Matthias Drochner wrote:
The value in the step part of the amplifier capabilities is a step size in 0.25db increments where a value of zero maps to 0.25db step rate and 1 is 0.5db. So they are valid if somewhat fine grained. The step size has no bearing on the range of numeric values accepted by the amplifier as a gain value. In all cases that will be 0 up to the number of steps specified.tnn%NetBSD.org@localhost said:kernel: integer divide fault trap, code=0That was me. I'm working on a fix, will be in the tree in a couple of minutes. (I'm not sure a mixer with step 0 or 1 is good for anything, but at least it should not crash.)
I've not long finished writing a HD Audio driver (UAA compliant) for windows so I can help out with any other spec ambiguities if you need it.
Mike