Re: [LAU] jack2 vs. jack1/zita vs. jack2+zita-a2j, performance differences

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
To: Jonathan E. Brickman <jeb@...>
Cc: linux-audio-user <linux-audio-user@...>
Date: Thursday, April 17, 2014 - 6:37 am

--f46d043be06caedc6e04f7374265
Content-Type: text/plain; charset=ISO-8859-1

On Wed, Apr 16, 2014 at 9:53 PM, Jonathan E. Brickman
wrote:

> Thought the below might be of interest to some. The last is the best :-)

Why are you using schedtool to run jackd ? In addition, I don't understand
what this test is supposed to do. Why would you run jackd on the dummy
device while using the zita bridge(s) to access an actual device? There is
also no need to specify -R to jackd since it is the default.

Where would you expect sound to come from? What is connected to the ports
represented by the zita bridges?

DSP load variance between1.3% and 29% under static conditions indicate
problems with the setup. The value will never be precisely static but it
should not vary by this much.

Running 3 yoshimi instances at once is specifically a good test for the
multi-processor abilities of jack2, since it can run them all in parallel
(which jack1 does not do). It is one of the scenarios in which jack2's
capabilities in this area are a real benefit.

I really don't understand the goal here. Why not just run jack -d alsa -d
hw:SB -r 48000 -p 32 ?

--f46d043be06caedc6e04f7374265
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

On Wed, Apr 16, 2014 at 9:53 PM, Jonathan E. Brickman <jeb@pond=
erworthy.com
> wrote:
Thought the below might be of interest=
to some. =A0The last is the best :-) =A0Hardware and clients identical, oc=
to 4GHz, test load is one yoshimi with jack_keyboard driving by way of a pa=
ssthrough mididings.
-------------------------------Item two is jack1-git, compiled with Zita libraries engaged, run like thi=
s:nohup schedtool -R -p 50 -e /usr/bin/jackd -A S=
B -R -c h -X alsa_midi -d dummy -r 48000 -p 32 &
Reported latency is 2ms. =A0Load-less, usage rating is =
1.3% through 29%, usually hanging in at 1.4% or so. =A0With the test load, =
it sits at 31.2% at silence. =A0No xruns. =A0But no actual sound came out :=
-)
Why are you using schedtool to=
run jackd ? In addition, I don't understand what this test is supposed=
to do. Why would you run jackd on the dummy device while using the zita br=
idge(s) to access an actual device? There is also no need to specify -R to =
jackd since it is the default.
Where would you expect sound to come from? What is connected=
to the ports represented by the zita bridges?=A0<=
div>DSP load variance between1.3% and 29% under static conditions indicate =
problems with the setup. The value will never be precisely static but it sh=
ould not vary by this much.
Running 3 yoshimi instances at once is specifical=
ly a good test for the multi-processor abilities of jack2, since it can run=
them all in parallel (which jack1 does not do). It is one of the scenarios=
in which jack2's capabilities in this area are a real benefit.
I really don't understand the goal here. Why not just ru=
n jack -d alsa -d hw:SB -r 48000 -p 32 ?

--f46d043be06caedc6e04f7374265--

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
[LAU] jack2 vs. jack1/zita vs. jack2+zita-a2j, performance d..., Jonathan E. Brickman, (Thu Apr 17, 1:54 am)
Re: [LAU] jack2 vs. jack1/zita vs. jack2+zita-a2j, performan..., Paul Davis, (Thu Apr 17, 6:37 am)
Re: [LAU] jack2 vs. jack1/zita vs. jack2+zita-a2j, performan..., Jonathan E. Brickman, (Thu Apr 17, 12:36 pm)