Jens Axboe wrote:
> > Audio drivers can use this for power saving, and also to turn off the
> > unused mixer input when it isn't being used to reduce noise in the output.
>
> Cool, I was unaware that such a system existed! Sounds like just the
> thing we need.
It really would be nothing to invent, and that would be worth doing if
the notifier.h isn't quite right for this jon.
> You don't always need to fire up the audio device when a CD is played,
> I frequently just use head phones plugged into the front of the drive.
I tried plugging headphones into one, and the output was horrible. I'm
surprised, because through the sound card, via two unshielded cables
draped across the whole computer, it sounded pretty good.
Another drive doesn't have a headphone socket or volume control.
Just a play button, oddly.
> > You could include volume requests, as a range of 101 command values (to
> > preserve the parameter as CD number). Can the IDE driver tell when a
> > drive doesn't honour volume settings, so the mixer should be doing it?
>
> Lets not talk IDE, but CD-ROM instead. All volume requests and CD
> play operations go through the uniform CD-ROM layer. A drive
> can fail the volume request naturally, if it doesn't support it.
I was thinking of a specific drive, the one I have at work. It's an
idea drive. All cd player apps with a volume slider, ranging from 0% to
100%, give this behaviour: 0% == no sound, 1-99% == fixed volume.
> At probe time we also detect whether the drive does support CD
> audio. Why would you need the specific volume? If there's some
> kind of output, fire up the audio device. Some CD drives don't
> even support a range of volumes, just on/off.
Precisely. You'd want the specific volume from the application that's
calling the uniform CD-ROM layer, so you can pass it to the audio mixer
in the case that the drive only supports on/off.
So my question is: can you detect the difference between a drive that
only does on/off and a drive that has its own volume control?
have a nice day,
-- Jamie
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Mon Jan 31 2000 - 21:00:24 EST