[sdiy] Simple MIDI-syncable LFO...?

[Richard Wentk]

At 15:56 18/10/2004 +0200, Fredrik Carlqvist wrote:

> >The only thing that's harder about a digital design is that you have to
> >write some code.
>
>... and learn how the microcontroller works and the assembly language and a
>development system and how to debug microcontrollers and where to get
>microcontrollers and how to get the program into the microcontroller...

I'm assuming people know all that stuff already. And even if they don't, 
I'm still not convinced it's all that much harder than getting your head 
round the analogue equivalents.

>There are many things one must learn before the microcontroller solution is
>"easier". Also, with a simple analog LFO, the huge ker_thunk can be reduced
>with an LFO speed pot. And you can make a soft sync so that there is no or
>at least very little ker_thunk.

I'm not just talking about an audible ker-thunk, I'm talking about the fact 
that the syncable design is inherently a kludge, in that you don't have 
genuine clock sync at all - just some random-ish reset event on top of a 
free-running analogue LFO whose frequency could be anything. Functionally 
they're just not comparable.

>The total part count is obviously lower with a microcontroller solution. It
>is also more versatile. But it can be much more expensive in terms of time
>(and money if you need a 16-bit DAC) and full of bugs.

Arguably, but once you have your design repeat costs are much lower, and 
the extra versatility makes the effort more than worthwhile.

Musically it's fantastically useful to have genuine clock sync. (And of 
course it can be sync from another LFO, not just MIDI.)

It's also interesting that with faster hardware the solution potentially 
scales to VCOs as well, so you can use some very similar ideas and code to 
track and follow the pitch of an incoming waveform from a VCO.

Richard