[PD] accuracy of signal/message-objects

Frank Barknecht fbar at footils.org
Mon May 7 21:31:12 CEST 2007


Hallo,
martin.peach at sympatico.ca hat gesagt: // martin.peach at sympatico.ca wrote:

> I think the bangs are independent but they don't get processed until any pending samples have been dealt with.
> That was the impression I got when testing [comport] by setting the DTR pin using [metro]: there is always a jitter of a few milliseconds. Maybe it's the OS that causes the jitter, but it was about the same on WinXP and linux.

I think, it's comport's fault: [metro] generates clock-delayed
messages. These are like messages tagged with a time-stamp referring
to Pd's internal clock. However an object needs to actually use the
time-stamps and "look at the clock" to see what time it is. Objects
like [vline~] or [delay] do this, but comport doesn't.

> Also looking at the pd code, as far as I could follow it, the
> control signals seem to be processed in between signal blocks, so
> that, for instance, a metro bang that happens in the middle of a
> signal block will not be processed until after that block, so yes,
> the metro can tick whenever it wants but the bang won't occur until
> after the current sample block has been processed. If it happens to
> tick in between sample blocks then the bang will happen at the
> correct time.

I think, basically that is correct, that's why evaluating the
time-stamps of clock-delayed message is important, if you want to
avoid jitter. 

Ciao
-- 
 Frank Barknecht                 _ ______footils.org_ __goto10.org__




More information about the Pd-list mailing list