FAQ Search Today's Posts Mark Forums Read
» Video Reviews

» Linux Archive

Linux-archive is a website aiming to archive linux email lists and to make them easily accessible for linux users/developers.


» Sponsor

» Partners

» Sponsor

Go Back   Linux Archive > 64 Studio > 64 Studio User

 
 
LinkBack Thread Tools
 
Old 07-13-2010, 12:00 PM
Ralf Mardorf
 
Default 3.0b and 3.3a amd64 MIDI latency test for several kernels

Hi

here's the ALSA MIDI latency test for 64 Studio 3.0 amd64 and 64 Studio
3.3 amd64.

An hw MIDI instrument's audio outputs recorded to audio tracks test will
follow.

What I call 'PCI card 1' might be PCI card 2 for the computer.
S/PDIF out from the card that I call #1, is connected to S/PDIF in of
the card I call #2, while master clock isn't set to S/PDIF in.
The cards still are 2 separated sound cards, but one virtual card.

################################################## ######################
64 Studio 3.0-beta3, kernel 2.6.29-1-multimedia-amd64
################################################## ######################

GNOME

$ sudo poff dsl-provider
$ cpufreq-selector -g performance
$ cat /etc/X11/xorg.conf | grep Driver
Driver "nv"

$ uname -a
Linux 64studio 2.6.29-1-multimedia-amd64 #1 SMP PREEMPT RT Thu Feb 26
16:18:05 UTC 2009 x86_64 GNU/Linux

$ alsa-midi-latency-test -l
Port Client name Port name
14:0 Midi Through Midi Through Port-0
16:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI
20:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI

PCI card 1, test 1

alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
9999; 1.01; 1.05
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.05 ms, which is great.

PCI card 1, test 2

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
5635; 1.05; 1.05
9999; 1.00; 1.05
> latency distribution:
1.0 - 1.1 ms: 9997
##################################################
1.1 - 1.2 ms: 3 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.05 ms, which is great.

PCI card 2, test 1

$ alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
9999; 1.00; 1.05
> latency distribution:
1.0 - 1.1 ms: 9998
##################################################
1.1 - 1.2 ms: 2 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.05 ms, which is great.

PCI card 2, test 2

$ alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
9999; 1.01; 1.05
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.05 ms, which is great.

################################################## ######################
64 Studio 3.0-beta3, kernel 2.6.31.5-rt18 - Trial 1
################################################## ######################

It's not the first time I got a timeout, but the first time for today
and the the first time ever that the system was frozen.

PCI card 1, test 1 - timeout

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.06; 1.06
9700; 1.01; timeout: there seems to be no connection between
ports 20:0 and 20:0

I had to turn of the computer and obviously the power connector for one
SATA HDD loosed.
Anyway, the timeout seems to be caused by another issue, because it
appeared for the other card during the Studio 3.3 alpha test too.

################################################## ######################
64 Studio 3.0-beta3, kernel 2.6.31.5-rt18 - Trial 2
################################################## ######################

GNOME

$ sudo poff dsl-provider
$ cpufreq-selector -g performance
$ cat /etc/X11/xorg.conf | grep Driver
Driver "nv"

$ uname -a
Linux 64studio 2.6.31.5-rt18 #1 SMP PREEMPT RT Sun Nov 8 23:44:41 CET
2009 x86_64 GNU/Linux

$ alsa-midi-latency-test -l
Port Client name Port name
14:0 Midi Through Midi Through Port-0
16:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI
20:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI

PCI card 1, test 1

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.06; 1.06
9999; 1.01; 1.06
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.06 ms, which is great.

PCI card 1, test 2

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.06; 1.06
9999; 1.01; 1.06
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.06 ms, which is great.

PCI card 2, test 1

$ alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.04; 1.04
9999; 1.00; 1.04
> latency distribution:
1.0 - 1.1 ms: 10000
##################################################
> SUCCESS
best latency was 0.99 ms
worst latency was 1.04 ms, which is great.

