[wdmaudiodev] Re: Audio input summing under Vista
- From: "Jeff Pages" <jeff@xxxxxxxxxxxxxxxx>
- To: <wdmaudiodev@xxxxxxxxxxxxx>
- Date: Wed, 22 Feb 2006 09:53:12 +1100
Hakon,
Thanks for your detailed reply. Our cards are principally used in the
professional audio environment, most notably sound broadcasting, and the input
summing node was requested by a customer using our card with broadcast
automation software, where audio coming from multiple sources (in their case
multiple satellite channels I think) needed to be available at a single wave
input device. Of course that could be done with an external mixer, but
professional-quality mixers aren't exactly cheap and, for a low-budget station,
having a sound card that included the mixer on its front end was a bonus for
them (and a sale for us!).
It's not a major issue for us, and my main concern is that any of our existing
customers who are using that feature may suddenly find it broken when they
upgrade to Vista. Physically each of our inputs has its own A/D converter, so
how they are presented is purely a firmware/driver issue. I have no problem
presenting each physical input as a separate device, but it's just the
backwards compatibility issue that's gnawing away at me.
By the way, will peakmeter nodes be supported under Vista? I've noticed under
build 5270 that they're not being mapped into the mixer API. This really is a
big issue for us, as one of our major customers relies heavily on having access
to peak meters.
Jeff
----- Original Message -----
From: Hakon Strande
To: wdmaudiodev@xxxxxxxxxxxxx
Sent: Wednesday, February 22, 2006 8:29 AM
Subject: [wdmaudiodev] Re: Audio input summing under Vista
Jeff,
With Vista and the UAA initiative we are taking big steps towards a more
transparent audio hardware ecosystem (I am referring to the consumer market
here - the 95% case) and through full discoverability of audio device
capabilities and the move towards independent devices we hope to create a
better user experience in our operating systems of the future for the regular
PC user.
One important aspect of the usability and discoverability of logical audio
devices exposed by an audio adaptor is that they are independent from each
other. By this I mean the audio solution has enough [converter] resources to
treat each logical input and output device/connector independently. In the
cases where multiple audio endpoints connect to a single DAC or ADC resource it
is important that the operating system is made aware of this through defined
driver OS interfaces so the user can be informed by the OS what device he can
expect to be active/operational. In the hardware realm, one of the UAA
technologies support the exposure of the system implementers intentions for
logical audio device support through a defined firmware method (read about Pin
Configuration registers in the HD Audio 1.0 spec or here if you want to learn
more about that).
Ideally all inputs and outputs on a PC are wholly independent and the
capabilities of the audio device completely discoverable and that is the
direction we are driving for Vista and beyond. What we want to enable is total
transparency through simple audio path designs in hardware that enable more
powerful OS audio policy/troubleshooting features and more flexibility with the
use of the audio device for our future OS features and for media applications
running on Windows.
This is equally true on the input side where when the ADC resources are
limited but the number of logical input devices exposed outnumbers the
available ADCs. While this in itself is not recommended, the muxed behavior is
preferable in the context of predictability and user friendliness. It is hard
for a normal user to understand why the integrated microphones on the laptop
still are recording after plugging in the headset into the Line/Mic jack on the
side for instance. In Vista we discourage the use of mixed input devices
sharing one ADC for usability reasons and our operating system will treat these
(when exposed by the driver) as muxed devices sharing one ADC.
It seems you are missing the ability to expose a multi input mixed device and
my question would be, why? What common user scenario is supported by mixing
multiple inputs to one ADC?
Sincerely,
Hakon Strande
PM Integrated, Internal, External, and Wireless Audio Devices
MediaTech/DMD/Windows Client/Microsoft
------------------------------------------------------------------------------
From: wdmaudiodev-bounce@xxxxxxxxxxxxx
[mailto:wdmaudiodev-bounce@xxxxxxxxxxxxx] On Behalf Of Jeff Pages
Sent: Sunday, February 19, 2006 4:41 PM
To: wdmaudiodev@xxxxxxxxxxxxx
Subject: [wdmaudiodev] Audio input summing under Vista
With our multichannel sound cards, we have a summing node (KSNODETYPE_SUM) in
the input topology so that multiple physical inputs can be summed into a single
waveIn device. Under Vista, each input bridge pin creates its own waveIn
device, so it would appear that summing multiple inputs to a single device is
no longer possible. Is this in fact the case, or is there something I'm missing?
Jeff Pages
Innes Corporation Pty Ltd
Other related posts:
