Doepfer

Hi list,

I've been thinking some more about new modules and I would like to 
share my thoughts with you. This one is about creating loose, funky, 
groovy, sloppy, wobbling, stumbling or even psychotic and mentally 
disturbed grooves and patterns out of an incoming dead-straight 
clock. You could even call it human feel :-))
   
The idea would be to have an external clock, have an internal clock 
synchronise to its tempo, and then have the possibillity too:

- divide the clock into groups/grooves/patterns of probably 1 - 8 
steps; 
- add a certain amount of positive or negative delay (can I say this? 
:-) too each individual step of the groove;
- vary the overall gate length of all steps with vc for extra rhythmic 
pumpification

This way you could (in theory) do crazy rhythmic things to your A-
100 system in real time while being locked to a master clock. You 
would never get out of control because every move you make would 
be related to your main beat at all times. Also it would be possible to 
work with triplets etc. (swing) by adjusting the master clock before 
you send it to the module, e.g. by using an A-163 or the v2 of the A-
151 (just have a '3' in there) or you could halve or double the 
resolution of the clock or use an unusual amount of steps in your 
pattern for different effects .

The typical delay-range for the individual steps would be somewhere 
between +/- 5 and 200 ms, so in this aspect this idea would require 
more precision than the A-162, which has a (+/-?) trigger delay of 
up to 10 seconds, and which is not very precise in the 15 - 50 ms 
range, where a lot of groove is to be found. But the range should 
probably be even wider than 50 ms, so you could not just do 
grooves, but also create really different rhythmic patterns. 
(Theoretically speaking, if you feed such a pattern to a sequencer 
pattern, I guess that would be 'pattern multiplication'?? :-)))   

I guess the user interface could be something like a clock input, 
output and reset, an input for vc gate length, 2 pots for vc gate 
length, 8 pots for +/- delaying individual steps and a rotary switch 
for the number of steps in the groove.

I'm no engineer, I only think about this from the functional point of 
view, so I don't know how doable it is (e.g. the real time combining 
of +/- delay and vc gate length) but I think a module like this could 
be very usefull musically, allthough I'm not sure if I'm missing 
anything obvious. 

Greetings

Anton

Eyyy Anton,

you snaffle my suggestions ;-))

I long time ago asked for something like that. It is not that easy to 
produce this out of a existing clock, because real "groove" requires 
triggers before(!) the beat. E.g. Jomox Xbase09 provides negative 
trigger delays like that, if you work with the internal clock and th 
XBase is the MIDI-Clock master.

What I would like to have would be some masterclock with several 
slave-modules which can provide different grooves. The clocks should be 
available for different dividing stages (shuffling with 1/8th is 
different from shuffling 16th).

My example are the different MIDI quantizing modes in Logic (16A to 16F, 
where A is straight beat, 16F is close to tripletts). You get real 
catchy grooves if some hihats are more tight to the straight beat (e.g. 
Mode 16B) and others, e.g shakers or tambourine have more swing (e.g. 
16E); the BD is quantized wit 8B or 8C.


I created a shuffling DIN-Sync clock by the following arrangement:

    *-linFM-->VCO----+------------>Clock out
    |                |
    |           Divide by 6 or 3
    |                |
    *----Inverter<---*

A VCO creates a squarewave which is used as Clock.
The clock is also divided (DIN-Sync has 48/96 ppq). The output of the 
divider provides a square which in inversion is down at the first 1/16, 
then down with the second 1/16 and so on. If this slow square is used as 
CM-Control voltage for the VCO itself, the Clock will start slower and 
the first (third, fitfth...) 1/16 will be longer -> the following second 
(fourth, sixth...) trigger will come delayed.
The clock while the second (fourth, sixth...) 1/16 will faster, the 
second (fourth, sixth...) 1/16 will be shorter and the following trigger 
will be come earlier.

I tested it with my Nordmodular and with the A100 and a TR-808. A 
swinging TR808 is real cool.



