[LAD] ALSA MIDI latency test results are far away from reality
Ralf Mardorf
ralf.mardorf at alice-dsl.net
Wed Jul 14 15:10:59 UTC 2010
On Wed, 2010-07-14 at 15:52 +0200, Clemens Ladisch wrote:
> Ralf Mardorf wrote:
> > 1.
> >
> > I disconnected all audio connections for JACK and connected hw MIDI in
> > to hw MIDI out.
>
> Is this connection through JACK or through ALSA, i.e., does it show up
> in the output of "aconnect -l"? From what I understand, JACK's sample-
> synchronous timing always adds latency, and might add period-related
> jitter depending on the implementation.
Qtractor at the moment is ALSA MIDI only and I even didn't use the -Xseq
bridge.
> > Yamaha DX7 --> Alesis D4 results in a 100% musical groove.
> > Yamaha DX7 --> PC --> Alesis D4 results in extreme latency,
>
> With a single MIDI loopback cable, the latency test program tests
> a "PC --> PC" connection. A more realistic test would be
> "PC1 --> PC2 --> PC1" (or just one PC if it has two inputs and two
> outputs).
I've got 1 USB MIDI and two PCI MIDI by 2 Terratec EWX 24/96 Envy24
based sound cards. I'll avoid to ever connect the USB MIDI to my machine
again (as I'll avoid to install ever a Windows to it again ;).
Next month I could buy another MIDI adaptor cable and run
alsa-midi-latency-test by using both PCI cards.
I bring myself to connect the USB MIDI one more time ;).
64 Studio 3.3 alpha
$ su -c "poff dsl-provider"
$ su -c "cpufreq-selector -g performance"
$ 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 USB Device 0x170b:0x11 USB Device 0x170b:0x11 MIDI 1
24:0 TerraTec EWX24/96 TerraTec EWX24/96 MIDI
########################################################################
### ONE OF THE PCI CARDS
########################################################################
$ 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
> sampling 10000 midi latency values - please wait ...
> press Ctrl+C to abort test
sample; latency_ms; latency_ms_worst
0; 1.05; 1.05
787; 1.06; 1.06
1227; 1.06; 1.06
8738; 1.07; 1.07
9999; 1.00; 1.07
> done.
> latency distribution:
...
1.0 - 1.1 ms: 9992
##################################################
1.1 - 1.2 ms: 8 #
> SUCCESS
best latency was 0.99 ms
worst latency was 1.07 ms, which is great.
########################################################################
### THE USB MIDI WITH AN UNLUCKY ENTRY FOR THE RTIRQ SCRIPT
########################################################################
Usually I keep Rui's entries for the script.
$ cat /etc/default/rtirq | grep RTIRQ
RTIRQ_NAME_LIST="rtc snd i8042"
RTIRQ_PRIO_HIGH=98
RTIRQ_PRIO_DECR=5
RTIRQ_RESET_ALL=0
RTIRQ_NON_THREADED="rtc snd"
# RTIRQ_HIGH_LIST="timer"
$ 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
> sampling 10000 midi latency values - please wait ...
> press Ctrl+C to abort test
sample; latency_ms; latency_ms_worst
0; 2.21; 2.21
9999; 1.99; 2.21
> done.
> latency distribution:
...
1.9 - 2.0 ms: 2 #
2.0 - 2.1 ms: 9997
##################################################
...
2.2 - 2.3 ms: 1 #
> SUCCESS
best latency was 1.95 ms
worst latency was 2.21 ms, which is great.
########################################################################
### PCI MIDI out --> USB MIDI in
########################################################################
$ alsa-midi-latency-test -Rrw=5 -i20:0 -o16:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
> sampling 10000 midi latency values - please wait ...
> press Ctrl+C to abort test
sample; latency_ms; latency_ms_worst
0; 2.03; 2.03
186; 2.04; 2.04
2726; 2.04; 2.04
9999; 1.99; 2.04
> done.
> latency distribution:
...
