[PD] dynamically determind real latency?

Josh Steiner joschi at eds.org
Thu Mar 21 09:20:57 CET 2002


sorry for the "real audio latency" mixup, i meant "actual audio latency",
as you gathered.  :) 

What I was thinking of was a patch that would be able to auto sense the
latency of the soundcard and report that back to my control workstation so
i could compensate for the latency (I'm trying to make this system as
machine agnostic, and easy to setup as possible).  I was wondering if
there is some consistant formula that you can apply to determine a cards
latency mathematically.  Something to do with number of buffers or
something?  I guess if anyone knows any document describing the flow of
data from pd to the psysical dsp and what latencies are in between, I'd
love to read it.  If its a chapter in that massive "Everything You Ever
Wanted to Know about Compuer Music" bible (I forget the actual title), let
me know where, since my roommate has a copy that he tries to hide from me.
:)

Thanks again.

Josh

On Thu, 21 Mar 2002, sme wrote:

> hi
> to understand you right, are you talking about realtime audio processing or
> Real Audio-playback of a RealPlayer?
> to test latency of Pd you probably will want to know also the latency of the
> soundcard, which is "before and behind" pd. you can use a stereo signal (of
> a dat-player or minidisc...) from outside the computer and split one channel
> directly to the dat´s inlet and send the other through your computer. the
> time difference of the two signals gives you the latency.
> "synchronisation" would be another problem ...
> sme.
> 
> 
> ----- Original Message -----
> From: "Josh Steiner" <joschi at eds.org>
> To: "Pure Data" <pd-list at iem.kug.ac.at>
> Sent: Thursday, March 21, 2002 3:57 AM
> Subject: [PD] dynamically determind real latency?
> 
> 
> > I've been trying to think of a way to determine real audio latency using a
> > pd patch.  This is for my brother's and my project to control 5 computers
> > from 1 workstation using pd.  I'm kind of at a loss as to how one might do
> > this, or if its even possible.  If any of you have any suggestions as to
> > how this might be done, I'd really appriciate it.
> >
> > thanks again,
> >
> > Josh
> >
> > ________________________________________________________________
> > Josh .. Yoshi .. Joschi .. xiphoidprocess.com .. eds.org/~joschi
> >
> >
> >
> 
> 
> 

________________________________________________________________
Josh .. Yoshi .. Joschi .. xiphoidprocess.com .. eds.org/~joschi




More information about the Pd-list mailing list