Subject: Re: [linux-audio-dev] Realtime
From: Jussi Laako (jussi_AT_jlaako.pp.fi)
Date: Thu Jun 29 2000 - 00:19:50 EEST
yodaiken_AT_fsmlabs.com wrote:
>
> Well, I am very interested in the question of whether the RTLinux model
> of breaking a RT application into tiny RT parts and user parts can work
> in the audio domain. Speaking with the authority that comes from
I've been testing that kind of model with my sonar software. So far I've
been only using my DAQ card as 8-channel AD-converter. I'm also planning to
include AGC stage in RTL module.
I will try to extend that to full-duplex procesessing chain in my upcoming
digital room correction system.
Lack of soundcard drivers is one of the biggest problems. And the most time
consuming task is to write drivers as RTL module. Soundcards tend to be more
complex compared to DAQ cards. And getting needed documentation about
soundcards is another problem. At least they don't come with full
documentation as my DAQ card did. (Yes yes, I know about the AudioPCI
driver!)
My biggest problem with standard Linux has been delays caused by devices
eating too much bus bandwidth. I tried to use 2.2.15pre LL patches, but
those didn't help at all. Some bandwidth limit / device would be nice.
As we all know, virtual memory (swap) is bad-thing(tm) for realtime systems
and RTL is good for avoiding problems caused by swapping.
- Jussi Laako
-- PGP key fingerprint: 161D 6FED 6A92 39E2 EB5B 39DD A4DE 63EB C216 1E4B Available at: ldap://certserver.pgp.com, http://keys.pgp.com:11371
This archive was generated by hypermail 2b28 : Wed Jun 28 2000 - 21:53:43 EEST