Synth-DIY Yahoo! Groups Archives

Thread

That 24db/octave thing...

2001-06-04 by John Papiewski

Hi all,

There's that contributed article on Egres I posted a long time ago,
about putting 2 VCFQ's in series.
I have only one vcfq in my system so I tried it by combining my system
with a couple others, I never actually got any interesting results. I
think I was doing it right.
Has anybody else tried this? If it turns out this thing is a crock then
I'll remove the article.

Thanks

JP

Re: That 24db/octave thing...

2001-06-04 by Sebastian Kuehnl

----- Original Message -----
From: "John Papiewski" <johnp@...>
To: "Serge Modular" <SergeModular@yahoogroups.com>
Sent: Monday, June 04, 2001 4:09 PM
Subject: [SergeModular] That 24db/octave thing...

: Hi all,
:
: There's that contributed article on Egres I posted a long time ago,
: about putting 2 VCFQ's in series.
: I have only one vcfq in my system so I tried it by combining my system
: with a couple others, I never actually got any interesting results. I
: think I was doing it right.
: Has anybody else tried this? If it turns out this thing is a crock then
: I'll remove the article.
:
: Thanks
:
: JP

About the 24dB module suggestions, I wanted to remind that all the things
like 24dB lowpass, variable slope, variable bandwidth (VCF2: 2 coupled 6dB
lo/hi filters anyway) and formants can be done with two combined VCFQ(X). So
my personal suggestion to Bill (and anyone else who wants to start with a
small system) would be to start with two VCFQ(X) and buy the more special
ones with the next upgrade.

However, now I'm not so sure about this anymore - John, maybe the filters in
the systems you combined differed too much? They should be identical really,
for the res loop. In general, only few filters sound good in series - I know
only the Sherman, the MS20 and the related Borg (very low noise), which do
sound good (formant sounds) even resonating in open loop. The Moog 904 does
not, I think (it does in coupled mode). Generally, cleaner sounding filters
just might not do it... I have to little experience with such.

Anyway, I would say Harvey owes the list a demo :->

Sebastian Kuehnl

_________________________________________________________
Do You Yahoo!?
Get your free @... address at http://mail.yahoo.com

Re: That 24db/octave thing...36 db per.....5 pole .......etc.

2001-06-04 by sonic@sonicboomuk.fsnet.co.uk

--- In SergeModular@y..., "Sebastian Kuehnl" <skuehnl@y...> wrote:

Show quoted textHide quoted text

> ----- Original Message -----
> From: "John Papiewski" <johnp@w...>
> To: "Serge Modular" <SergeModular@y...>
> Sent: Monday, June 04, 2001 4:09 PM
> Subject: [SergeModular] That 24db/octave thing...
>
>
> : Hi all,
> :
> : There's that contributed article on Egres I posted a long time

ago,

> : about putting 2 VCFQ's in series.
> : I have only one vcfq in my system so I tried it by combining my

system

> : with a couple others, I never actually got any interesting

results. I

> : think I was doing it right.
> : Has anybody else tried this? If it turns out this thing is a

crock then

> : I'll remove the article.
> :
> : Thanks
> :
> : JP
>
>
> About the 24dB module suggestions, I wanted to remind that all the

things

> like 24dB lowpass, variable slope, variable bandwidth (VCF2: 2

coupled 6dB

> lo/hi filters anyway) and formants can be done with two combined

VCFQ(X). So

> my personal suggestion to Bill (and anyone else who wants to start

with a

> small system) would be to start with two VCFQ(X) and buy the more

special

> ones with the next upgrade.
>
> However, now I'm not so sure about this anymore - John, maybe the

filters in

> the systems you combined differed too much? They should be

identical really,

> for the res loop. In general, only few filters sound good in

series - I know

> only the Sherman, the MS20 and the related Borg (very low noise),

which do

> sound good (formant sounds) even resonating in open loop. The Moog

904 does

> not, I think (it does in coupled mode). Generally, cleaner sounding

filters

> just might not do it... I have to little experience with such.

Dunno if its too off subject , but the OSCar filter is 2 x 12 db per
oct. filters ,with seperation control , resonance & filter drive
level .(I'm sure everyone is aware how critical the drive level is to
any filters operation ).The OSCar filter is a beauty .I.M.O one of
the best .I think the Wasp may be similar , without the seperation
control , and those synths are death on 2 batteries -pity they're so
flakey....
The Fenix filters 1 & 2also work very nicely together , but are
identical units .Funnily enough , it appears the two 12 db per o
filters sound sexier than the (5 pole ?) 24 db per oct. filter 3
(Chris Whitten seemed to agree on the better feel of these ,
supposedly weaker filters).
The other usefull aspect about the Fenix is the resonance is the
most finely settable and usable designs I've ever seen . the control
over the amount/response of the resonance is very slinky & very happy
to oblige in any way it can.
EMS filters also sound rather nice in series , but the filters on
all the synthis I have are slightly differant . If the right 2 are
used , beautifull results with 36 db for the first octave , 24 db per
octave below that cut off filter effects ensue...........

On the Serge issue . I dont think that two Serge vcfq's will do
the "sound" I suspect people are looking for .I think it will have a
beautifully precise sound , and be very capable of hi-fi subtle
filtering , but I dont think it'll ever have the "meaty/fat/phunky
sound of some other machines. I'd say this of the EMS stuff
too .However , I'm not sure it's totally down to the filter
itself......
P.K.

