own volume control, the synth's output can be used to modulate a CV parameter (as per DrJustice's tip) using the volume control as an attenuator. This means audio modulation of any filter, or even a one osc patch modulating its own pitch.

[PES]

I just did a test using the pedal connectors as CV inputs.

Don't know about safety since the pedal voltages are +3.3 volts and the CV signals sent were +/-5v - if you try this at home your MB will most likely blow up!

Anyway, works like a charm. LFOs are cut halfways since the inputs won't accept negative voltages. Same goes for feeding audio. There is some smoothing going on too, so it's not full resolution (audio rate) like the CV inputs, but it's plenty fast.

So for envelopes or square LFOs or controllers putting out positive CV it's all good; we've got two assignable CV inputs.


edit: I quite like using a triangle LFO into this, so the cutting has its merits too.

[MarkSteiner]

I just did a test using the pedal connectors as CV inputs.

Don't know about safety since the pedal voltages are +3.3 volts and the CV signals sent were +/-5v - if you try this at home your MB will most likely blow up!

Anyway, works like a charm. LFOs are cut halfways since the inputs won't accept negative voltages. Same goes for feeding audio. There is some smoothing going on too, so it's not full resolution (audio rate) like the CV inputs, but it's plenty fast.

So for envelopes or square LFOs or controllers putting out positive CV it's all good; we've got two assignable CV inputs.


edit: I quite like using a triangle LFO into this, so the cutting has its merits too.

Wow, thank you!  This really expands things well!

[fede bts]

Would it be possibile, in your opinion (I am talking to the ones who have better technology competence than me), to improve the expression pedal inputs in order to attenuate and convert the incoming CV signals? I am thinking of a software upgrade obviously, not a hardware hack. Let's say I am feeding an LFO into the EXP Pedal whose signal is +/- 5v: would it be possible to convert it from a bipolar signal to "monopolar" one, and to attenuate its level in order to fit the 3,3v limitations imposed by the Matrixbrute? I read somewhere that there is some kind of microprocessor Between the EXP Pedal Input and the PCB, but I have literally no clue of what it can do. SO forgive if this is a relatively silly question. Just wondering if Arturia can actually do something in order to fulfill the "CV inputs to destination of choice" feature, which seems like many of us have been asking for in the past few months.