> I guess the user interface could be something like a clock input, 
> output and reset, an input for vc gate length, 2 pots for vc gate 
> length, 8 pots for +/- delaying individual steps and a rotary switch 
> for the number of steps in the groove.
I think, this can be done with several A155 plus A142 (and maybe other 
modules) already? If you use the A155, you can use one row for creating 
CVs for the Clock-VCO, which work similar like the described DIN-Sync 
shuffling.

Florian

let me know how you plan to travel backwards in time to do negative 
quantization

On 2 Dec 2005, at 12:49, Florian Anwander wrote:

> 
> Eyyy Anton,
> 
> you snaffle my suggestions ;-))

Hi Florian

Quite honestly, I don't see how our approaches could be more 
different. We have both been thinking about a groove-module but I 
think that's where the similarity ends

You suggest templates, where I suggest the possibility to create my 
own grooves, and also the possibillity to change them live in real 
time. Not only the timing, but even the gate length. It doesn't get 
more different than that... 

Then you suggest different grooves at the same time, but I suggest 
one at a time. For polyrhythm that would mean buying more modules 
(as it allways does :-)

It's a subjective thing to say, but creating my own grooves live is 
more the modular feel and what I would like to use my A-100 system 
for. But the A-162 just won't do for this purpose
> 
> I long time ago asked for something like that. It is not that easy to 
> produce this out of a existing clock, because real "groove" requires 
> triggers before(!) the beat.

that's what I mean with +/- delay

> where A is straight beat, 16F is close to tripletts). You get real 
> catchy grooves if some hihats are more tight to the straight beat (e.g. 
> Mode 16B) and others, e.g shakers or tambourine have more swing (e.g. 
> 16E); the BD is quantized wit 8B or 8C.

This is the meaning of polyrhythm: each element its own groove. But 
I want to customise a groove to my own taste, right down to the 
millisecond (the A-162 e.g. just can't do this). But maybe the next 
time I feel like making wild sweeps, just to experiment so a timing-
module for me has to be versatile  

> I tested it with my Nordmodular and with the A100 and a TR-808. A 
> swinging TR808 is real cool.

Clever patching Florian, I'm gonna try it 

> I think, this can be done with several A155 plus A142 (and maybe other 
> modules) already? If you use the A155, you can use one row for creating 
> CVs for the Clock-VCO, which work similar like the described DIN-Sync 
> shuffling.

I bet you a crate of dutch beer it won't be the same :-))

Anton

Hi Anton

>>you snaffle my suggestions ;-))
> Quite honestly, I don't see how our approaches could be more 
> different. 
Sorry, Anton, it should be kidding. I never intended to embarrasse you.

I also never wanted to say that your intention of creating a groove is 
the same like mine. I only wanted to describe, what I could imagine.

>>I think, this can be done with several A155 plus A142 (and maybe other 
>>modules) already? If you use the A155, you can use one row for creating 
>>CVs for the Clock-VCO, which work similar like the described DIN-Sync 
>>shuffling.
> I bet you a crate of dutch beer  it won't be the same :-))
You are right: of course it is not the same.

The main problem is, that the Seq-outCV-controls-VCO method influences 
the tempo. This should not happen to my opinion.

> The typical delay-range for the individual steps would be somewhere 
> between +/- 5 and 200 ms, so in this aspect this idea would require 
> more precision than the A-162, which has a (+/-?) trigger delay of 
> up to 10 seconds, and which is not very precise in the 15 - 50 ms 
> range, where a lot of groove is to be found. 
It is technically very simple to add an switch and an alternative 
capacitor to reduce the delay time range. Like the range switches of the 
A140 envelopes.



Regards, Florian


PS:
 > I bet you a crate of dutch beer
Never, never, never bet for dutch beer with someone born and living in 
Bavaria. ;-)

On 2 Dec 2005, at 15:30, Florian Anwander wrote:

> Sorry, Anton, it should be kidding.

I'm not at all offended 

> The main problem is, that the Seq-outCV-controls-VCO method influences 
> the tempo. This should not happen to my opinion.

I agree, and even stronger, to me that makes this method quite 
useless, because the most important thing to me would be staying 
locked into the main beat and being able to do unorthodox things at 
the same time without getting lost. Think of the more extreme flows 
you hear in d&b/garage/grime grooves, or some of the stranger 
hiphop beats.  

