[sdiy] Roland Juno 106

Sun Feb 18 15:40:27 CET 2024

It's an interesting topic, Rutger!  As someone who's spent a bit of time 
digitally modelling analogue circuitry I can tell you that the digital model 
only becomes as complicated as necessary to capture all of the important 
(perceptible) features of the analogue sound.  You essentially only keep 
adding parameters if you feel that something is missing from the model, and 
from that point of view you know which parameter to tweak to change a 
particular aspect of the sound if it's not quite right.  When you try to 
clone the original analogue circuit directly it's a bit different...

You can get as many of the original components as you can, and substitute 
modern supposedly "equivalent" components for those that you can no longer 
get.  You build it up, test it and find that it sounds kind of similar to 
the original 80's analogue electronics but it sounds noticeably different in 
several areas.  Now comes the question, what do you tweak in this mass of Rs 
Cs and op-amps to make it sound *exactly* the same!?!?!?  There's maybe 50 
or so analogue components in there, all with subtly different affects on the 
sound, but *you* didn't design or simulate the original hardware so it might 
not be obvious which capacitor or resistor needs tweaking to brighten the 
decay of a sound, or if the gm of that particular OTA is a bit too high, 
etc.

My point is that the analogue electronics is as complicated as the original 
circuit is.  A digital model is only as complicated as you make it, and you 
should know what all those parameters do because you added them to achieve a 
certain goal.  Taking the example of the Juno 106 chorus...  I'd far rather 
be given the task of nailing that sound digitally (I've done it and it's not 
hard,) than trying to find some modern "equivalent" BBDs that are close 
enough and then endlessly tweaking the design to get a faithful copy of the 
original with modern analogue electronics.

-Richie,

-----Original Message----- 
From: Rutger Vlek via Synth-diy
Sent: Sunday, February 18, 2024 10:45 AM
To: SDIY List
Subject: Re: [sdiy] Roland Juno 106

Interesting discussion! What fascinates me is the following:

Instruments are being 'cloned', and some clones are closer to the original 
than others (as Roland's JU-06 resembles a Juno 106 but not quite). If one 
is interested in exhaustively duplicating every single aspect of the 
original (including the UX-part), why not build the exact same instrument? I 
know, that's challenging given some part may no longer be available, but 
then the engineering question is: how do I make it work with modern parts 
without losing any of the properties of the old circuit.

However, some properties of the original synth were actually not 
intentionally designed, but rather emerged as a consequence of other 
engineering decisions and properties of available parts. The original Juno 
106 that we've become to like so much is only partially the result of clever 
engineering. The original Juno also had unintended properties, some of which 
turned out to be musically favourable/characterful. These were 'lucky 
accidents'. However, some unintended properties have no musical value 
whatsoever, or are even detrimental. So how do we differentiate between 
these two categories?

It's easy to state that a clone is not resembling the original circuit, 
because in many cases it isn't. Whether that's a problem is another thing. 
Do we need all the quirks of the original, or only those quirks that have 
musical value? If someone states that a clone 'musically' falls short of 
mimicking the original, then we have to deepen our understanding of what 
actually makes a good instrument. However, the question is to what extent we 
can achieve that. And if we take musical instrument design to such a high 
degree of 'controlled engineering', don't we miss out on the chance to make 
our own 'lucky accidents' that fuel the gear lust of future generations?

Rutger

Op ma 12 feb 2024 om 01:04 schreef brianw <brianw at audiobanshee.com>:
On Feb 11, 2024, at 1:50 AM, Gordon wrote:
> On Sat, Feb 10, 2024 at 04:42:08PM -0800, brian wrote:
>> What's "the digital part"?
>
>> There's no need to 'clone' the firmware, since that digital aspect 
>> doesn't really affect the sound.
>
> Okay, so you're saying that the way the control ranges interact, the way 
> the envelopes work over particular ranges, and the way the LFO works - and 
> the way that is all mixed together to control the oscillators and 
> envelopes - you're saying that this "doesn't really affect the sound"?
>
> That's an interesting take on it.

Thanks for the details. If you forgive me my engineering focus on semantics, 
then what you're saying is that the digital envelopes and digital LFO, as 
well as the specifics of the frequency control for the DCO, is what needs to 
be recreated. i.e. It's not a 'clone' of the CPU, per se, or the entire 
firmware, but the audible aspects and characteristics of the signals 
generated by the CPU - the envelopes and frequency controls - that are 
critical.

>> I'm just curious what you're saying has yet to be reproduced with modern 
>> technology...
>
> Nobody seems to really understand what the voice CPU is doing, so they try 
> to "improve" it. This is nothing short of disastrous. If you compare the 
> JU-06 to a real Juno 106 for example, it doesn't sound anything like the 
> "real deal" because it just doesn't get the envelopes even close to 
> correct.
>
> That's leaving out all the "higher order effects" like the inherent 
> wobblyness that the LFO has, the variation in the envelope times, and so 
> on. There's a lot going on in that 2kB of code and 1.5kB of lookup tables.

This makes sense.

Thanks again,

Brian

________________________________________________________
This is the Synth-diy mailing list
Submit email to: Synth-diy at synth-diy.org
View archive at: https://synth-diy.org/pipermail/synth-diy/
Check your settings at: https://synth-diy.org/mailman/listinfo/synth-diy
Selling or trading? Use marketplace at synth-diy.org

-- 
This email has been checked for viruses by AVG antivirus software.
www.avg.com