Subject: Re: [linux-audio-dev] latencytest 0.42 with PNG support released
From: Benno Senoner (sbenno_AT_gardena.net)
Date: Wed Jul 26 2000 - 21:00:35 EEST
On Wed, 26 Jul 2000, David Olofson wrote:
> On Tue, 25 Jul 2000, Benno Senoner wrote:
> [...]
> > The 2.2.16-lowlatency RPM I posted today works great,
>
> I'm just downloading it. I'll also try the latest 2.4 stuff.
attention that Udo replaced the RPM today with one compiled with APM OFF
because the other caused up to 2.5msec periodic latency spikes which
could make my 2.1msec audio thread dropout.
David, can you run a latencytest session on the RPM to see if the green spikes
are absent. If yes then the problem may be 2.2.16 issue.
> >
> > and I tried my usual triggering of 20 voices (my MIDI keyboard
> > does not seem to be able to manage more than 20 active notes)
>
> What about sustain pedal support? ;-) (That's a controller, so you
> need to do it in the sampler.) I have 88 keys + pedal here - but
> I doubt that my 20 GB 7200 rpm drive can deal with it without relying
> on the chache. *hehe*
I have only a toy keyboard : simple note-on velocity, program change,
pitchbend , modulation , volume , no sustain pedal.
So I can't try. But I could emulate sustain through the modweel
(if modwheel >= 64 then sustain)
currently disksampler does not even support note releases
(it simply cuts off the sound at note-off), so the event system is the first
thing I have to implement.
Later sustain will be quite easily to do, since if sustain goes to on then
queue up all note-offs in a list, and if sustain goes to off , then process
all note offs.
>
> > I hoped at least that the disk stream dropped out
> > (256KB buffers = 3sec latencies per track), but no I was unable
> > to generate the "associated stream not ready" message.
>
> Looks great! :-)
It seems that our baby will work very well on a dedicated disk since linux
does not show up these high disk latencies when only reading data
(the writes are the problem), so we may achieve quite a good performance
on a dedicated disk. (that's an overall recommendation for all HD recording
software because the disk is not deterministic, so it's better to let only
one or few apps control it)
David: did you look at my proposed event managing sytem ?
(keeping the out LAD for now because we the talks are too specific)
Benno.
>
>
> //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 -' `-> david_AT_linuxdj.com -'
This archive was generated by hypermail 2b28 : Wed Jul 26 2000 - 20:50:45 EEST