PCI card 2, test 2

$ alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.04; 1.04
6247; 1.07; 1.07
9999; 1.00; 1.07
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.07 ms, which is great.

################################################## ######################
64 Studio 3.0-beta3, kernel 2.6.31.6-rt19
################################################## ######################

NOTE, I've got this kernel version for openSUSE too.
I did a comparison, also for USB MIDI with and without hrtimer.

GNOME

$ sudo poff dsl-provider
$ cpufreq-selector -g performance
$ cat /etc/X11/xorg.conf | grep Driver
Driver "nv"

$ uname -a
Linux 64studio 2.6.31.6-rt19 #1 SMP PREEMPT RT Fri Dec 25 22:37:59 CET
2009 x86_64 GNU/Linux

$ alsa-midi-latency-test -l
Port Client name Port name
14:0 Midi Through Midi Through Port-0
16:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI
20:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI

PCI card 1, test 1

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
9999; 1.01; 1.05
> latency distribution:
1.0 - 1.1 ms: 10000
##################################################
> SUCCESS
best latency was 0.99 ms
worst latency was 1.05 ms, which is great.

PCI card 1, test 2

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.06; 1.06
9999; 1.01; 1.06
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 1.00 ms
worst latency was 1.06 ms, which is great.

PCI card 2, test 1

$ alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.04; 1.04
9999; 1.01; 1.04
> latency distribution:
1.0 - 1.1 ms: 10000
##################################################
> SUCCESS
best latency was 0.99 ms
worst latency was 1.04 ms, which is great.

PCI card 2, test 2

$ alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.06; 1.06
9999; 1.01; 1.06
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.06 ms, which is great.

##########################################
USB
##########################################

$ hwinfo --usb
01: USB 00.0: 0000 Unclassified device
SysFS BusID: 3-1:1.0
Speed: 12 Mbps
Attached to: #5 (Hub)

$ hwinfo --usb
01: USB 00.0: 0000 Unclassified device
SysFS BusID: 1-1:1.0
Speed: 12 Mbps
Attached to: #7 (Hub)

I tried to find the first USB slot. I've got no idea what will be the
the first one.
I did two connections and kept 'SysFS BusID: 1-1:1.0' for the test.

$ alsa-midi-latency-test -l
Port Client name Port name
14:0 Midi Through Midi Through Port-0
16:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI
20:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI
24:0 USB Device 0x170b:0x11 USB Device 0x170b:0x11 MIDI 1

USB MIDI, test 1

$ alsa-midi-latency-test -Rrw=5 -i24:0 -o24:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.70; 1.70
1; 1.95; 1.95
2; 1.98; 1.98
3; 1.99; 1.99
5; 2.00; 2.00
21; 2.00; 2.00
61; 2.00; 2.00
102; 2.00; 2.00
223; 2.01; 2.01
251; 2.03; 2.03
9999; 2.00; 2.03
> latency distribution:
1.7 - 1.8 ms: 1 #
...
1.9 - 2.0 ms: 1 #
2.0 - 2.1 ms: 9998
##################################################
> SUCCESS
best latency was 1.70 ms
worst latency was 2.03 ms, which is great.

USB MIDI, test 2

$ sudo chgrp audio /dev/hpet
$ sudo sysctl -w dev.hpet.max-user-freq=64
$ sudo modprobe snd-hrtimer

$ alsa-midi-latency-test -Rrw=5 -i24:0 -o24:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.57; 1.57
1; 1.96; 1.96
2; 1.98; 1.98
3; 1.99; 1.99
4; 1.99; 1.99
5; 1.99; 1.99
9; 2.00; 2.00
12; 2.00; 2.00
15; 2.01; 2.01
3763; 2.02; 2.02
6763; 2.02; 2.02
9263; 2.03; 2.03
9999; 2.00; 2.03
> latency distribution:
1.6 - 1.7 ms: 1 #
...
1.9 - 2.0 ms: 3 #
2.0 - 2.1 ms: 9996
##################################################
> SUCCESS
best latency was 1.57 ms
worst latency was 2.03 ms, which is great.

