Re: [LAD] Floating point Denormals: C++ and Waf

On Thu, 2012-08-02 at 11:28 -0400, Paul Davis wrote:
>
>
> On Thu, Aug 2, 2012 at 10:46 AM, David Robillard <d@email-addr-hidden>
> wrote:
>
> Is there a reason Jack can't do this for everything? I am not
> really
> keen on putting a bunch of mysterious assembler crap in a host
> meant to
> be a relatively clean example, and it's even worse to make
> plugins have
> to do this...
>
> JACK doesn't get a chance to handle the output of a plugin until the
> host hands it over to JACK and the process cycle ends. Just have a
> reverb plugin outputting denormals and then have the host apply a gain
> value and boom .. JACK can't play a role in that signal flow.

I mean set the flush-to-zero flag (and/or any other processor state
stuff required).

I know Ardour 'injects' non-zero data to fix denormals, but last I
tinkered with this, setting that flag made the problems go away. This
was many processor generations ago, though.

-dr

_______________________________________________
Linux-audio-dev mailing list
Linux-audio-dev@email-addr-hidden
http://lists.linuxaudio.org/listinfo/linux-audio-dev