1.9 - 2.0 ms: 2 #
2.0 - 2.1 ms: 9998
##################################################
> SUCCESS
best latency was 1.92 ms
worst latency was 2.04 ms, which is great.
########################################################################
### USB MIDI out --> PCI MIDI in
########################################################################
$ alsa-midi-latency-test -Rrw=5 -i16:0 -o20:0
> alsa-midi-latency-test 0.0.3
> set_realtime_priority(SCHED_FIFO, 99).. done.
> clock resolution: 0.000000001 s
> sampling 10000 midi latency values - please wait ...
> press Ctrl+C to abort test
sample; latency_ms; latency_ms_worst
0; 1.14; 1.14
1; 1.93; 1.93
2; 1.99; 1.99
5; 2.00; 2.00
6; 2.00; 2.00
13; 2.01; 2.01
22; 2.03; 2.03
36; 2.03; 2.03
51; 2.03; 2.03
65; 2.03; 2.03
265; 2.04; 2.04
1160; 2.04; 2.04
2267; 2.07; 2.07
9999; 1.96; 2.07
> done.
> latency distribution:
...
1.1 - 1.2 ms: 1 #
...
1.9 - 2.0 ms: 31 #
2.0 - 2.1 ms: 9967
##################################################
2.1 - 2.2 ms: 1 #
> SUCCESS
best latency was 1.14 ms
worst latency was 2.07 ms, which is great.
$ cd /etc/init.d
$ ./rtirq status
PID CLS RTPRIO NI PRI %CPU STAT COMMAND
388 FF 95 - 135 0.0 S< irq/8-rtc0
1119 FF 90 - 130 0.0 S< irq/21-ICE1712
1115 FF 89 - 129 0.0 S< irq/20-ICE1712
381 FF 85 - 125 0.0 S< irq/1-i8042
380 FF 84 - 124 0.0 S< irq/12-i8042
102 FF 50 - 90 0.0 S< irq/9-acpi
539 FF 50 - 90 0.0 S< irq/16-ohci_hcd
588 FF 50 - 90 0.0 S< irq/19-ehci_hcd
590 FF 50 - 90 0.0 S< irq/22-ahci
601 FF 50 - 90 0.0 S< irq/14-ide0
603 FF 50 - 90 0.0 S< irq/17-ohci_hcd
604 FF 50 - 90 0.0 S< irq/22-ohci1394
608 FF 50 - 90 0.0 S< irq/18-ohci_hcd
611 FF 50 - 90 0.0 S< irq/17-ohci_hcd
614 FF 50 - 90 0.0 S< irq/18-ohci_hcd
943 FF 50 - 90 0.0 S< irq/7-parport0
1670 FF 50 - 90 0.2 S< irq/18-nvidia
1984 FF 50 - 90 0.0 S< irq/26-eth0
4 FF 49 - 89 0.0 S< sirq-high/0
5 FF 49 - 89 0.0 S< sirq-timer/0
6 FF 49 - 89 0.0 S< sirq-net-tx/0
7 FF 49 - 89 0.0 S< sirq-net-rx/0
8 FF 49 - 89 0.0 S< sirq-block/0
9 FF 49 - 89 0.0 S< sirq-tasklet/0
10 FF 49 - 89 0.0 S< sirq-sched/0
11 FF 49 - 89 0.0 S< sirq-hrtimer/0
12 FF 49 - 89 0.0 S< sirq-rcu/0
18 FF 49 - 89 0.0 S< sirq-high/1
19 FF 49 - 89 0.0 S< sirq-timer/1
20 FF 49 - 89 0.0 S< sirq-net-tx/1
21 FF 49 - 89 0.0 S< sirq-net-rx/1
22 FF 49 - 89 0.0 S< sirq-block/1
23 FF 49 - 89 0.2 S< sirq-tasklet/1
24 FF 49 - 89 0.0 S< sirq-sched/1
25 FF 49 - 89 0.0 S< sirq-hrtimer/1
26 FF 49 - 89 0.0 S< sirq-rcu/1
- Ralf
>
>
> Regards,
> Clemens
More information about the Linux-audio-dev
mailing list