I run the USB test with '-R -r -w=20' too, but the result didn't vary
significantly.

Clock resolution for the test seems to be the same as without hrtimer.
When making music there's less jitter when using hrtimer, but the USB
device without or with hrtimer has to much jitter to make music.
I can't confirm that the USB device is great. For my needs it's
unusable.
For my Swissonic USB MIDI the results seem to be better, the less
latency there is for JACK. The integrated ATI Radeon X1250-based
graphics seems to be better, when using USB MIDI, but the NVIDIA GeForce
7200GS PCX, anyway, here USB MIDI is (was?)unusable.
FWIW I had a PATA HDD that is broken now and replaced by a second SATA
HDD, I dunno if it was unnoticeable broken the years before and might
have had impact to the issue.
Watching the recorded waveforms, there was much more jitter for the
recorded audio from the MIDI instruments, but the alsa-midi-latency-test
results. I never tested PCI MIDI on this machine by making music.

################################################## ######################
64 Studio 3.0-beta3, kernel 2.6.31.12-rt20
################################################## ######################

GNOME

$ sudo poff dsl-provider
$ cpufreq-selector -g performance
$ cat /etc/X11/xorg.conf | grep Driver
Driver "nv"

$ uname -a
Linux 64studio 2.6.31.12-rt20 #1 SMP PREEMPT RT Sun Feb 7 13:55:55 CET
2010 x86_64 GNU/Linux

$ alsa-midi-latency-test -l
Port Client name Port name
14:0 Midi Through Midi Through Port-0
16:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI
20:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI

PCI card 1, test 1

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
9999; 1.01; 1.05
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.05 ms, which is great.

PCI card 1, test 2

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.07; 1.07
9999; 1.00; 1.07
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.07 ms, which is great.

PCI card 2, test 1

$ alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
9999; 1.01; 1.05
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.05 ms, which is great.

PCI card 2, test 2

$ alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.06; 1.06
9999; 1.00; 1.06
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.06 ms, which is great.

################################################## ######################
64 Studio 3.3 alpha, kernel 2.6.31-2-multimedia-amd64
################################################## ######################

LXDE

$ su -c "poff dsl-provider"
$ su -c "cpufreq-selector -g performance"
^C
I guess it's set to performance by default, don't know how to control
it, because even hwinfo for some distros doesn't show the actual
frequency scaling all the time.
$ cat /etc/X11/xorg.conf | grep Driver
Driver "nvidia"

$ uname -a
Linux 64studio 2.6.31-2-multimedia-amd64 #1 SMP PREEMPT RT Thu Oct 1
16:14:20 BST 2009 x86_64 GNU/Linux

$ alsa-midi-latency-test -l
Port Client name Port name
14:0 Midi Through Midi Through Port-0
16:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI
20:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI

PCI card 1, test 1

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
2205; 1.08; 1.08
9999; 1.01; 1.08
> latency distribution:
1.0 - 1.1 ms: 9994
##################################################
1.1 - 1.2 ms: 6 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.08 ms, which is great.

PCI card 1, test 2

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.06; 1.06
4061; 1.07; 1.07
6394; 1.timeout: there seems to be no connection between ports
20:0 and 20:0

Aaaaaaaaaaargh! I even didn't touch the MIDI adaptor cable connections,
but just started a test again.
The system didn't freeze and I didn't touch any jack after the timeout,
but anyway it passed.

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
1134; 1.05; 1.05
6240; 1.06; 1.06
9999; 1.00; 1.06
> latency distribution:
1.0 - 1.1 ms: 9995
##################################################
1.1 - 1.2 ms: 5 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.06 ms, which is great.

PCI card 2, test 1

$ alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
6232; 1.06; 1.06
9999; 1.01; 1.06
> latency distribution:
1.0 - 1.1 ms: 9998
##################################################
1.1 - 1.2 ms: 2 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.06 ms, which is great.

PCI card 2, test 2

spinymouse@64studio:~$ alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.07; 1.07
9999; 1.01; 1.07
> latency distribution:
1.0 - 1.1 ms: 9991
##################################################
1.1 - 1.2 ms: 9 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.07 ms, which is great.

