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

[Mark Knecht]

On 9/28/05, Dmitry S. Baikov <c0ff at konstruktiv.org> wrote:
> >  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.
> >
> It's hard to find light laptop with dedicated controller at a reasonable
> price :)

I just sold mine on Ebay this morning. (At least I think it wasn't
shared memory. Hardly remember anymore. Anyway, it's gone...)

> Is it possible to do 8in/8out and 64-sample buffers with RME Multiface
> CardBus on a laptop with shared video memory?

Depends on the laptop but possibly. Why so stringent about 64/2? Do
you _really_ require sub-3mS?

>
> And the only one(and expensive one) I found is with nvidia controller
> (Sony Vaio VGN-S480). Not sure about nv TurboCache - is it a shared
> memory solution?

Couldn't say.
>
> Seems i want too much to be in one unit.
>

Possibly, but even shared memory probably won't kill you if you're
just running Ardour, etc.

- Mark