Re: audio volume with multiple audio queues
Re: audio volume with multiple audio queues
- Subject: Re: audio volume with multiple audio queues
- From: Brian Willoughby <email@hidden>
- Date: Tue, 12 Aug 2008 16:58:54 -0700
When you do your own mixing, where do you get the data, and what
format are you processing? How are you converting -32768..32767 to
float for the audio queue? The resulting values should all be -1.0
<= x < 1.0. Any mistakes made here in your code will give different
results than when allowing the Audio Queue API to read and mix files
directly from disk.
How have you confirmed that you are not getting any clipping with 2
audio queues? You cannot rely on your ears alone. The only
definitive method is to use digital output and store the values on
another digital interface - or you could use an audio device with
calibrated level meters, although that method may not be able to
distinguish 0dBFS from 'over.' In other words, perhaps you are
getting clipping with 2 audio queues, but you just don't realize it
because the amount of clipping is much less.
I'm not sure it's safe to assume that the default gain is the maximum
of 1.0 when your only option to be sure is to access the Playback
Gain for each Audio Queue. Since you're having trouble, I would
recommend that you stop making assumptions and confirm the Playback
Gain. It can be queried as well as set.
Of course, I realize that you are here precisely because you want to
confirm your assumptions, but it seems that definitive answers from
the Audio Queue developers are not forthcoming. I think you still
might be able to find your own answers by carefully eliminating the
variables.
Brian Willoughby
Sound Consulting
On Aug 8, 2008, at 16:30, Peter Thom wrote:
I mean, when for example I'm adding 2 sine waves at full amplitude in
one audio queue, their max values are 32767 + 32767 (or -32768 + -
32768), and that creates heavy clipping.
if I use 2 audio queues, and have a full amplitude sine wave in each
queue, I don't get any clipping.
My understanding is that the audio queue gain is by default at 1.0
(maximum), so how would I make it louder?
Peter
On Aug 8, 2008, at 2:48 PM, Brian Willoughby wrote:
Of course you can.
If you are combining multiple files into one queue and getting a
different result than when the system combines multiple audio
queues, then there is either something wrong with your code or your
perception. I suppose it's possible that you're listening on a
laptop, and the built-in audio compensation is reacting
differently, although that is unlikely, but if you use the digital
outputs or a well-known audio interface, then you should be able to
get identical results either way. There is a Playback Gain for
each Audio Queue, have you examined these? Unless the overall
graph has some kind of dynamic plugin inserted, there should be no
automatic gain with Audio Queues.
On Aug 8, 2008, at 14:29, Peter Thom wrote:
Bump,
can I reach the same volume with one audio queue as I can with
multiple queues?
On Aug 6, 2008, at 10:26 AM, Peter Thom wrote:
When I run multiple audio queues (for example playback multiple
audio files at the same time), I can reach a higher volume than if
I just added the buffer signals into one audio queue (which clips).
Is there any way to achieve higher volumes with just one audio
queue? Like, is there a volume setting that I can increase?
_______________________________________________
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