The other main function of the module would be to finetune the feel 
of basslines, melodies, hihats etc. and creating all kinds of crazy 
loops. Also it can be refreshing to get out of the 4/8/16 grid and use 
a bit of 3, 5, 6 or 7, and just experiment in general. That's really what 
this is all about. I think this groove module could help to create some 
really new sounds and music. That's also why I suggested the 
extended quantizing module some time ago. I see those two ideas as 
related, or complementary. The quantizer idea basically would offer 
more grids which can be used for the cv-output of a sequencer, 
while the groove-module would offer more and flexible grids which 
can be used for the clock input of a sequencer. 

You could say that the groove-module approach is better because 
you can tweak each value individually, and the tuning-module idea 
depends upon a template (the n-tet) with only 1 variable n (same 
function as the number of steps-switch in groove module). But on 
the other hand tweaking the individual value of each step in a tuning 
would require so many pots that this would probably become 
unpractical. Definitely if you would let n go up to 24 (although 
personally I wouldn't mind having that possibility at all). But then 
again you could refine the n grids and make them more usable with 
an editable scale-mask, which is actually quite necessary for higher 
values of n. 

It's just such a pity that a lot of people have no idea what's behind scales 
and tunings, although it is right at the core of all music. That keeps 
you from exploring and you end up with the same notes over and over 
again while there are so many more notes... :-))

> > The typical delay-range for the individual steps would be somewhere 
> > between +/- 5 and 200 ms, so in this aspect this idea would require 
> > more precision than the A-162, which has a (+/-?) trigger delay of 
> > up to 10 seconds, and which is not very precise in the 15 - 50 ms 
> > range, where a lot of groove is to be found. 
> It is technically very simple to add an switch and an alternative 
> capacitor to reduce the delay time range. Like the range switches of the 
> A140 envelopes.

Yes, I've been thinking about such a modification. That's how I got this 
idea in the first place. 

What would be the preferred method to change the delay-range of the 
A-162 so the level indicators of the pot would roughly translate to ms 
x 10 (level 1 10 ms, level 5 somewhere around 50ms etc.) ?? Do you 
happen to know?  

> PS:
>  > I bet you a crate of dutch beer
> Never, never, never bet for dutch beer with someone born and living in 
> Bavaria. ;-)

Ok.. I won't, I have far too much respect for German beer ;-)

Greetings

Anton

To speak the truth I did not read all the groovalizer messages in detail but
how about this solution:

Clock a VC trigger delay (e.g. A-142 or A-162 with reduced timing capacitors
and the planned universal vactrol module) with the fixed (i.e. non-grooving)
master clock. A random or any other control voltage (sequencer, LFO,
Midi-to-CV, Theremin ...) is used to vary the delay time in the desired
range. The output of the trigger delay is the new grooving master clock.

Even clock dividing/clock sequencing functions (A-160, A-161, modified A-163
for DC coupling) can be used. One can choose if the clock dividing module is
connected pre/post the trigger delay mentioned above. In the first case
(i.e. the clock dividing module is driven by the non-grooving master clock)
an independend grooving unit can be used for each stage.

Not sure if this will solve all the grooving problems but you should try it.

For the planned synced LFO (A-143-4 ?, still in the planning stage) we will
see if it is possible to introduce grooving features.

