Re: Another newbie question -- format of raw data
Re: Another newbie question -- format of raw data
- Subject: Re: Another newbie question -- format of raw data
- From: William Stewart <email@hidden>
- Date: Mon, 31 Jul 2006 18:39:42 -0700
On 31/07/2006, at 1:03 PM, Bill Cunning wrote:
Hi all,
I have been fighting this beast for a while now, and progress is
being made slowly but surely, but It seems I have hit a brick wall.
I can capture the audio data using the ‘guts’ more or less of
complex play through and another sample … and the raw audio from
the mac à> to server à> back to mac works fine. The issue seems to
be when I try and send it to a PC or run it through an ADPCM Codec
that all heck breaks loose . mac -> codec-enc ---- codec-dec – mac
fails with loud screeching noises and sending it off to a pc gives
about the same results.
It appears I am getting 32bit audio data
This will be floats with a range of -1 < 0 < 1
from the hardware (Bytes per frame == 4) single channel and the raw
data appears to be BigEndian (00 00 xx xx) ---
I found the flags to set 16bit capture but setunitproperties fails
if I try and change it -- that would be question one … is it
possible to change the #bytes
Not just this way. The ASBD is a rather abstract structure with many
internal dependencies - for instance, if bytes Per Frame changes,
then bytes per packet needs to change and probably bits per channel -
if you are going from float to int, then the flags also change,
etc... This has been discussed many times on the list so a search
through the archives should throw some light on this.
in captured stream? If I configure from midi configuration utility
it seems to let me set it to 16 bit, but this seems to be ignored
when I open the device / audio unit
The other question is is this data stream really just LinearPCM or
is there some other kind of encoding going on.
From the device or an AU - Linear PCM (float)
If it was just audio data, it would seem I could do something like
value/256 à out à value *256 and it would work
slowly - CA API services also provide efficient conversions to
convert to/from float/int formats - the AudioConverter being the main
API that does this - you can also set these different lpcm formats on
the output unit and it will handle it for you - have a look at the
DefaultOutputUnit code in /Developer/Examples/CoreAudio - it also has
the various different set ups for ASBDs for float, 16bit int, 24 bit
int.
Bill
.. but this (test case that has no use) causes playback to fail. Is
there some other kind of information encoded in the bitstream that
is getting lost when I try and process it this way.
Thanks for your help
Bill
iLinc Communications
Bill Cunning
Senior Software Developer
email@hidden
www.ilinc.com
165 Jordan Road
Troy, NY 12180
877.960.1700, X454
_______________________________________________
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
--
mailto:email@hidden
tel: +1 408 974 4056
________________________________________________________________________
__
"Much human ingenuity has gone into finding the ultimate Before.
The current state of knowledge can be summarized thus:
In the beginning, there was nothing, which exploded" - Terry Pratchett
________________________________________________________________________
__
_______________________________________________
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