################################################## ######################
openSUSE 11.2, kernel 2 2.6.31.6-rt19
################################################## ######################

GNOME

I've got no idea how to disconnect the Internet access.
Even if I set up CPU freq scaling to performance by default:
> su -c"cpufreq-set -gperformance"
> cat /etc/X11/xorg.conf | grep Driver
Driver "nvidia"

> uname -a
Linux suse11-2 2.6.31.6-rt19 #1 SMP PREEMPT RT Wed Nov 18 16:59:26 CET
2009 x86_64 x86_64 x86_64 GNU/Linux

> alsa-midi-latency-test -l
Port Client name Port name
14:0 Midi Through Midi Through Port-0
16:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI
24:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI

PCI card 1, test 1

> alsa-midi-latency-test -Rrw=5 -i24:0 -o24:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.06; 1.06
9999; 1.00; 1.06
> latency distribution:
1.0 - 1.1 ms: 9996
##################################################
1.1 - 1.2 ms: 4 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.06 ms, which is great.

PCI card 1, test 2

> alsa-midi-latency-test -Rrw=5 -i24:0 -o24:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
92; 1.05; 1.05
9999; 1.00; 1.05
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.05 ms, which is great.

PCI card 2, test 1

> alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
85; 1.05; 1.05
694; 1.08; 1.08
4658; 1.08; 1.08
9999; 1.00; 1.08
> latency distribution:
1.0 - 1.1 ms: 9962
##################################################
1.1 - 1.2 ms: 38 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.08 ms, which is great.

PCI card 2, test 2

> alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
41; 1.06; 1.06
2306; 1.07; 1.07
9999; 1.00; 1.07
> latency distribution:
1.0 - 1.1 ms: 9966
##################################################
1.1 - 1.2 ms: 34 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.07 ms, which is great.

##########################################
USB
##########################################

I connected the USB MIDI to the same USB slot, as for the 64 Studio
test, but the output of hwinfo differed.

> su -c "hwinfo --usb"
10: USB 00.0: 0401 Multimedia audio controller
SysFS BusID: 2-1:1.0
Speed: 12 Mbps
Attached to: #5 (Hub)

> alsa-midi-latency-test -l
Port Client name Port name
14:0 Midi Through Midi Through Port-0
16:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI
20:0 USB Device 0x170b:0x11 USB Device 0x170b:0x11 MIDI 1
24:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI

USB MIDI, test 1

> alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.79; 1.79
1; 1.96; 1.96
2; 1.99; 1.99
4; 2.00; 2.00
9; 2.00; 2.00
12; 2.02; 2.02
54; 2.03; 2.03
321; 2.03; 2.03
410; 2.04; 2.04
886; 2.04; 2.04
1695; 2.05; 2.05
3158; 2.05; 2.05
9999; 1.99; 2.05
> latency distribution:
1.8 - 1.9 ms: 1 #
1.9 - 2.0 ms: 4 #
2.0 - 2.1 ms: 9993
##################################################
2.1 - 2.2 ms: 2 #
> SUCCESS
best latency was 1.79 ms
worst latency was 2.05 ms, which is great.

USB MIDI, test 2

> su -c "chgrp audio /dev/hpet"
> su -c "sysctl -w dev.hpet.max-user-freq=64"
> su -c "modprobe snd-hrtimer"

> alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 2.06; 2.06
9999; 1.99; 2.06
> latency distribution:
1.9 - 2.0 ms: 5 #
2.0 - 2.1 ms: 9994
##################################################
2.1 - 2.2 ms: 1 #
> SUCCESS
best latency was 1.94 ms
worst latency was 2.06 ms, which is great.

Cheers!

Ralf

_______________________________________________
64studio-users mailing list
64studio-users@lists.64studio.com
http://lists.64studio.com/mailman/listinfo/64studio-users
 
Old 07-13-2010, 12:00 PM
Ralf Mardorf
 
Default 3.0b and 3.3a amd64 MIDI latency test for several kernels

Hi

here's the ALSA MIDI latency test for 64 Studio 3.0 amd64 and 64 Studio
3.3 amd64.

