[wdmaudiodev] Re: Real Time Processing
- From: "Bert Schiettecatte" <bert@xxxxxxxxxxxx>
- To: <wdmaudiodev@xxxxxxxxxxxxx>
- Date: Thu, 3 Mar 2005 09:53:05 +0100
Are you sure you need to use Windows XP as your OS? You might be better
of with Linux or writing a command line DOS program for what you need to
do? It sounds like you want direct access to the sound interface without
going through layers of abstraction.
Bert Schiettecatte
> -----Original Message-----
> From: wdmaudiodev-bounce@xxxxxxxxxxxxx
> [mailto:wdmaudiodev-bounce@xxxxxxxxxxxxx] On Behalf Of
> Pieterjan Demarcke
> Sent: donderdag 3 maart 2005 0:34
> To: wdmaudiodev@xxxxxxxxxxxxx
> Subject: [wdmaudiodev] Re: Real Time Processing
>
>
> Hello,
>
> thank you for the extremely fast answers! (btw I'm from
> Belgium, not such a hard guess if you look at my name indeed ;-))
>
> This ASIO interface indeed seems to be a better solution than
> WDM but there's only 1 problem: for my project i can (have
> to) use a computer owned by the university and i don't think
> buying a new professional soundcard is an option. So I'll
> have to stick with a standard one. Maybe this site
(http://www.asio2ks.de/) is a solution?
To clarify what the purpose is of the project: we have to make a
profilometer (measures relief at microscale) with a dvd OPU (optical
pickup unit, the laser head and surrounding electronics of a
dvd-player). The OPU uses the astigmatic method of focus error
detection, so it is possible to measure the amount of out-of-focus by
summing and substracting 4 specific output signals of the OPU. The
resulting signal (DC because the sensed variable is changing relatively
slowly) is transformed to an AC signal so the sound card can detect the
amplitude. The pc then sends a rescaled value of that amplitude to the
line-out to activate the focus actuator of the OPU. When the OPU is back
in focus, we can move the OPU for example 1 micrometer further and after
all that we are ready to sample again. (in fact this is simplified: the
opu is moving at constant velocity, but the idea is the same) The big
problem now with a latency is the
speed: assuming we want to have a resolution of 1 micrometer and there
is a latency of 2 ms, we quickly find a scanning speed of 1 mm/2 sec.
This is still acceptable because we never scan further than 1 cm.
So i'm looking for the fastest solution with a normal sound card. If it
seems to be that the minimum latency is for example 10 ms, so be it. But
the faster the better of course.
thank you very much for your input and goodnight (for those in europe)
Pieterjan.
******************
WDMAUDIODEV addresses:
Post message: mailto:wdmaudiodev@xxxxxxxxxxxxx
Subscribe: mailto:wdmaudiodev-request@xxxxxxxxxxxxx?subject=subscribe
Unsubscribe:
mailto:wdmaudiodev-request@xxxxxxxxxxxxx?subject=unsubscribe
Moderator: mailto:wdmaudiodev-moderators@xxxxxxxxxxxxx
URL to WDMAUDIODEV page:
http://www.wdmaudiodev.de/
******************
WDMAUDIODEV addresses:
Post message: mailto:wdmaudiodev@xxxxxxxxxxxxx
Subscribe: mailto:wdmaudiodev-request@xxxxxxxxxxxxx?subject=subscribe
Unsubscribe: mailto:wdmaudiodev-request@xxxxxxxxxxxxx?subject=unsubscribe
Moderator: mailto:wdmaudiodev-moderators@xxxxxxxxxxxxx
URL to WDMAUDIODEV page:
http://www.wdmaudiodev.de/
Other related posts:
