Re: [FFADO-user] About the latency on Windows...

[Patrick N. <pat...@gm...>]

On 5/27/08, Arnold Krille <ar...@ar...> wrote:
>
>
> Congratulations! And I do believe you when you say that. That is why I
> worded "doubt". On the other hand ppalmers has an statement from a
> tc-electronics guy who said that the the windows kernel can't schedule
> faster
> then 10msecs (which would be almost as using 100Hz as tick-frequency in the
> linux-kernel).
> I don't know if that statement is true, because I am not a windows
> developer.
> I don't even run windows long enough to do extensive testings. Firing up
> the
> old MS for taxes is enough...
> Maybe it is possible to be faster on the plain windows kernel, maybe they
> use
> some ugly hacks (which would make support an even greater nightmare), maybe
> they even change half of the win-kernel to make it rt-able (which would
> explain why they all need a different driver:)...


(Sorry, I missed the beginning of this thread.)

I know my MOTU provides an ASIO driver for Windows.  It is supposed to give
the programmer direct access to the audio hardware.  I don't know if that
helps at all with the scheduler's latency.

http://en.wikipedia.org/wiki/Audio_Stream_Input/Output

Patrick