Subject: Re: [linux-audio-dev] XAP spec & PTAF comments [merge]
From: torbenh_AT_gmx.de
Date: Fri Feb 21 2003 - 12:46:34 EET
On Sat, Feb 08, 2003 at 11:55:00PM +0100, David Olofson wrote:
> On Saturday 08 February 2003 13.17, Steve Harris wrote:
> [...brachless clamp...]
> > Its not an instruction its just a bit of maths using fabs().
>
> Yes, of course! I keep forgetting that fabs() is an FPU operation. :-)
> (Life's not that nice with ints on most CPUs...)
>
>
> [...]
> > > Besides, if you have seriously heavy plugins in combination with
> > > this "a few effects at a time" behavior, the host could probably
> > > optimize this a bit without plugins explicitly supporting it. It
> > > means the host has to test buffers and figure out a clean way of
> > > activating and deactivating plugins without side effects, but if
> > > the plugins, or whole sub nets of plugins can be disabled, it's
> > > still a big win. ...and doesn't require any API support
> > > whatsoever, apart from the (de)activation stuff, which is needed
> > > anyway.
> >
> > You still have unpredicatable CPU load, which makes it pretty
> > useless.
>
> If you want to make sure you *cannot* overload the CPU with the
> current net, just have the host test it by turning all silent buffers
> into cleared non-silent buffers.
>
> Varying CPU load is not a problem in RT systems, but non-determinism
> is. They may look confusingly similar from a practical POV, but
> they're different things.
yes... you would only get more CPU time for GUI updates, if you
had silent meshes, which is a feature.
-- torben Hohn http://galan.sourceforge.net -- The graphical Audio language
This archive was generated by hypermail 2b28 : Fri Feb 21 2003 - 15:06:07 EET