Doepfer

--- Dieter Doepfer <hardware@doepfer.de> wrote:
> From: "Dieter Doepfer" <hardware@doepfer.de>
> To: "Bakis Sirros" <synth_freak_2000@yahoo.com>
> Subject: AW: [Doepfer_a100] Re: more news about the
> a154 and other modules from doepfer.
> Date: Fri, 16 Jan 2004 16:17:12 +0100
> 
> Please forward ...
> 
> > -----Urspr�ngliche Nachricht-----
> > Von: Tommy DOG [mailto:mrsvomit@tommydog.com]
> > Gesendet: Donnerstag, 15. Januar 2004 20:57
> > An: Doepfer_a100@yahoogroups.com
> > Betreff: [Doepfer_a100] Re: more news about the
> a154 and other modules
> > from doepfer.
> >
> >
> > --- In Doepfer_a100@yahoogroups.com, Bakis Sirros
> > <synth_freak_2000@y...> wrote:
> > > dieter says:
> > >
> > <Snip>
> > >" to emulate the
> > > behaviour of the original Buchla module it is
> true
> > > that 2 separate shift registers are the right
> solution.
> >
> > This Isn't clear whether this was an oversight in
> design or a
> > mistake that should be corrected.
> 
> It depends upon the point of view. You could call it
> another feature or a
> mistake whatever you want. We knew that the Buchla
> uses two shift registers
> but we were of the opinion that it is not necessary.
> So we did not oversee
> that but believed that the version with one shift
> register is an equivalent
> solution. It was no problem to introduce a second
> shift register and so we
> did it (even though we think that it is not
> necessary, simply two different
> types of random sequences appear but one could not
> say that one is better
> than the other, that's the basic of random - or
> better pseudo-random as the
> Buchla approach - like any approch with feedback
> shift registers - is only
> pseudo-random in contrast to "real" random like the
> A-118 or other
> semiconductor noise based designs).
> 
> To have a 100% copy of the Buchla module 2 shift
> registers are required. But
> to have only one shift register that is triggered
> from both clock sources
> creates also random voltages but a little bit
> different as each register is
> triggered by its own clock. From our point of view
> the version with one
> register is "more random" as both clock sources
> cause an advance to the next
> state.
> 
> > Is this a adjustment that users can make?
> 
> The shift registers are made with a PIC
> (microcontroller). So it depends
> upon the PIC in the module which version is
> included. We could try to start
> a competition (like wine tasting): test two A-149-1
> and find out which is
> the version with one and which the version with 2
> shift registers (of course
> with the corrected rising edge trigger) :-)
> 
> Best wishes
> Dieter Doepfer
> 
> 


=====
Bakis Siros
Parallel Worlds / Polariton
Athens-Greece
[Doepfer_a100] group owner
http://www.rubber.gr
http://www.ward12.com

__________________________________
Do you Yahoo!?
Yahoo! Hotjobs: Enter the "Signing Bonus" Sweepstakes
http://hotjobs.sweepstakes.yahoo.com/signingbonus