Re: Few sample rate conversion questions
Re: Few sample rate conversion questions
- Subject: Re: Few sample rate conversion questions
- From: William Stewart <email@hidden>
- Date: Wed, 28 Jan 2009 12:38:28 -0800
On Jan 28, 2009, at 5:42 AM, Mike Kluev wrote:
On Tue, 27 Jan 2009 17:20:01 -0800
William Stewart <email@hidden> wrote:
On Jan 27, 2009, at 12:21 AM, Mike Kluev wrote:
A side observation: I noticed that I could mimic behavior of
AUVarispeed variable playback rate by using AUConverter and
updating it's source/destination format's rate (the property
that updates formats is not read-only with AUConverter). This
leads to a question why to have AUVarispeed at all?
for dynamic control of the playback rate. Seriously, if you are
wanting to do a basic sample rate conversion you should use
AUConverter
But is dynamically changing playback rate of AUVarispeed any different
than changing srcFormat.mSampleRate / dstFormat.mSampleRate
ratio of AUConverter on the fly?
Yes - you have a substantial change (a new audio converter), so there
is alot of memory thrashing (allocating and so forth) when doing this
with AUConverter and you can very easily get gaps in playback...
Normally, with AUs we don't allow you to change the stream format
while the AU is Initialised (because changing a format involves
allocations and that is what Init is for). With the converter we have
to be more flexible, so we have logic in there to guard against a
change like this and the need to render - which is why you might get
gaps in playback if you do this while the converter is in a rendering
chain
Changing the rate is a parameter change in the varispeed and can be
done at any time. Internally it uses a different construction and
interface to the underlying rate converter to enable it to do this.
Bill
How?
Mike
_______________________________________________
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