On Sat, 08 Jul 2000, Khimenko Victor wrote:
> In <[EMAIL PROTECTED]> Juhana Sadeharju
>([EMAIL PROTECTED]) wrote:
[...]
> > Lets not talk about 4 ms latency but 7 ms from practical point.
>
> Why ?
I agree with Juhana; the interesting figure is 7 ms. The reason for
this is that if 7 ms happens often enough to be registered within the
period of a normal benchmark/test, then it's happening way too often!
7 ms has to be regarded as the practical worst case latency, and it's
too much even if it was guaranteed never to occur twice within say, 5
seconds. (Allowing RT thread to catch up so that another 7 ms peak
doesn't generate a drop-out.) Once in a few weeks may be acceptable.
Once in 10 hrs, and we don't have a seriously usable solution.
//David
.- M u C o S --------------------------------. .- David Olofson ------.
| A Free/Open Multimedia | | Audio Hacker |
| Plugin and Integration Standard | | Linux Advocate |
`------------> http://www.linuxdj.com/mucos -' | Open Source Advocate |
.- A u d i a l i t y ------------------------. | Singer |
| Rock Solid Low Latency Signal Processing | | Songwriter |
`---> http://www.angelfire.com/or/audiality -' `-> [EMAIL PROTECTED] -'