An hw MIDI instrument's audio outputs recorded to audio tracks test will
follow.

What I call 'PCI card 1' might be PCI card 2 for the computer.
S/PDIF out from the card that I call #1, is connected to S/PDIF in of
the card I call #2, while master clock isn't set to S/PDIF in.
The cards still are 2 separated sound cards, but one virtual card.

################################################## ######################
64 Studio 3.0-beta3, kernel 2.6.29-1-multimedia-amd64
################################################## ######################

GNOME

$ sudo poff dsl-provider
$ cpufreq-selector -g performance
$ cat /etc/X11/xorg.conf | grep Driver
Driver "nv"

$ uname -a
Linux 64studio 2.6.29-1-multimedia-amd64 #1 SMP PREEMPT RT Thu Feb 26
16:18:05 UTC 2009 x86_64 GNU/Linux

$ alsa-midi-latency-test -l
Port Client name Port name
14:0 Midi Through Midi Through Port-0
16:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI
20:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI

PCI card 1, test 1

alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
9999; 1.01; 1.05
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.05 ms, which is great.

PCI card 1, test 2

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
5635; 1.05; 1.05
9999; 1.00; 1.05
> latency distribution:
1.0 - 1.1 ms: 9997
##################################################
1.1 - 1.2 ms: 3 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.05 ms, which is great.

PCI card 2, test 1

$ alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
9999; 1.00; 1.05
> latency distribution:
1.0 - 1.1 ms: 9998
##################################################
1.1 - 1.2 ms: 2 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.05 ms, which is great.

PCI card 2, test 2

$ alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
9999; 1.01; 1.05
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.05 ms, which is great.

################################################## ######################
64 Studio 3.0-beta3, kernel 2.6.31.5-rt18 - Trial 1
################################################## ######################

It's not the first time I got a timeout, but the first time for today
and the the first time ever that the system was frozen.

PCI card 1, test 1 - timeout

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.06; 1.06
9700; 1.01; timeout: there seems to be no connection between
ports 20:0 and 20:0

I had to turn of the computer and obviously the power connector for one
SATA HDD loosed.
Anyway, the timeout seems to be caused by another issue, because it
appeared for the other card during the Studio 3.3 alpha test too.

################################################## ######################
64 Studio 3.0-beta3, kernel 2.6.31.5-rt18 - Trial 2
################################################## ######################

GNOME

$ sudo poff dsl-provider
$ cpufreq-selector -g performance
$ cat /etc/X11/xorg.conf | grep Driver
Driver "nv"

$ uname -a
Linux 64studio 2.6.31.5-rt18 #1 SMP PREEMPT RT Sun Nov 8 23:44:41 CET
2009 x86_64 GNU/Linux

$ alsa-midi-latency-test -l
Port Client name Port name
14:0 Midi Through Midi Through Port-0
16:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI
20:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI

PCI card 1, test 1

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.06; 1.06
9999; 1.01; 1.06
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.06 ms, which is great.

PCI card 1, test 2

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.06; 1.06
9999; 1.01; 1.06
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.06 ms, which is great.

PCI card 2, test 1

$ alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.04; 1.04
9999; 1.00; 1.04
> latency distribution:
1.0 - 1.1 ms: 10000
##################################################
> SUCCESS
best latency was 0.99 ms
worst latency was 1.04 ms, which is great.

PCI card 2, test 2

$ alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.04; 1.04
6247; 1.07; 1.07
9999; 1.00; 1.07
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.07 ms, which is great.

################################################## ######################
64 Studio 3.0-beta3, kernel 2.6.31.6-rt19
################################################## ######################

NOTE, I've got this kernel version for openSUSE too.
I did a comparison, also for USB MIDI with and without hrtimer.

