27 Jan
2010
27 Jan
'10
4:20 p.m.
On 01/27/2010 09:11 PM, Stéphane Letz wrote:
Le 27 janv. 2010 à 21:02, Jörn Nettingsmeier a écrit :
On 01/27/2010 08:39 PM, alex stone wrote:
Well "jack2's apparent overhead," is something new for me, and would
require some deeper test/feedback to understand better. Moreover without more precise
description of xruns occurrence (at what DSP CPU does it start to happen.. etc..) , what
kind of setup (jack2 version, jack configuration, applications used....), it is again hard
to understand/correct things.
Stéphane
It's been a good day, and i've enjoyed
the stability. I used jack1
before, because it gave me fewer challenges, xruns and occasional pops
and spits, than jack2, which i ended up having to ease out to
48000/512/3, to reach near the same performance. (i'm on 48000/256/3
with jack1)
for some more anecdotal evidence, this is about what i was seeing during
my jack1/2 comparison tests. jack2 generally required one buffer size
step more to achieve the same xrun robustness as jack1. but i'm
generally able to use much lower latencies down to 64 (or 128 in the
jack2 case), unless i use jconvolver, which forces me to go to at least
1024 so as not to max out the cpu.
i didn't think too much about jack2's apparent overhead, since it has
the benefit of scaling to smp, which usually affects the
single-processor case (my box is a single-core amd64).
it would be interesting to see if torben's approach is able to deliver
the same latencies as jack1, while adding smp support.
regards,
jörn