Best wishes
Dieter Doepfer

> -----Ursprüngliche Nachricht-----
> Von: Doepfer_a100@yahoogroups.com
> [mailto:Doepfer_a100@yahoogroups.com]Im Auftrag von Anton Coops
> Gesendet: Freitag, 2. Dezember 2005 20:48
> An: Doepfer_a100@yahoogroups.com
> Betreff: Re: [Doepfer_a100] suggestion: groovalizer
>
>
> On 2 Dec 2005, at 15:30, Florian Anwander wrote:
>
> > Sorry, Anton, it should be kidding.
>
> I'm not at all offended
>
> > The main problem is, that the Seq-outCV-controls-VCO method influences
> > the tempo. This should not happen to my opinion.
>
> I agree, and even stronger, to me that makes this method quite
> useless, because the most important thing to me would be staying
> locked into the main beat and being able to do unorthodox things at
> the same time without getting lost. Think of the more extreme flows
> you hear in d&b/garage/grime grooves, or some of the stranger
> hiphop beats.
>
> The other main function of the module would be to finetune the feel
> of basslines, melodies, hihats etc. and creating all kinds of crazy
> loops. Also it can be refreshing to get out of the 4/8/16 grid and use
> a bit of 3, 5, 6 or 7, and just experiment in general. That's really what
> this is all about. I think this groove module could help to create some
> really new sounds and music. That's also why I suggested the
> extended quantizing module some time ago. I see those two ideas as
> related, or complementary. The quantizer idea basically would offer
> more grids which can be used for the cv-output of a sequencer,
> while the groove-module would offer more and flexible grids which
> can be used for the clock input of a sequencer.
>
> You could say that the groove-module approach is better because
> you can tweak each value individually, and the tuning-module idea
> depends upon a template (the n-tet) with only 1 variable n (same
> function as the number of steps-switch in groove module). But on
> the other hand tweaking the individual value of each step in a tuning
> would require so many pots that this would probably become
> unpractical. Definitely if you would let n go up to 24 (although
> personally I wouldn't mind having that possibility at all). But then
> again you could refine the n grids and make them more usable with
> an editable scale-mask, which is actually quite necessary for higher
> values of n.
>
> It's just such a pity that a lot of people have no idea what's
> behind scales
> and tunings, although it is right at the core of all music. That keeps
> you from exploring and you end up with the same notes over and over
> again while there are so many more notes... :-))
>
> > > The typical delay-range for the individual steps would be somewhere
> > > between +/- 5 and 200 ms, so in this aspect this idea would require
> > > more precision than the A-162, which has a (+/-?) trigger delay of
> > > up to 10 seconds, and which is not very precise in the 15 - 50 ms
> > > range, where a lot of groove is to be found.
> > It is technically very simple to add an switch and an alternative
> > capacitor to reduce the delay time range. Like the range
> switches of the
> > A140 envelopes.
>
> Yes, I've been thinking about such a modification. That's how I got this
> idea in the first place.
>
> What would be the preferred method to change the delay-range of the
> A-162 so the level indicators of the pot would roughly translate to ms
> x 10 (level 1 10 ms, level 5 somewhere around 50ms etc.) ?? Do you
> happen to know?
>
> > PS:
> >  > I bet you a crate of dutch beer
> > Never, never, never bet for dutch beer with someone born and living in
> > Bavaria. ;-)
>
> Ok.. I won't, I have far too much respect for German beer ;-)
>
> Greetings
>
> Anton
>
>
>
>
> Yahoo! Groups Links
>
>
>
>
>
>

On 3 Dec 2005, at 14:43, Dieter Doepfer wrote:

> To speak the truth I did not read all the groovalizer messages in detail but

the first mail has a short description

> how about this solution:
> 
> Clock a VC trigger delay (e.g. A-142 or A-162 with reduced timing capacitors
> and the planned universal vactrol module) with the fixed (i.e. non-grooving)
> master clock. A random or any other control voltage (sequencer, LFO,
> Midi-to-CV, Theremin ...) is used to vary the delay time in the desired
> range. The output of the trigger delay is the new grooving master clock.
>
You're absolutely right, modding the delay time of the A-162 and 
adding vactrol vc for time and length would give me the features I 
want, but I would personally use this only for short grooves of up to 
4 steps where I can use one A-151 and a mixer to trigger the A-162. 
Bigger patterns of 5-8 steps would require big patches, and using a 
sequencer to trigger the A-162 would not be that great, because you 
want to be able to trigger the sequencer itself with the resulting 
grooves.. 

So the module idea is really about having a simple intuitive user 
interface instead of a big patch, something that you could use live 
on stage without getting confused. 

> For the planned synced LFO (A-143-4 ?, still in the planning stage) we will
> see if it is possible to introduce grooving features.

