[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: skype problems

Lauri wrote:
Lauri wrote:
Did you start Skype using the hijacker? It should report detecting the
bug and closing the session.

I've tried both ways. Renameing the hangup.wav got rid of that problem so skype can now be run multiple times pre reboot. skype_dsp_hijacker, if it works at all (only one audio device although I can snooker kmix into showing me 2 they are shared controls always)

/etc/asound.conf is the global file. There was also a file for normal
user, I can't remember it's name. ALSA's config file.

But this stuff can be put into /etc/asound.conf as well I assume, so I'll give that a try.

I might add that this same stuttering and endless echo of small fractions of an incoming word, completely drowning out the normal incoming sound, also plagues ekiga/gnomemeeting.

I am using a stereo headset with a boom mic, and I'm reported to sound great on the other end with skype, but while I sound great in my own ears, I am all chopped up on the other end when running ekiga.

When running skype at home, I was able to arrive at mixer settings using kamix that resulted in my not hearing myself, but I transmitted just fine. If I could hear myself then we had echo's, which is what I think is killing me now, but with this laptops ATI IXP ac97 lashup, a setting that stops me from hearing myself, also stops the transmission. There seems to be no way to isolate it into a true 4 wire circuit where what I say goes out only, and the incoming pcm is to my earphones only.

If there is a specific stanza I can add to /etc/asound.conf that would accomplish that, please tell which one in that "ALSA Project - the C library reference: PCM (digital audio) plugins" from <http://www.alsa-project.org/alsa-doc/alsa-lib/pcm_plugins.html> will accomplish this. Its only about 10 pages of very very small type :-(

I think I know what I need to do, but do not understand alsa well enough to recognize what it is I need to setup to accomplish this 4 wire setup, or even if its possible to do this on this laptop.


I have a script for starting Skype. It kills started Skype processes and starts a new one using skype_dsp_hijacker, using the 2nd soundcard.
To remove choppyness, I use the dmix'ing code and try to keep capture as
low as possible (Skype needs it, otherwise your voice won't reach other
person). Everything works fine with echo123, but not when talking to
other people.

This works fine for me, on both, FC4 and FC5.


I'm gradually narrowing this thing down, and what it seems to be coming down to is that this ATI-IXP chipset used in this HP lappy is incapable of sustained full duplex operation.

So now my question is: Is there a hidden software switch that might enable this?

Those switches in audacity's prefs that would be enabled to get what would be simultainious play, only serve to generate almost exactly the same effect in my headphone earpieces as I'm suffering with skype. The other switch, the one that enables playback of other tracks for doing voiceovers and such, produces little sound in the phones while recording, and that skypey static can be heard way down in level, but does mix them properly when the new track is played back, and apparently without the added static. I have to admit that when 3 of me are talking at the same time, its as close to gibberish as can be. :-)

So, back to the original question, is this chipset even capable of full duplex operation? Its identified as an ATI-IXP, or conexant id 30.
An lspci looks like this:
00:14.3 ISA bridge: ATI Technologies Inc IXP SB400 PCI-ISA Bridge
00:14.4 PCI bridge: ATI Technologies Inc IXP SB400 PCI-PCI Bridge
00:14.5 Multimedia audio controller: ATI Technologies Inc IXP SB400 AC'97 Audio Controller (rev 02)

Cheers, Gene

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]