[PD-ot] how low (latency) can you go?
Hans-Christoph Steiner
hans at eds.org
Mon Dec 18 03:07:03 CET 2006
On Dec 16, 2006, at 1:14 PM, Bryan Jurish wrote:
>
> On 2006-12-16 19:02:08, Hans-Christoph Steiner <hans at eds.org>
> appears to have written:
>> And always, when talking about latency, I feel the need to point
>> out the speed of sound: 340 m/s or .34 m/ms. So if your speakers
>> are 2m away from you, that's 6ms of latency. You could spend days
>> tweaking your machine to get 3ms less latency, or you could move
>> 1m closer to your speakers. Puts things into perspective...
>
> indeed 'tis true. but 5ms vs. 10ms makes a major subjective/
> perceptual difference if we're talking about playing an electro-
> acoustic instrument in and getting a munged signal out...
>
Now some promised references:
---------------------------------------------
"As already noted in 3.1, the precedence effect says that sounds are
perceived as coming from a single source when they occur within 25-35
ms time window (Haas, 1949). Later research has suggested that we can
subconsciously perceive latencies below 10 ms, but in musical
instrument performance context latency tolerance can be even higher
than 50 ms. Chew et al. (2004) have studied duo performances, and in
four-hand piano experiment they have found that delays below 50 ms
were generally considered tolerable, and that when local delay was
introduced, the tolerance increased to 65 ms.
They also concluded that latency tolerance is dependant on tempo
(slower tempo equals higher tolerance), and on instrumentation. For
example, if a synthesizer pad type sound is used in ensemble
performance, the musician has to anticipate chord changes so that
slow attack is matched with the song tempo (corresponding latencies
of several hundred milliseconds), whereas transient sounds result
lower latency tolerance levels. Musical genre affects also tolerance,
because it often presets tempo and instrumentation choices, and it is
clear that a loop-based tightly quantized machine groove is less
tolerant than a rubato style ballad. Individual playing styles do
have an impact as well."
http://www.tml.tkk.fi/Opinnot/T-111.5080/2006/Paperit/
Latency_Kleimola.pdf
---------------------------------------------
"[T]he perception of the latency between an user action and the
corresponding reaction... shows a very high degree of precision: it
was shown that variations in feedback delay of 20ms are, although not
consciously noticed, compensated for..."
This article has some good info on jitter too.
http://gsd.ime.usp.br/~lago/masters/latency-paper.pdf
---------------------------------------------
"Since sound travels at about 1 ms per foot, latency of 7 ms is
roughly equal to the maximum separation between members of a string
quartet. In practice, latency of 10 ms is generally imperceptible,
as long as the variation in latency (i.e. jitter) is kept small."
http://www.midi.org/about-midi/tutorial/tutor.shtml
---------------------------------------------
For realtime games (e.g. FPS, RTS), latencies greater than 50-75ms
are noticeable. Upper threshold of manageable latency is 100-200ms.
(paraphrase)
http://www.ibr.cs.tu-bs.de/users/wellnitz/papers/netgames2005/
netgames2005-talk.pdf?backurl=%2Fcm%2Fbib.html&lang=de
http://www.ibr.cs.tu-bs.de/users/wellnitz/papers/netgames2005/
netgames2005.pdf?backurl=%2Fcm%2Fbib.html&lang=de
---------------------------------------------
So, my whole point with all this is to say that you really don't need
to worry so much about latency to have a very playable computer
instrument. So if you want to make art, spend your time making art
instead of getting a few less milliseconds of latency in your setup.
If you what to improve the technology, then please publish what you
did (blog it, put it on a website, article, magazine, textbook...),
and write code and get it out there.
.hc
------------------------------------------------------------------------
"[W]e have invented the technology to eliminate scarcity, but we are
deliberately throwing it away to benefit those who profit from
scarcity." -John Gilmore
More information about the PD-ot
mailing list