[sdiy] Groove jitter in step sequencers

[cheater cheater]

It would be interesting to make something that simulates different
kinds of groove. You basically have a single master clock, and then a
"groove clock" which is around the same frequency as the master clock,
maybe more, maybe less. This groove clock has programmable jitter. The
final clock output works like this:

1. the master clock sends out a pulse which is latched somewhere
2. the pulse waits until the next groove clock happens and then is
sent out to the clock output


You could do the programmable-jitter clock this way: have a
current-out DAC working at clock rate. Have a capacitor. Every time a
new groove clock pulse happens, the capacitor is unloaded. Then, the
DAC pulses the capacitor with some sort of current. Of course the
higher the value, the higher the charge is going to be at the
capacitor after the pulse happens. Then, start discharging the
capacitor (or maybe just always have it grounded through a
current-limiting resistor?). When the capacitor falls below a certain
value, checked by a comparator, have the circuit send out a pulse.

You could have samples, maybe even looped samples, playing back
through the DAC. You could for example sample the jitter of the 808 by
checking when it does respond to interrupts (I'm sure someone creative
would find a way to do it). You could also make the sample index reset
every fourth bar, or something like that, to make the groove more
mechanical. Of you could have a looping sample of a length which is
not a divisor of the bar length, to have a nice jitter that does not
repeat and that "softens up" the rhythm.

Finally, both clocks could be run by analogue VCOs for more fun.

Cheers,
D.

On 11/11/2010, dancemachine <dancemachine at gmail.com> wrote:
> feel that midi groove.   since i started using midi for clock only I
> am having great results.  all thanks for flame clockwork, which gets
> all weird as soon as you send it notes instead of clock only
>
>
> On Thu, Nov 11, 2010 at 11:31 AM, Florian Anwander
> <fanwander at mnet-online.de> wrote:
>> Hi Nick
>>
>>> Hello, anybody have anything to say about the inherent 'groove' and or
>>> 'pocket of old hardware sequencers?
>>> MSQ700?
>>> MSQ100?
>>
>> No inherent groove here. Everything which could be "inherent groove" like
>> in
>> a TR808, is overridden by the inherent delay which is provided by the
>> length
>> of MIDI note events and the serial transmission of MIDI. Each MIDI Note
>> event takes 1 ms. If you play a three note chord, the third note will come
>> 2
>> ms after the first one. This will cause more unwanted "inherent groove"
>> than
>> any wiggling inside the devices clock reading could do.
>>
>> Florian
>>
>