Keep us posted! I think the module might be handy for 'true positive and negative 
delay' so to speak :-))

Thanks and greetings,

Anton

> Best wishes
> Dieter Doepfer
> 
> 
> > -----Ursprüngliche Nachricht-----
> > Von: Doepfer_a100@yahoogroups.com
> > [mailto:Doepfer_a100@yahoogroups.com]Im Auftrag von Anton Coops
> > Gesendet: Freitag, 2. Dezember 2005 20:48
> > An: Doepfer_a100@yahoogroups.com
> > Betreff: Re: [Doepfer_a100] suggestion: groovalizer
> >
> >
> > On 2 Dec 2005, at 15:30, Florian Anwander wrote:
> >
> > > Sorry, Anton, it should be kidding.
> >
> > I'm not at all offended
> >
> > > The main problem is, that the Seq-outCV-controls-VCO method influences
> > > the tempo. This should not happen to my opinion.
> >
> > I agree, and even stronger, to me that makes this method quite
> > useless, because the most important thing to me would be staying
> > locked into the main beat and being able to do unorthodox things at
> > the same time without getting lost. Think of the more extreme flows
> > you hear in d&b/garage/grime grooves, or some of the stranger
> > hiphop beats.
> >
> > The other main function of the module would be to finetune the feel
> > of basslines, melodies, hihats etc. and creating all kinds of crazy
> > loops. Also it can be refreshing to get out of the 4/8/16 grid and use
> > a bit of 3, 5, 6 or 7, and just experiment in general. That's really what
> > this is all about. I think this groove module could help to create some
> > really new sounds and music. That's also why I suggested the
> > extended quantizing module some time ago. I see those two ideas as
> > related, or complementary. The quantizer idea basically would offer
> > more grids which can be used for the cv-output of a sequencer,
> > while the groove-module would offer more and flexible grids which
> > can be used for the clock input of a sequencer.
> >
> > You could say that the groove-module approach is better because
> > you can tweak each value individually, and the tuning-module idea
> > depends upon a template (the n-tet) with only 1 variable n (same
> > function as the number of steps-switch in groove module). But on
> > the other hand tweaking the individual value of each step in a tuning
> > would require so many pots that this would probably become
> > unpractical. Definitely if you would let n go up to 24 (although
> > personally I wouldn't mind having that possibility at all). But then
> > again you could refine the n grids and make them more usable with
> > an editable scale-mask, which is actually quite necessary for higher
> > values of n.
> >
> > It's just such a pity that a lot of people have no idea what's
> > behind scales
> > and tunings, although it is right at the core of all music. That keeps
> > you from exploring and you end up with the same notes over and over
> > again while there are so many more notes... :-))
> >
> > > > The typical delay-range for the individual steps would be somewhere
> > > > between +/- 5 and 200 ms, so in this aspect this idea would require
> > > > more precision than the A-162, which has a (+/-?) trigger delay of
> > > > up to 10 seconds, and which is not very precise in the 15 - 50 ms
> > > > range, where a lot of groove is to be found.
> > > It is technically very simple to add an switch and an alternative
> > > capacitor to reduce the delay time range. Like the range
> > switches of the
> > > A140 envelopes.
> >
> > Yes, I've been thinking about such a modification. That's how I got this
> > idea in the first place.
> >
> > What would be the preferred method to change the delay-range of the
> > A-162 so the level indicators of the pot would roughly translate to ms
> > x 10 (level 1 10 ms, level 5 somewhere around 50ms etc.) ?? Do you
> > happen to know?
> >
> > > PS:
> > >  > I bet you a crate of dutch beer
> > > Never, never, never bet for dutch beer with someone born and living in
> > > Bavaria. ;-)
> >
> > Ok.. I won't, I have far too much respect for German beer ;-)
> >
> > Greetings
> >
> > Anton
> >
> >
> >
> >
> > Yahoo! Groups Links
> >
> >
> >
> >
> >
> >
> 
> 
> 
> 
>  
> Yahoo! Groups Links
> 
> 
> 
>  
> 
> 
> 
>

I guess the planned multiple synced lfo module could do that if the 
cv-input for phase shift would be implemented in such a way that the 
2nd lfo could be sent back in time (sort of) relative to the 1st lfo.  
This could then perhaps be controlled by feeding it a negative 
voltage or something like that

Another cv-input for variable pulse width of the rectangle output 
would add versatility if you would want to use this module for 
creating grooves. However if you want to go all the way in creating 
grooves of say 8 patterns you would need two A-155, one for 
creating the groove (timing, maybe gate length) and one for the 
pattern that you apply the groove on.

