Subject: Re: [ecasound] ecamegapedal, full-duplex, usb audio
From: Mikko Nummelin (mnummeli_AT_cc.hut.fi)
Date: Mon Sep 09 2002 - 16:40:33 EEST
On Mon, 9 Sep 2002, janne halttunen wrote:
> On Mon, 09 Sep 2002 13:12:15 +0100
> Daniel Barlow <dan_AT_telent.net> wrote:
>> This is a full-duplex-capable device. I can't see (using strace)
>> anywhere in ecamegapedal that it's calling the SNDCTL_DSP_SETDUPLEX
>> ioctl, though I may be stracing the wrong thing.
>> I'm using Linux 2.4.20-pre5-ac3 with the in-kernel audio drivers
> Even if the device is capable of full-duplex doesn't guarantee that it
> is supported by the driver. Perhaps you could test with ecasound: first
> playback, then recording, and finally playback and recording.
> ecasound -i -i test.wav -o /dev/dsp
> ecasound -i /dev/dsp0 -o test2.wav
> ecasound -a 1 -i test.wav -o /dev/dsp \
> -a 2 -i /dev/dsp0 -o test2.wav
> I am thinking about writing somesort of a test-script for this common
> category of problems. Don't hold your breath, though :)
Full-duplex is _not possible_ in OSS/Free-drivers which ship normally with
2.4.x -kernel-series. In order to do recording and playback at the same
time, one must either
(i) get ALSA drivers from http://www.alsa-project.org and follow
installation instructions carefully.
(ii) buy commercial OSS-drivers
(iii) Compile the kernel from 2.5.x development series (they include
ALSA), namely the latest up-to-date, which is 2.5.33 . _THIS IS RISKY_! I
have myself tested it, and it is very difficult to compile, and even after
that, I get lots of boot-time error messages, but the systems however
keeps up. Some people who have tested the development kernels more
thoroughly have made bug-reports of even corrupted hard drive file
systems, so don't say later that you had not been warned!
Mikko Nummelin
-- To unsubscribe send message 'unsubscribe' in the body of the message to <ecasound-list-request_AT_wakkanet.fi>.
This archive was generated by hypermail 2b28 : Mon Sep 09 2002 - 16:41:34 EEST