GNOME

$ sudo poff dsl-provider
$ cpufreq-selector -g performance
$ cat /etc/X11/xorg.conf | grep Driver
Driver "nv"

$ uname -a
Linux 64studio 2.6.31.6-rt19 #1 SMP PREEMPT RT Fri Dec 25 22:37:59 CET
2009 x86_64 GNU/Linux

$ alsa-midi-latency-test -l
Port Client name Port name
14:0 Midi Through Midi Through Port-0
16:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI
20:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI

PCI card 1, test 1

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
9999; 1.01; 1.05
> latency distribution:
1.0 - 1.1 ms: 10000
##################################################
> SUCCESS
best latency was 0.99 ms
worst latency was 1.05 ms, which is great.

PCI card 1, test 2

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.06; 1.06
9999; 1.01; 1.06
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 1.00 ms
worst latency was 1.06 ms, which is great.

PCI card 2, test 1

$ alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.04; 1.04
9999; 1.01; 1.04
> latency distribution:
1.0 - 1.1 ms: 10000
##################################################
> SUCCESS
best latency was 0.99 ms
worst latency was 1.04 ms, which is great.

PCI card 2, test 2

$ alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.06; 1.06
9999; 1.01; 1.06
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.06 ms, which is great.

##########################################
USB
##########################################

$ hwinfo --usb
01: USB 00.0: 0000 Unclassified device
SysFS BusID: 3-1:1.0
Speed: 12 Mbps
Attached to: #5 (Hub)

$ hwinfo --usb
01: USB 00.0: 0000 Unclassified device
SysFS BusID: 1-1:1.0
Speed: 12 Mbps
Attached to: #7 (Hub)

I tried to find the first USB slot. I've got no idea what will be the
the first one.
I did two connections and kept 'SysFS BusID: 1-1:1.0' for the test.

$ alsa-midi-latency-test -l
Port Client name Port name
14:0 Midi Through Midi Through Port-0
16:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI
20:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI
24:0 USB Device 0x170b:0x11 USB Device 0x170b:0x11 MIDI 1

USB MIDI, test 1

$ alsa-midi-latency-test -Rrw=5 -i24:0 -o24:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.70; 1.70
1; 1.95; 1.95
2; 1.98; 1.98
3; 1.99; 1.99
5; 2.00; 2.00
21; 2.00; 2.00
61; 2.00; 2.00
102; 2.00; 2.00
223; 2.01; 2.01
251; 2.03; 2.03
9999; 2.00; 2.03
> latency distribution:
1.7 - 1.8 ms: 1 #
...
1.9 - 2.0 ms: 1 #
2.0 - 2.1 ms: 9998
##################################################
> SUCCESS
best latency was 1.70 ms
worst latency was 2.03 ms, which is great.

USB MIDI, test 2

$ sudo chgrp audio /dev/hpet
$ sudo sysctl -w dev.hpet.max-user-freq=64
$ sudo modprobe snd-hrtimer

$ alsa-midi-latency-test -Rrw=5 -i24:0 -o24:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.57; 1.57
1; 1.96; 1.96
2; 1.98; 1.98
3; 1.99; 1.99
4; 1.99; 1.99
5; 1.99; 1.99
9; 2.00; 2.00
12; 2.00; 2.00
15; 2.01; 2.01
3763; 2.02; 2.02
6763; 2.02; 2.02
9263; 2.03; 2.03
9999; 2.00; 2.03
> latency distribution:
1.6 - 1.7 ms: 1 #
...
1.9 - 2.0 ms: 3 #
2.0 - 2.1 ms: 9996
##################################################
> SUCCESS
best latency was 1.57 ms
worst latency was 2.03 ms, which is great.

I run the USB test with '-R -r -w=20' too, but the result didn't vary
significantly.

