Re: Preroll semantics & clarification of kAudioUnitProperty_OfflineRender
Re: Preroll semantics & clarification of kAudioUnitProperty_OfflineRender
- Subject: Re: Preroll semantics & clarification of kAudioUnitProperty_OfflineRender
- From: Kyle Sluder <email@hidden>
- Date: Wed, 21 Dec 2011 12:55:14 -0800
On Wed, Dec 21, 2011 at 1:16 AM, Heinrich Fink <email@hidden> wrote:
> So my actual question was if some hosts compensate for drift against the wall clock also during playback, so that the begin and end times of a longer clip would actually be accurate to a clock outside the device’s clock. But I guess that such decisions have to be accessible as user options (such as those from Logic as you mentioned). From your comments I understand that users would probably accept a slight time off or losing a few frames due to sync at the boundary of their clips in favor of bit-accurate and high-quality playout during the clip.
I studied this problem a bit while building a (never-released) radio
automation program.
You just need to anticipate drift. The CPU might get bogged down, your
software might have a bug, the temperature in the room might exceed
the video card's operating parameters… In the case of radio
automation, it's important to never have dead air. But often the
schedule is set before the durations of the program's elements are
known—for example, a spot might be scheduled for 30 seconds, but it
really only lasts 29.
This is where the audio engine's understanding of syncing ends and
your app's domain knowledge begins.
--Kyle Sluder
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Coreaudio-api mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden