[LAD] discussing HZ (kernel timer frequency) and midi timing

Florian Schmidt mista.tapas at gmx.net
Fri May 4 10:18:25 UTC 2007

On Friday 04 May 2007, Mulyadi Santosa wrote:
> Hi
> Sorry for late reply..
> > I have
> > dev.rtc.max-user-freq = 8192
> > in my sysctl.conf.
> > Before seq24 timing seemed to be changing between laid back and forward.
> > Haven't tested since setting.
> > Do
> > cat /proc/sys/dev/rtc/max-user-freq
> > to see your setting now.
> > echo 8192 > /proc/sys/dev/rtc/max-user-freq
> > changes setting.
> Thank you, even this small hint gave me a valuable direction. It makes
> me believe there are many applications that do rely on non-kernel-timer
> source.. in this case RTC.
> May I ask, could seq24 reports somekind of note playing delay? or xruns?

seq24 does _not_ use RTC for timing. It uses an approach based on sleep() (in 
some way or the other), so setting the rtc-max-frequency does nothing for 
seq24 performance. To improve seq24 performance over the vanilla kernel you 
need a kernel with the system timer frequency set back to something like 
1000Hz (vanilla has 200Hz, which is too unprecise for MIDI timing based on 
sleeping short amounts of time)..


Palimm Palimm!

More information about the Linux-audio-dev mailing list