[linux-audio-user] latency: video card contribution

Mark Knecht markknecht at gmail.com
Wed Sep 28 16:18:45 EDT 2005


On 9/28/05, Dmitry S. Baikov <c0ff at konstruktiv.org> wrote:
> Hello!
>
> Another set of questions for experienced Linux Audio Users.
> Mainly it's related to laptop performace.
> It seems the choice of video system for modern laptop consists of two
> main alternatives:
> 1) dedicated high performance controller (nvidia/ati) with closed source
> drivers
> 2) shared memory controller (intel) with open source drivers
>
> People on Windows forums (no choice for Apples) prefer dedicated
> controller (with own video memory) because shared memory video degrades
> performance and increase latencies (they say, and in windows).
> I suppose, under Linux the things are different, because minimal
> possible latency is directly related to interrupt processing: closed
> source drivers have arbitrary interrupt paths, surely are written to
> maximise video performance and thus, should play a bad role in latency.
> Moreover they cannot be fixed. Open source ones at least can be fixed.
>
> Or I am completely wrong and shared video memory makes it bad on a
> hardware side (locking pci bus, for example)?

Shared memory is not the highest performance alternative in any
operating system. When the video memory is part of system memory then
the processor the video controller fight for memory bandwidth. This
slows both down.

>
> So, the question is: what to choose, integrated intel solution or
> ati/nvidia one (in this case, nvidia is preferred, because of driver
> quality).
>

Choose a good controller with a bit of dedicated video memory. For
purely audio apps you don't need all that much, but if you're going to
run video apps or do multimedia stuff then you'll want more.

HTH,
Mark




More information about the Linux-audio-user mailing list