Elektron Musical Instruments

> > well, the nord modulars (especially, the micro modular, the one that i
> know
> > anything about) have only a small few knobs, which is definitely not
> enough
> 
> but the sidstation only has four :) .. i can program that fine.

Only has four built-in.  It takes 2^7 "virtual knobs."


> Software would make it even better imho!

Actually, I shy away from a workstation-based GUI.  And what I'd prefer is
this: a "test sequence" menu button.  So that you can start sequences from the
front panel without using any piece of external midi equipment.

So you could just sit there and play whole songs with ONLY the SIDstation.
with 100 different sequences to the song.  With separate knob settings for
every single sequence.  

> well yes, i attach modules up to eachother (oscillators, filters) with
> virtual patch leads. It's not possible to program the thing from the front
> panel, but i don't mind. This is much neater than using real patch leads!. I
> can then assign the parameters to the knobs up front, as well as controller
> numbers. And i think this visual approach is surely the only way to handle
> real modular programming with something so small. The real great thing is
> this tiny size. You save 99 patches on there, then use it live as a synth or
> a sound processor.

but the small is non-multi-timbral :(  Nor is the SIDstation, completely,
but those wave tables rock.  I mean, they're a sequencer.

> the sidstation can be programmed very comfortably from the front, as the
> architecture doesn't become as complex. The oscillators are either routed
> through the filter or not, and then modulated with LFOs. I love the
> sidsound! .. but if it was a modularsid, (wow now that'd be cool!) i'd want
> an editor :)

does those nord thingies have sync'd oscillators?  do you have any idea
how they work on the SID?  does anyone?  if I ring mod three triangle
waves, does the last one ring mod the first?  or does the ring-mod'd 
output of the previous oscillator mod the first?  What about sync, I have
the same question.  

Has anyone sat down and generated a transfer function of the SID filter
using a pulse waveform as an impulse?  How about a 2-dimensional sampling of
all 8 filter combinations at 2^4*2^4 filter cutoff and resonance settings?
Has anyone characterized that?  Where can I see it posted, I'd like to know
what the transfer functions look like....because until you do that, you'll
never get the SID filter sound from any emulation.  Your "Chimera" will be
screeching loud because you didn't have the right deadspots in your filter.
Or would it be the other way around?  

Once you've got the filter figured out, then you can go on and try and
solve the Megahertz-bandwidth-output problem.  I'll give you a hint,
it takes more than an entire Intel PII350 worth of computation power to
even get close to the SID sound.  (I should know, I pegged reSID at 16X
in order to hear some AMJ tracks.  Damn if that isn't the one of the
most beautiful sounds in the world in the Assembly 96 song.)

-N