A second reason for not being an audio device is that historically the SDR code has had to do weird sound card specific things to deal with the different audio drivers and mixer setups. I've got somewhat of a feeling if that we keep Windows grubby hands off our precious AD data we'll be better off.
A third reason is no one knows if a generic USB audio device can do 24 bit 192 khz sampling which is one of our prime goals with the AD chips we're looking at (Wolfson, TI, and Cirrus)
Right now, I'm writing code to suck the AD data out of the USB and stuff it into PowerSDR. It is simply using the USB code from the Xylo folks and we'll see how well it works. If someone wants to go off and look into doing a generic USB audio device with the FX2 and sorting out if we can do that and still also do other control functions on the FX2 that would be terrific, if my code gets obsoleted, no problem. At the moment we do not have a lot of FX2 or USB experts on the crew, so we're seeing if we can make it work with what we know how to do at the moment.
Regards,. Bill (kd5tfd) At 09:08 PM 1/22/2006, Christopher T. Day wrote:
The way these things usually go, when the driver first discovers the presence of the USB node, it asks it what it can do. The node may give multiple answers each with a different USB Device Type. The upper level driver then attaches a generic driver for the proper USB Device Type for each answer; all this is supplied by the OS. From the point of view of an application, it sees only the specific USB Device Type it is interested in. From the point of view of the node, it has to demultiplex the commands for the different Device Types, but that is what I thought the on-board 8501 of the FX2 chip was for. [Anyone who knows better, please don't hesitate to point out my ignorance.] Chris