Das liegt in der speziellen Architektur des Blofeld begründet. Wenn die VCA Envelope auf 0 geht, dann "vergisst" der Blofeld alles, was mit dieser Stimme/diesem Sound zu tun hat.
Hier ist es sehr schön erklärt:
https://www.gearslutz.com/board/showpost.php?p=9322447&postcount=4
Hier ist noch meine Antwort von anno Ofenrohr im Waldorf User-Forum:
> In polyphonic mode yes. In monophonic mode I would call it a bug. Free
> running LFO means free running right? I expect in monophonic mode that
> 1 LFO is used all the time, not depending on voice triggering, I mean
> *the same* LFO. As it is now it looks like with every new trigger,
> even in monophonic mode, a different LFO is used?
> Well, I guess you know what I mean
I know what you mean, well, at least I am quite sure about that.
Point is, in a synth as implemented with the same philosophy as the
Blofeld (I am sure, the other DSP based Waldorf synths behave the
same here, but I haven´t checked) there is nothing like *the same*
LFO.
The Blofeld is a multitimbral synth with dynamic voice allocation, even
the DSP and RAM ressources are dynamically allocated. So technically it
is quite different to anything in the league of an old style analog
polyphonic synth.
I am not saying it would not be possible to implement it in a way it
would behave like expected, but then a differen approach would be
needed.
In the Blofeld, as with other Waldorf synths, the voice allocation is
tied to the VCA envelope. When a voice is triggered, an instance of
that voice with all the needed settings is created and the calculation
starts. When the VCA envelope is reaching 0 in the end of the release
stage, calculation stops and all the ressources (DSP and RAM) are
released. The next triggered voice could be a differen sound with
different setting. There may be some fancy caching going on in the
background, only Stenz and Wolfram may know that...
There is no way that one voice hands over LFO status information to
another, since when the VCA envelope has reached 0 in the end of the
release stage, it is gone. The new voice is not in existence and there
is nothing implemented to let the LFO run further in the background.
Well, this is not fully true, there is the sync parameter in the LFOs.
This syncs the phase of the LFOs of a part together. This leads to
the workaround -> see below.
In a synth, where you have real hardware, or maybe even in a DSP VA
synth, with a fixed architecture you may have a LFO which continues
in the background, even if no voice is sounding.
For the Blofeld a Global LFO (per part maybe) would be needed, this
would eat up DSP ressources all the time.
> > And there is a workaround available. More on that later as well.
>
> THAT I am interested in
You have to use the LFO sync, that´s obvious. In addition you need
to create a continuous time stream, that is, make sure that at least
voice is allocated for the time you want a continous LFO pattern.
Option 1:
Have a soustained note, which is silent. That could be done with
clever use of modifiers or just an ADSR setting like 0 20 0 0 or
velocity. Press and hold one note, then play along your chords
or patters.
Option 2:
Have the notes overlapping, this is just setting the release to the
right timing. If needed you could use an ADSR setting like 0 20 1 127.
You have to prevent that the part gets deallocated. This works for the
arp patter in mono mode and works fine.
> @ Stefan: yes it is common: Jupiter 4 for instance, Korg PS3100,
> Jupiter 6, Memorymoog etc. etc. Polyphonic synths, 1 or 2 LFOs. But
> they run (in free running mode) independent from voice triggering.
The examples are not really valid, since those synths are quite limited
in architecture (not in sound or coolness). They only offer global LFOs.
They are not multitimbral, I think the JP6 offers two parts.
More valid synths would be Xpander, Andromeda, SunSyn, maybe OB-8 (here
the page 2 options offered a rate modulation of the LFOs)...
