net nog iets gevonden op het net,
> Hi,
> I just got the Waldorf Pulse, which has DCOs (digitally controlled
oscillators). To my knowledge DCOs are analog oscillators (analog
components produce the signal), the only difference between DCOs and
VCOs (voltage controlled oscillators) being that the pitch of the DCOs
is controlled by a digital clock rather than a voltage.
This is too simplistic a comparison. There are many different
implementations of "DCO" out there, some of which you'd be willing to
call a DCO and others you wouldn't. I'll give you one example: in the
Roland JX8P and JX10 synths, there is a digital clock (square wave)
that is run through an analog "waveshaper" to produce the saw and
pulse waves. Is this then a DCO or a digital oscillator or what?
> Bearing this in mind, shouldn't the two be indistinguishable from
each other in terms of tone color?
Individual notes may be indistinguishable, yes. But we don't often
play just one note on a synthesizer.
> Why VCOs are said to be more warm sounding then? Is it because of
the instability in pitch, which causes the pitch to drift and when
more than one oscillator is in use, creates warm detuned "drifting"
effects? But can you really hear the difference?
Hell yes! I refer you to the Oberheim OBX which we used to
intentionally leave slightly out of tune so that chords would sound
even fatter than usual.
> How about emulating the drifting with a sample and hold LFO or a
slew limited version of a sample&hold LFO, whose frequency is
controlled by another LFO, how close can you get?
It's been done. See the Roland JD-800 synth. First, it has an
"analog feel" parameter which is separately settable per each of its
four oscillators. All this parameter does is randomize the INITIAL
pitch of each new note. Small values of this parameter give very
subtle results. With large values, you can hear the random S/H-like
behavior of this control with each new keystrike. Secondly, the JD
has a special LFO waveform that I haven't really seen on other synths,
along the lines that you describe above. It sounds like a normal sine
wave LFO whose pitch is being continuously randomly varied with plenty
of slew in there so you don't hear sudden changes in the waveform. If
you route a little bit of this LFO to oscillator pitch, again
independently for each of the four oscillators, you get a simulation
of the pitch instability of analog synthesizers.
Now, having said all of this, do the Roland JD-800's oscillators sound
analog? Hell no. It's a fantastic synth and I love mine to death,
but it'll never ever pass as a classic Moog or Oberheim.
> And isn't it some times better to have 100 % stabile oscillators
which do stay in tune when you want them to?
Of course!
Another aspect of analog oscillators that you haven't mentioned is
that when you strike a key, the waveform begins playing at a random
location in its cycle. On most digital implementations (up until
recently, that is) the waveform would always begin at its zero point
and continue from there. It's not too hard to randomize the phase of
the waveform before playing it in a virtual analog synth
implementation, but most manufacturers didn't bother.
I don't know if any of this will help, or just add to your confusion,
but here it is. Good luck.
Regards,
Michael Erwin
Synthank
www.synthank.com