Clock resolution for the test seems to be the same as without hrtimer.
When making music there's less jitter when using hrtimer, but the USB
device without or with hrtimer has to much jitter to make music.
I can't confirm that the USB device is great. For my needs it's
unusable.
For my Swissonic USB MIDI the results seem to be better, the less
latency there is for JACK. The integrated ATI Radeon X1250-based
graphics seems to be better, when using USB MIDI, but the NVIDIA GeForce
7200GS PCX, anyway, here USB MIDI is (was?)unusable.
FWIW I had a PATA HDD that is broken now and replaced by a second SATA
HDD, I dunno if it was unnoticeable broken the years before and might
have had impact to the issue.
Watching the recorded waveforms, there was much more jitter for the
recorded audio from the MIDI instruments, but the alsa-midi-latency-test
results. I never tested PCI MIDI on this machine by making music.

################################################## ######################
64 Studio 3.0-beta3, kernel 2.6.31.12-rt20
################################################## ######################

GNOME

$ sudo poff dsl-provider
$ cpufreq-selector -g performance
$ cat /etc/X11/xorg.conf | grep Driver
Driver "nv"

$ uname -a
Linux 64studio 2.6.31.12-rt20 #1 SMP PREEMPT RT Sun Feb 7 13:55:55 CET
2010 x86_64 GNU/Linux

$ alsa-midi-latency-test -l
Port Client name Port name
14:0 Midi Through Midi Through Port-0
16:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI
20:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI

PCI card 1, test 1

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
9999; 1.01; 1.05
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.05 ms, which is great.

PCI card 1, test 2

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.07; 1.07
9999; 1.00; 1.07
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.07 ms, which is great.

PCI card 2, test 1

$ alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
9999; 1.01; 1.05
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.05 ms, which is great.

PCI card 2, test 2

$ alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.06; 1.06
9999; 1.00; 1.06
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.06 ms, which is great.

################################################## ######################
64 Studio 3.3 alpha, kernel 2.6.31-2-multimedia-amd64
################################################## ######################

LXDE

$ su -c "poff dsl-provider"
$ su -c "cpufreq-selector -g performance"
^C
I guess it's set to performance by default, don't know how to control
it, because even hwinfo for some distros doesn't show the actual
frequency scaling all the time.
$ cat /etc/X11/xorg.conf | grep Driver
Driver "nvidia"

$ uname -a
Linux 64studio 2.6.31-2-multimedia-amd64 #1 SMP PREEMPT RT Thu Oct 1
16:14:20 BST 2009 x86_64 GNU/Linux

$ alsa-midi-latency-test -l
Port Client name Port name
14:0 Midi Through Midi Through Port-0
16:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI
20:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI

PCI card 1, test 1

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
2205; 1.08; 1.08
9999; 1.01; 1.08
> latency distribution:
1.0 - 1.1 ms: 9994
##################################################
1.1 - 1.2 ms: 6 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.08 ms, which is great.

PCI card 1, test 2

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.06; 1.06
4061; 1.07; 1.07
6394; 1.timeout: there seems to be no connection between ports
20:0 and 20:0

Aaaaaaaaaaargh! I even didn't touch the MIDI adaptor cable connections,
but just started a test again.
The system didn't freeze and I didn't touch any jack after the timeout,
but anyway it passed.

$ alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
1134; 1.05; 1.05
6240; 1.06; 1.06
9999; 1.00; 1.06
> latency distribution:
1.0 - 1.1 ms: 9995
##################################################
1.1 - 1.2 ms: 5 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.06 ms, which is great.

PCI card 2, test 1

$ alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
6232; 1.06; 1.06
9999; 1.01; 1.06
> latency distribution:
1.0 - 1.1 ms: 9998
##################################################
1.1 - 1.2 ms: 2 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.06 ms, which is great.

PCI card 2, test 2

spinymouse@64studio:~$ alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.07; 1.07
9999; 1.01; 1.07
> latency distribution:
1.0 - 1.1 ms: 9991
##################################################
1.1 - 1.2 ms: 9 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.07 ms, which is great.

