Hi Tom. (Howdy Neighbor!)
My objective this week (month) is to learn how the sound card interface
works. Right now the Azve Input is working except I am not getting the audio I
expect. I'm just getting noise.
From 1984 to 1995 I programmed PCs in Assembly Language and C. In 95 I bought
my first version of VC++. So I have been programming in VC++ for over 10 years.
I have been programming microcontrollers since 1973. I have programmed most
of the popular uPs. Right now I am using PIC18s for the controllers in my
projects.
I am 74 and I have been retired for 10 years now. With gas prices like they
are I can't afford to go anywhere so i am staying home and spending time om
electronic(ham radio) projects. It's just something to do. It's a challenge but
really not important.
Bob Macklin
Kent, Wa.
Tom Lackamp <[EMAIL PROTECTED]> wrote:
Bob,
I almost hesitate to make this wacky suggestion, but maybe
it might help, at least with your initial experiments.
I've been fooling around with an MP3 player, mostly as an
exercise to re-familiarize myself with the finer points of Delphi.
(I've been playing with microcontrollers for the last few years,
and haven't done any PC programming. I'm very rusty.)
I found a really nifty DLL that helps immensely. It's called
bass.dll, and is available for free for personal use from
Un4Seen:
http://www.un4seen.com
The feature that might make bass.dll attractive for your
preliminary work is the ease of data retrieval from the
sound card. Give the DLL the BASS_ChannelGetData
command, and you have a buffer full of data.
It also provides some stand alone DSP tools, but mostly
geared toward music stuff. But it provides simple hooks for
you to add your own routines into the DSP processing chain.
I doubt if bass.dll would be your final solution, but it would
get you up and running with something relatively quickly and
easily that lets you get on with your experiments.
Good luck!
Tom Lackamp
Tukwila
(Howdy, neighbor! :-) )
Bob Macklin wrote:
> Alberto,
>
> Thanks,
>
> I am a retired EE. Of my 40 years as a designer I spent 30 of them in
> magnetic recording, both analog and digital, both tape and disk. I
> retired 10 years ago.
>
> It took me a little while today to remember that Intel data is low
> byte first. After swaping bytes on the 16 bit data I saw the bias on
> both 8 and 16 bit data.
>
> But now I know I am just geeting silence, not what is on the "Line In"
> input. And the mux selection is not a part of the Wave LIB functions.
> I thought it might be in the AUX functions but I have not found it yet.
>
> I see more noise in the 16 bit data than I do in the 8 bit data. But
> it is just noise.
>
> Bob Macklin
> Seattle, Wa.
>
> */i2phd <[EMAIL PROTECTED]>/* wrote:
>
> --- In [email protected]
> <mailto:soft_radio%40yahoogroups.com>, "Bob Macklin"
> <[EMAIL PROTECTED]>
> wrote:
> >
> >
> > Does anyone know of a place to find out how this data really works.
> > The only books I have found are multimedia reference manual from the
> > early 90's, Ie, pre-Windows.
>
> Bob, this site http://www.borg.com/~jglatt/tech/wave.htm
> <http://www.borg.com/%7Ejglatt/tech/wave.htm>
> will tell you much more than you need to know about the wave audio
> data formats.
>
> You are correct when saying that 8-bit samples are biased. The range
> is from 0 to 0xff, and the silence corresponds to 0x80.
> 16-bit samples must be looked at as signed integers, ranging from
> -32768 to +32767. Same for 24-bit samples, that range form -8388608 to
> +8388607
>
> 73 Alberto I2PHD
>
>
>