Anton

On 2 Dec 2005, at 13:45, Jilt van Moorst wrote:

> let me know how you plan to travel backwards in time to do negative 
> quantization
> 
> 
> 
>  
> Yahoo! Groups Links
> 
> 
> 
>  
> 
> 
> 
>

Hi Jilt

> let me know how you plan to travel backwards in time to do negative 
> quantization
;-)
In fact all sequzencers which provide a "negative delay time", either 
don't do the negative delay on the first beat (old Linn or Korg drum 
machines or Jomox Xbase09 in SyncSlave mode), or have count in bars 
(Logic) or simply delay the clock (for output) itself and are able to 
set some beats before the clock.
It is the same as with through zero flanging. You cannot get backwards 
in time, but you can delay your non modulated signal.

Florian

> It is the same as with through zero flanging. You cannot get backwards 
> in time, but you can delay your non modulated signal.

I would even say talking about positive and negative delay for 
refining a groove only makes sense if you have a really intuitive 
interface to do it because like you say, in the end it's all relative

Anton

> 
> Florian
> 
> 
> 
> 
>  
> Yahoo! Groups Links
> 
> 
> 
>  
> 
> 
>

Allo!

I've been thinking on this Groovalizer concept and I think it can be  
accomplished with very few modules and several different ways with  
the tools we have (or soon will have) available.

What we want to have is a sequence of notes with the possibility of  
sounding before or after the beat.  Right?

I really love my A-146 Variable Waveform LFO.  It produces a variable  
width pulse wave and a variable saw to triangle to ramp wave.  I  
patch these into Envelope Generators.    The variable width pulse  
gives a variable Gate for note durations.  But the variable saw wave  
allows you to trigger the EG at different times in the cycle.   
Comparing the pulse to the saw you are able to trigger the EG's  
simultaneously or at ever widening intervals.  A little knob  
twiddling can be fun!.

If you use the triangle setting as the starting point of the first  
cycle, then you can vary whether the next trigger occurs after or  
BEFORE the beat!  Time Travel aka Rubato!

As soon as the Universal Vactrol is available we may be able to  
remove the protoplasmic element from the knob twiddling and get  
repeatable variations using a Sequencer, for instance.

So Dieter, we pray that your UV will come with patch-point  
instructions for the A-146.  Pretty Please?

Similar effects may be obtainable with the A-142 Decay/Gate.  It  
produces an inverse gate when the delay is done and it is Voltage  
Controlled!  This one doesn't even need patch-points!  Is this the  
elusive Groovalizer?

Also perhaps the A-162 Dual Trigger Delay.

sfrules

Hi sfrules

> gives a variable Gate for note durations.  But the variable saw wave  
> allows you to trigger the EG at different times in the cycle.   
> Comparing the pulse to the saw you are able to trigger the EG's  
> simultaneously or at ever widening intervals.  A little knob  
> twiddling can be fun!.
> 
> If you use the triangle setting as the starting point of the first  
> cycle, then you can vary whether the next trigger occurs after or  
> BEFORE the beat!  Time Travel aka Rubato!

This is basically a nice idea, but it lacks completely the possibility 
of getting this synced to an external master-clock.

Also a groove requires to move definite clocks to another position and 
leaving all other notes on the same position.
E.g for a classical jack beat, you take a 4/4 beat counted in 1/16th. 
The beat number 4,8,12, and 16 need to be delayed around 10 % of the 
lenght of one 1/16th note. The beat number 2, 6, 10, and 14 should be 
delayed 5% only. Of course you can manage all this with an 
extraordinariuos amount of logical modul settings, but I have to admit, 
that I would prefer something more simple. I don't want to waste modules 
for 800 Euro only for getting a swing clock, which could be realized 
with a simple one chip solution.

Florian