25 Aug
2017
25 Aug
'17
9:05 a.m.
On Fri, 25 Aug 2017 10:15:26 +0200, Ralf Mattes wrote:
Am Freitag, 25. August 2017 08:25 CEST, John Rigg <j(a)jrigg.co.uk>
schrieb:
To idiotic high values, that long that playing out the buffer could be
considered to be annoying?
This is the context you seem to ignore:
On Thu, 24 Aug 2017 17:35:33 -0500, Chris Caudle wrote:
On 08/24/17 22:30, Ralf Mardorf wrote:
And, just to note: there are valid use cases for running jack without
low latency demands. As a matter of fact, unless you do realtime
(i.e. on-stage) audio processing/manipulation there's no need for
low latency. Actually, it' s a pretty good idea to increase buffer
sizes (read: high latency) when doing recordings. FreeBSD is not real-time capable, you need very
long latency.
FreeBSD has a preemptive scheduler and facilities for allowing
realtime privlleges for processes. Are you sure you're not
confusing it with one of the other *BSDs? I think the original message said that jackd kept
playing old audio.
If it was just jack playing out the buffers with old data I would
expect at most maybe 30 ms of audio (e.g. 512 samples per period, 3
periods, 48k samples/second) which would not really be recognizable
audio.
Wouldn't it be around 15 ms, since 30 ms would be round-trip latency?
To get something audible of e.g. annoying long 128 ms you need 4096
frames, 3 periods, 48k sample rate. 128 ms are 1/16 note at 117 BPM.
And assuming I should be mistaken, than it would be 256 ms, 1/8 note at
117 BPM.
The OP does use that long latency, if playing out the buffer should be
that long audible, or the issue is not related to playing out the
buffer.
Regards,
Ralf
PS:
On Fri, 25 Aug 2017 06:25:03 +0000, John Rigg wrote:
FreeBSD has a preemptive scheduler and facilities for
allowing
realtime privlleges for processes. Are you sure you're not
confusing it with one of the other *BSDs?
I installed FreeBSD because it has got a driver for the RME HDSPe AIO.