Re: [Jack-Devel] Very low latency audio

On Fri, Sep 20, 2013 at 1:35 PM, John Reed <[hidden]> wrote:

> On my netbook, I was able to start jackd with
> jackd -P62 -dalsa -dhw:0 -r44100 -p16 -n2
> but the hardware driver didn't support -p8 or -p4.
>
Indeed very few cards support that low buffer size: apparently one of the
Creative SoundBlaster series starts on -p8 but xruns internally...
I've heard that some RME gear is capable of running at such low latencies
too. No experience with those RME interfaces (the PCI-E stuff IIRC).


> -p16 already requires a lot of fast interrupts, so I assume that to use
> this reliably, I would have to be very careful about setting all of the
> kernel switches, etc.
>
Yes it does: and in doing so creates conciderable DSP overhead. That said,
I've ran jack at -r96000 -p16 -n2. That's an output latency of 0.3ms...
pretty good!


> Is there any hope for getting -p8 or -p4 working, even if I have to write
> my own custom device driver?
>
I'll leave this to more experienced hands: I'm not sure.


> Or has anyone done any work in connecting xenomai to jack? Would I be
> better off writing a xenomai application and then find a way to connect it
> to an ALSA driver directly?
>
Nope, but I would be interersted in results / experiences if you do decided
to go that route.

My background is my final-year-college project, investigating real-time
threading / messaging for audio-applications.
Code available: https://github.com/harryhaaren/fypRealtimeCppPrograming

A system had to be tuned to handle these latencies without Xruns: perhaps
interesting to see what I had to do to get a 6 year old laptop running
0.3ms reliably :)
Blog post: http://openavproductions.com/real-time-latency-tuning/

Interesting project idea, I'll keep an eye on it! -Harry

1379681614.14790_0.ltw:2, <CAKudYbNyqNdXmp2LjpFoXUSAcYKcfwLoaLJi+DBBVj2x5m=uZA at mail dot gmail dot com>