Archive of the former Yahoo!Groups mailing list: The Yamaha AN1x Synthesizer mailing list

previous by date index next by date
previous in topic topic list  

Subject: RE: OSC. Drift

From: Kevin Conder <kwconder@...>
Date: 2005-03-05

Ed Edwards wrote:

> Now you've gone and dipped into the pool of the unresearched.
> How do analog oscillators drift, and where do we find that data?

The data are out there. Try http://scholar.google.com/ and search for
"oscillator drift" or "thermal drift". After reading all those links,
you should ask for college credit!

> The worstest thing about random is that computers don't like random.

I'm not a mathematician but I play one on the computer...

Computers are deterministic machines, that means generally when you ask
it to do something then it will always produce the same result. So
getting it to be random, or to produce different results based on the
same input, is difficult. The best we can do is to use Psuedo Random
Number Generators (PRNGs).

A PRNG will produce a long list of numbers. However, the list never
changes. So, to appear random, you need to select a different starting
place on the list each time you read it. Your starting place on the
list is called a seed value.

Seed values are usually chosen based on something always changing like
the current time, line noise, or even a lava lamp!
(http://www.wired.com/wired/archive/11.08/random.html )

> One idea is to map velocity or key to change the envelope generator
> in some way. Sort of pseudo-random will result.

Since humans never hit a key with the same force (except when emulating
Jerry Lee Lewis's "Great Balls of Fire"), I think velocity would make a
good seed value!


=====
--
Kevin Conder

__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com