[sdiy] wanting "good" envelopes, not "bad"

[Jim Patchell]

    Zipper noise is probably the biggest problem I can think of.  In order
to keep the zipper noise down, you with either need a high sampling rate, or
a filter.

    One way I considered doing a "software EG" was more of a compromise
between hardware and software.  Use a standard "EG" core, ie, a single pole
voltage controlled filter (well, that is what they are, aren't they?).  Two
control voltages per EG, one to control the level, the other to control the
time constant.  Uses fewer components, but all the advantages of full
hardware, but, still more parts pure software EG.

    -Jim

Kenneth Martinez wrote:

> I'm finally resuming work on my synth project...I'd like to finalize the
> circuits and design PCBs soon, so this is the time to make any
> changes...
>
> At this point I'm still planning to use software-generated envelopes, as
> they have several advantages - they're easily controlled and
> "calibrated" (no hardware trimming needed to get consistency between
> multiple envelopes), I can allow several per voice, they can easily be
> made multistage or looped, they can modulate any CV without needing
> extra hardware summing circuits, etc.  I own some synths with software
> envelopes (e.g. Matrix-1000, JP-6, MKS-50, Prophet-600, used to have an
> Xpander) and some with hardware envelopes (e.g. Pro-One, Prophet-5,
> various Moogs, used to have an OB-8).
>
> I often hear complaints about how software envelopes don't sound as good
> as hardware.  Can anyone please quantify what makes the software
> envelopes "bad" and the hardware ones "good"?
>
> - Is it the fastest attack and release time?  Certainly some synths like
>
>   the Prophet-600 have a sluggish attack; but the JP-6's minimum attack
>   and release times are fast enough to generate an audible pop, and to
>   my ears it sounds pretty close to the hardware equivalents.
>
> - Is it the shape of the attack and/or decay?  I know the Prophet-600's
>   linear decay isn't popular (it's not my favorite either); but the
> JP-6,
>   Xpander, etc have realistic-sounding exponential decay.
>
> - Is there some other characteristic I haven't mentioned?