>

>
>
> _________________________________________________________
> Do You Yahoo!?
> Get your free @... address at http://mail.yahoo.com

Re: That 24db/octave thing...

2001-06-07 by thornburgh73@yahoo.com

--- In SergeModular@y..., "Sebastian Kuehnl" <skuehnl@y...> wrote:

Show quoted textHide quoted text

> ----- Original Message -----
> From: "John Papiewski" <johnp@w...>
> To: "Serge Modular" <SergeModular@y...>
> Sent: Monday, June 04, 2001 4:09 PM
> Subject: [SergeModular] That 24db/octave thing...
>
>
> : Hi all,
> :
> : There's that contributed article on Egres I posted a long time

ago,

> : about putting 2 VCFQ's in series.
> : I have only one vcfq in my system so I tried it by combining my

system

> : with a couple others, I never actually got any interesting

results. I

> : think I was doing it right.
> : Has anybody else tried this? If it turns out this thing is a

crock then

> : I'll remove the article.
> :
> : Thanks
> :
> : JP
>
>

...

>
> Anyway, I would say Harvey owes the list a demo :->
>
> Sebastian Kuehnl
>
>

Done. I have two examples, one varying the resonance (generated
by external feedback) from none to beyond self-oscillation.
The second consists of several filter sweeps. In both there is a
square wave input from an NTO. A square wave usually causes
problems in filter sweeps because the first few harmonics are
spaced quite far apart, so it's hard to maintain a constant
amplitude when doing a sweep. What I do at the
end of the second file is fade the oscillator input to silence. The
amplitude of the self-oscillation stays remarkably consistent, except
at the very end.

I checked John's site again, everything seems correct. I would
be interested if someone else would try this patch so I can
figure out exactly what is the problem?

The basic benefit of this approach is once the filter reaches
self-oscillation you have pretty good control of the amplitude
(headroom) of this oscillation. You can increase it gradually without
too much distortion. actually the sound of this filter
configuration is quite "clean" because no audio-rate distortion
is necessary to limit the oscillation amplitude, if this makes
sense. It's like using a compressor vs. clipping distortion,
except applied to the feedback path.

one unfortunate aspect of this filter is you can't overdrive the
input. if the input is louder than the specified self-oscillation
amplitude, the resonant peak disappears :( If I want distorted
filter sounds, I use two VCFQ's in series, with some fuzz distortion
(top wave multiplier) in between, without any external feedback.
this is very useful but a quite different sound.

My opinion on this filter business is I would much rather use
two 2-poles even in a series configuration than use a 4-pole, because
you get five options for the filter (lowpass, lowpass-band, bandpass,
highpass-band, and highpass). However, I'm all for different
filter designs, the more (high-quality) modules to choose from
the better! I *really* wish Rex would develop wavetable and sampler
modules (controlled by analog clock, of course.)

No other processing was done to these sounds, so there's a bit of
noise. Anyway, hope you enjoy the files.

--Harvey

Re: That 24db/octave thing...

2001-06-08 by John Papiewski

Wow, great, thanks! Very cool.

thornburgh73@... wrote:

Show quoted textHide quoted text

>
> Done. I have two examples, one varying the resonance (generated
> by external feedback) from none to beyond self-oscillation.
> The second consists of several filter sweeps. In both there is a
> square wave input from an NTO. A square wave usually causes
> problems in filter sweeps because the first few harmonics are
> spaced quite far apart, so it's hard to maintain a constant
> amplitude when doing a sweep. What I do at the
> end of the second file is fade the oscillator input to silence. The
> amplitude of the self-oscillation stays remarkably consistent, except
> at the very end.
>
> I checked John's site again, everything seems correct. I would
> be interested if someone else would try this patch so I can
> figure out exactly what is the problem?
>
> The basic benefit of this approach is once the filter reaches
> self-oscillation you have pretty good control of the amplitude
> (headroom) of this oscillation. You can increase it gradually without
> too much distortion. actually the sound of this filter
> configuration is quite "clean" because no audio-rate distortion
> is necessary to limit the oscillation amplitude, if this makes
> sense. It's like using a compressor vs. clipping distortion,
> except applied to the feedback path.
>
> one unfortunate aspect of this filter is you can't overdrive the
> input. if the input is louder than the specified self-oscillation
> amplitude, the resonant peak disappears :( If I want distorted
> filter sounds, I use two VCFQ's in series, with some fuzz distortion
> (top wave multiplier) in between, without any external feedback.
> this is very useful but a quite different sound.
>
> My opinion on this filter business is I would much rather use
> two 2-poles even in a series configuration than use a 4-pole, because
> you get five options for the filter (lowpass, lowpass-band, bandpass,
> highpass-band, and highpass). However, I'm all for different
> filter designs, the more (high-quality) modules to choose from
> the better! I *really* wish Rex would develop wavetable and sampler
> modules (controlled by analog clock, of course.)
>
> No other processing was done to these sounds, so there's a bit of
> noise. Anyway, hope you enjoy the files.
>
> --Harvey
>
> To unsubscribe from this group, send an email to:
> SergeModular-unsubscribe@yahoogroups.com
>
> Keep on Patchin'!
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/