################################################## ######################
openSUSE 11.2, kernel 2 2.6.31.6-rt19
################################################## ######################

GNOME

I've got no idea how to disconnect the Internet access.
Even if I set up CPU freq scaling to performance by default:
> su -c"cpufreq-set -gperformance"
> cat /etc/X11/xorg.conf | grep Driver
Driver "nvidia"

> uname -a
Linux suse11-2 2.6.31.6-rt19 #1 SMP PREEMPT RT Wed Nov 18 16:59:26 CET
2009 x86_64 x86_64 x86_64 GNU/Linux

> alsa-midi-latency-test -l
Port Client name Port name
14:0 Midi Through Midi Through Port-0
16:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI
24:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI

PCI card 1, test 1

> alsa-midi-latency-test -Rrw=5 -i24:0 -o24:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.06; 1.06
9999; 1.00; 1.06
> latency distribution:
1.0 - 1.1 ms: 9996
##################################################
1.1 - 1.2 ms: 4 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.06 ms, which is great.

PCI card 1, test 2

> alsa-midi-latency-test -Rrw=5 -i24:0 -o24:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
92; 1.05; 1.05
9999; 1.00; 1.05
> latency distribution:
1.0 - 1.1 ms: 9999
##################################################
1.1 - 1.2 ms: 1 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.05 ms, which is great.

PCI card 2, test 1

> alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
85; 1.05; 1.05
694; 1.08; 1.08
4658; 1.08; 1.08
9999; 1.00; 1.08
> latency distribution:
1.0 - 1.1 ms: 9962
##################################################
1.1 - 1.2 ms: 38 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.08 ms, which is great.

PCI card 2, test 2

> alsa-midi-latency-test -Rrw=5 -i16:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
41; 1.06; 1.06
2306; 1.07; 1.07
9999; 1.00; 1.07
> latency distribution:
1.0 - 1.1 ms: 9966
##################################################
1.1 - 1.2 ms: 34 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.07 ms, which is great.

##########################################
USB
##########################################

I connected the USB MIDI to the same USB slot, as for the 64 Studio
test, but the output of hwinfo differed.

> su -c "hwinfo --usb"
10: USB 00.0: 0401 Multimedia audio controller
SysFS BusID: 2-1:1.0
Speed: 12 Mbps
Attached to: #5 (Hub)

> alsa-midi-latency-test -l
Port Client name Port name
14:0 Midi Through Midi Through Port-0
16:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI
20:0 USB Device 0x170b:0x11 USB Device 0x170b:0x11 MIDI 1
24:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI

USB MIDI, test 1

> alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 1.79; 1.79
1; 1.96; 1.96
2; 1.99; 1.99
4; 2.00; 2.00
9; 2.00; 2.00
12; 2.02; 2.02
54; 2.03; 2.03
321; 2.03; 2.03
410; 2.04; 2.04
886; 2.04; 2.04
1695; 2.05; 2.05
3158; 2.05; 2.05
9999; 1.99; 2.05
> latency distribution:
1.8 - 1.9 ms: 1 #
1.9 - 2.0 ms: 4 #
2.0 - 2.1 ms: 9993
##################################################
2.1 - 2.2 ms: 2 #
> SUCCESS
best latency was 1.79 ms
worst latency was 2.05 ms, which is great.

USB MIDI, test 2

> su -c "chgrp audio /dev/hpet"
> su -c "sysctl -w dev.hpet.max-user-freq=64"
> su -c "modprobe snd-hrtimer"

> alsa-midi-latency-test -Rrw=5 -i20:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
sample; latency_ms; latency_ms_worst
0; 2.06; 2.06
9999; 1.99; 2.06
> latency distribution:
1.9 - 2.0 ms: 5 #
2.0 - 2.1 ms: 9994
##################################################
2.1 - 2.2 ms: 1 #
> SUCCESS
best latency was 1.94 ms
worst latency was 2.06 ms, which is great.

Cheers!

Ralf

_______________________________________________
64studio-devel mailing list
64studio-devel@lists.64studio.com
http://lists.64studio.com/mailman/listinfo/64studio-devel
 

Thread Tools




All times are GMT. The time now is 02:07 PM.

VBulletin, Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO ©2007, Crawlability, Inc.
Copyright 2007 - 2008, www.linux-archive.org