Minimum reasonable latency Was: Re: ZynAddSubFX was: Re: [linux-audio-dev] some new soundfiles on-line

Fons Adriaensen fons.adriaensen at
Sat May 14 10:57:44 UTC 2005

On Sat, May 14, 2005 at 11:58:45AM +0200, Benno Senoner wrote:

> Since we cannot increase the speed at which the sound travels and even 
> DACs add some latency (1msec or so)
> I see any effort to reduce latency below 2-3msec quite useless.
> We know that halving the number of frames (soundcard period size) we run 
> the audio software at is doubling the IRQ rate
> and you can quickly achieve a point where a significative chunk of CPU 
> is burnt in the IRQ handlers.
> It could change if CPU get considerably faster but then you begin to hit 
> other limits, eg
> the PCI DMA minimum transfer sizes which limits your minimum soundcard 
> period size.
> Plus what happens if your speakers are 1m away from the listener ? 
> Another 3msecs added to the total
> "latency between hitting a key and hearing the sound".
> If you move the speakers another 50cm away from the ear ? Another 
> 1.5msecs delay get added.

This message came in while I was playing my new piano samples with
Linuxsampler (thanks Benno !). Jackd set to 2 * 256, that's 5.3ms,
about 0.6ms of DAC delay, let's say 1.0ms for midi and 3ms for speaker
to ear distance. That adds up to 9.9ms. Now the piano _is_ a percussive
instrument, the things I was playing were fast enough for my fingers,
and yet I had _no problem at all_ with this delay.

Now I've had some training as an organist, and then you learn to adapt
to much longer latencies - 30ms is quite normal. But I don't think that
is the determining factor.
> Humans are more sensitive to differences than to absolute values.

Right. And every musician (hopefully) learns sooner or later to play
along with others, and that also involves latencies. You just learn
to anticipate. 

> Eg in physics you feel the acceleration but not the speed.

That's because you feel force, and we have F = m * a.

> Of course some instrument players, especially those that play percussive 
> sounds are more sensitive than
> others, but the question is how small is this time ? Is there scientific 
> data around that can state:
> "if latency of the virtual percussion instrument is higher then X then 
> the instrument is unusable for 30% of professional musicians" ?

Not scientific, but I'd expect the most sensitive players to be those
that play acoustic string instruments. They are very near to the sound
source, and there's virtually nothing between them and the strings.


More information about the Linux-audio-dev mailing list