Alan Stern wrote: > For example, the lsusb output shows that interfaces 1 and 9 both use > endpoint 1-out, and interfaces 2 and 10 both use endpoint 3-in. This > is a violation of the USB spec (unless the audio class has an exception > that allows it). This is not the first audio device that violates the spec ... I'd guess the driver is supposed to know which interfaces cannot be used in parallel. > In addition, the maxpacket sizes for endpoint 3-in on interface 10 are > 384 and 592 bytes for 16-bit and 24-bit respectively, rather than 196 > and 296 bytes. Why is that? The sample clocks are usually not synchronized to the bus clock, so the max packet sizes must be at least a little larger than theoretically needed. For practical purposes, there should be no more than one or two additional audio frames per USB packet, so rounding up as much as done here should not be necessary. Regards, Clemens PS: The SB Audigy 2 NX, in full speed mode, expects the OS to be able to send eight channels ot 16-bit data at 48 kHz. I'll have to dig it out and test it with the latest kernel. -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
