[linux-audio-dev] Re: [linux-audio-user] denormals or ... ?

[Tim Goetze]

[Leonard Ritter]

>On Tue, 2007-03-13 at 10:08 -0500, Dave Phillips wrote:
>> So, if anyone has anything to add to Chris's assessment, I'd like to 
>> know about it. If the problem is indeed related to denormals, is there
>> a 
>> way to fix it ? Comments and suggestions are most welcome.
>
>i made a few experiences regarding to denormals in Aldrin/libzzub.
>numerous plugins had their own fixes for denormalization (and i know
>that CAPS has a fix too, which is to add a -250dB square wave to avoid
>the value to go off the scale). AMD cpus do not have that problem most
>of the time, so the problem relates mostly to Intel cpus.

Unfortunately, there seems to be a denormal problem with at least 
AmpIV in CAPS.  The fix you mention works reasonably well for most 
plugins, but in the Amp series there's a simple dc-blocking IIR 
running at 8x fs that as far as I remember isn't 'renormalized' -- I 
keep meaning to verify the problem and this likely culprit, but I 
never seem to get around to it.

Dave, I think your problems may well be caused by denormals; in fact I 
think it's the most probable cause.  Does the problem persist if you 
remove AmpIV from the processing setup?  The other CAPS plugins 
_should_ be safe (can't guarantee 100% though).

Cheers, Tim