Re: [LAU] Session management with NSM

From: Fons Adriaensen <fons@email-addr-hidden>
Date: Thu Sep 04 2014 - 20:29:18 EEST

On Thu, Sep 04, 2014 at 03:34:57PM +0100, Rui Nuno Capela wrote:
 
> as said, automatic latency compensation in qtractors audio recording
> is a one-time settlement--it gets settled at the time you take the
> recording. nevermore it gets auto-adjusted later on, ie. the take
> offset value, as read from jack_port_get_latency*(), won't ever
> change thereafter. that means you're in charge to tweak it later if
> you please, by summoning the clip properties dialog and changing the
> offset value as you wish.

You only get correct latency compensation if you compensate both
when recording (the samples are older than they appear), and when
playing back (the samples will be output later). In both cases you
need to apply the _current_ value (Li, resp. Lo).

OTOH, there is never any need to store the offset. When recording,
just apply it to the stored position and forget about it - the
samples are now at their correct position (the one that corresponds
to having no latency).

> rest assured that, in qtractor sequencer model, ever since its
> inception, all time locations do honor the midi/music metric scale
> resolution.

For audio tracks that makes no sense. Aligning the start of the
track doesn't mean the content will be aligned, and that is what
matters if you want to align everything to the MIDI timescale.

Ciao,

-- 
FA
A world of exhaustive, reliable metadata would be an utopia.
It's also a pipe-dream, founded on self-delusion, nerd hubris
and hysterically inflated market opportunities. (Cory Doctorow)
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user@email-addr-hidden
http://lists.linuxaudio.org/listinfo/linux-audio-user
Received on Fri Sep 5 00:15:02 2014

This archive was generated by hypermail 2.1.8 : Fri Sep 05 2014 - 00:15:02 EEST