> So I played with the Shark audio a bit.  I found 4 problems, the first
> two of which I've fixed:
Great!  I'm not surprised there are still problems.  I had not tried
recording at all, and only some playing.

> 3) There is an interesting hardware issue.  Both the play and record
> channels must run at >=22050Hz, or both <22050Hz; otherwise the wrong
> crystal is used on the play channel, and it plays at half or double
> speed.  (This was masked in some cases by #1.)
Well, I guess we should set play and record to the same frequency
at all times then.  At least that is an easy fix.

> 4) At 44KHz, I get `warbling' output, which sounds very similar to the
> output of the SB driver before it did auto-initialize DMA.
I've had similar experiences.  Are you getting data over NFS?
Or using the network a lot?  As far as I can determine the ISA
bus in the shark cannot sustain both the ethernet controller and
the sound chip doing dma at the same time at 44kHz.  I checked
the time between interrupts and they are often far too long.
And it doesn't seem to be because of lost interrupts because
the DMA pointer is at the right place when the interrupt arrives.
I can only conclude that the sound chip cannot do DMA at the
rate it wants to.  Maybe I'm missing something obvious?

        -- Lennart