1 Mar
2011
1 Mar
'11
11:38 p.m.
On 03/01/2011 10:41 PM, Kim Cascone wrote:
Hello Kim,
PulseAudio is not a clusterthingy and after having re-read your article
I don't really understand your criticism on those convoluted layers upon
layers. You say it yourself, one only needs ALSA/JACK (or FFADO/JACK if
you're dealing with FireWire devices).
I wonder how
we could evangelize Linux Audio more effectively to the
outside world?
- Mark
Mark
I address some of the issues still plaguing Linux Audio in an article I
wrote almost two years ago,
http://ow.ly/1s687Z
many of these issues, such as the clusterf*ck that is Pulse Audio as
well as the convoluted layers upon layers of Linux audio, have either
not been resolved or just gotten worse
and as I see it prevent many from adopting Linux as a
serious audio
production platform - unless you are willing to pop the hood, roll up
your sleeves and get your hands dirty
Imho I think you can add conservatism, sheer ignorance or simple denial
that stuff can be done with Linux ;)
mind you this is if one is NOT using one of the better
distros such as
AVLinux or PureDyne (both of which works pretty much out of the box) and
are attempting to multi-purpose their environment - by sharing
administrative and music making tasks with a distro like Ubuntu
I have no issues trying to attempt to multi-purpose my environment. I
log out of my day-to-day account and log in to my music account and I'm
done. I could run a script to set up my day-to-day account for music
making purposes also but I prefer keeping them separate.
but other than the complexity of disentangling various
messes that arise
with xruns, tweaking conf files and testing various kernels I'm very
pleased with my current setup -- now if I could only get my xruns under
control! ;)
Personally I barely test kernels. Basically all real-time/lowlatency
kernels currently available for Ubuntu are no good imho so I've decided
to install one from someone who knows what he's doing (Jof from Tango
Studio) and indeed, works like a charm on all my set-ups (except for the
nouveau regression in the latest RT patch set). As for tweaking conf
files, at the moment there are very few files you might have to tweak,
/etc/security/limits.d/audio.conf (or /etc/security/limits.conf) for the
necessary permissions, maybe /etc/systcl.conf for some performance gain
and /etc/fstab to add noatime to the necessary mounts. If you want to
use a FireWire device it might take some more tweaking and when you're
audio related IRQ's are crowded you might have to use the rtirq script.
Enough rambling :)
Best,
Jeremy