Re: [fluid-dev] DSP testing

[Josh Green]

On Tue, 2004-03-30 at 02:21, Peter Hanappe wrote:
> 
> > I have a feeling there are probably some issues with stereo output in
> > regards to the phase offset between right/left speakers. Is there
> > currently a way to guarantee multiple voices to all start at the same
> > time? Playing the same note can often sound different for each note on.
> 
> No, there currently is any way to guarantee they start together. It
> could happen that the audio thread intervenes between the initialization
> of the first voice and the initialization of the second voice, in which
> case the two voices are shifted by 64 samples. I should test whether
> this really happens and think of a way to fix it if it does. Good point.
> 
> What suprises me is that it makes that big a difference that you hear
> it. Is there any way you could record it?
> 

As Tim Goetze mentioned, a 64 sample offset is enough to expose
different phasing effects between samples. I'll see about recording an
example.

> 
> > I've noticed that the damping parameter has no effect (at least as far
> > as I can hear). Perhaps its just really subtle in its effect?
> 
> Did you try the version in CVS? I find that the reverb has improved
> and the effects seem more intuitive now.
> 

I just downloaded it and gave it a try. Wow! Much better sounding reverb
:)  Instruments sound much more vibrant now. I still can't notice any
change though when modifying the damping parameter. Perhaps I should try
the original Freeverb to see if it has the same effect (or non-effect
rather).

> > Is the TODO file up to date? I'd like to see about helping with some of
> > the items on that list.
> 
> I cleaned up the TODO file a bit. Don't add too much stuff to it. It's
> demotivating to not see the list shrink.
> 

Right, I was thinking of helping to shrink it :)

I went and changed the filter turn-off optimization to keep the filter
on, once it turns on (now in CVS). I used the voice->filter_startup
variable to determine this. I notice, though, that there is still the
potential for a click when it gets turned on. I was curious why the
filter doesn't always interpolate between values, even on startup? If
the "old" values of the filter were initialized to un-audible values (Q
= 0, FC = 20khz) this may prevent clicking when the filter gets switched
on. An alternative is to do away with this all together and always have
the filter enabled. I wonder how often this optimization is actually
used, it doesn't seem worth it if its going to cause clicks.

I wasn't able to figure out how to get the filter to interpolate during
startup though (either that or it still causes a click).

> Cheers,
> Peter

Cheers.
        Josh Green