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

Re: Real-time category

Jeff Larson wrote:
> But this is a very unusual situation.  I doubt anyone running a 
> host like Bidule or Live with even one extra plugin like an amp modeler
> or reverb can use a 16 sample buffer.

I usually never need to go below 64 samples, and its only a matter of 
the amount of processing I want to do (I rarely use plugins, I prefer my 
own tweakings). If I just do what a looper is doing, a low vector size 
won't be a problem nowadays. Of course if you open yourself for the 
potential of a computer you want to use it all. Then you have to decide 
if you want the short latency or the impulse reverb. You might not be 
able to have both, and if you add a hardware looper to your set ;-)

And not to forget as soon you start to go into really soffisticated FFT 
based processing your signal vector size could go up to 4096 samples. 
But this is a completely different kind of fun... (And would apply to 
any RTOS-based computer as well)
Has anyone on the list ever experimented with a RTOS Linux kernel?

> The original point was that an RTOS will be better for triggering because
> it can usually give you lower latency.

The original point was the claim at the Loopers Delight web site that 
software could not be called a realtime looper.
Now I know its there to start interesting threads about tiny details 
which probably are not so important in a real lif(v)e situation in the 
end (so please keep i there ;-)



  [][]  [][][]  [][]  [][][]  [][]  [][][]  [][]  [][][]

         Stefan Tiedje
     Electronic Composition

     \\\   /|() ()|\
     ))))) )|  |  |( \\\
     ///    \ \_/)/ )))))
             \___/   ///

--(_|_ ----|\-----|-----()---
-- _|_)----|-----()----------

14, Av. Pr. Franklin Roosevelt, 94320 Thiais, France
Phone at CCMIX   +33-1-49 77 51 72