Testing audio latency of modern operating systems

Archived from groups: rec.audio.pro (More info?)

In article <gX%Od.27102$QG6.470468@twister2.libero.it> ODIOLOSPAM@nospamnospam.org writes:

> I would like to test the audio latency value for the various operating
> systems trying to modify also their hardware configuration (i.e. CPU,
> memory, sound card...).

I do this with a dual channel oscilloscope - a real one, not a PC
program. I connect a generator to the input and one channel of the
scope, and the output to the other channel of the scope. I put in a
pulse waveform and look at the time difference between when it goes in
and comes out.

Simple and realistic.


--
I'm really Mike Rivers (mrivers@d-and-d.com)
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo

Archived from groups: rec.audio.pro (More info?)

In article <cuiod4$ief$1@panix2.panix.com> kludge@panix.com writes:

> I find if you are looking for a really accurate measurement, though,
> lissajous method is best. Adjust the pulse repetition rate until the
> two outputs are 180' out of phase, then use the pulse rate to calculate
> the latency. Much more precise than the ruler-on-the-grid method.

Pretty clever, but when dealing with latencies of 10 milliseconds or
so, you don't need to get any more accurate than what can be read on
the screen. And if it's less than .5 ms, I (and most others) don't
usually care about it.

But "no latency" monitoring usually isn't, unless there's a direct
analog connection.




--
I'm really Mike Rivers (mrivers@d-and-d.com)
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo