octal-dev
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: nope! still fresh


From: Timo Tossavainen
Subject: Re: nope! still fresh
Date: Tue Oct 31 01:18:02 2000

On Mon, 30 Oct 2000, Knos wrote:

> damn I'm a frenchman in finland and we still haven't got our first snow ;)

You were in the wrong place at the right time. It just snowed on sunday
and monday here in Tampere *<:)

Some thoughts on avoiding the zipper effect in control data: You could
make the control-rate higher than the event-rate in patterns and then 
interpolate the control-values when the user asks for it. The machines
control could then be updated faster than there's data in the pattern. 
The key is that you don't update the controls with too big steps
at a time. 

The machine programmer could also, at least in principle, take care of
this. Filters also have a problem; when you change the coefficients
on the fly you get transients. The internal state isn't what you'd
get from the data using the new filter parameters, but it behaves
normally after a while.

For instance if the machines work on vectors of 64/128 samples then
SR/vectorsize is going to be a lot faster than the rate of events in the
pattern. I don't think that it'll even cause a noticeable performance hit. 

I'm still out of audio stuff for a while... I currently have to research,
study and teach at the same time. 

Timo








reply via email to

[Prev in Thread] Current Thread [Next in Thread]