This is about RX and not TX.
This has caused some headaches. I want to use one trx and one antenna for this test. (Icom 7300 and a Hyendfed)Now the problem is that Direwolf and QtSoundModem both use the sound card. Now you can’t both use the same sound card at the same time. So we will have to use two virtual sound cards and route the audio to these sound cards. For this I use “pactl” with this I can manipulate the PulseAudio server.
pactl load-module module-virtual-sink sink_name=direwolf And pactl load-module module-virtual-sink sink_name=qtsoundmodem
Actually we are making a virtual audio card for Direwolf and QtSoundModem. Now we need to route the Audio from the input to the virtual audio card. This is possible with PavuControl.
ADEVICE pulse ACHANNELS 1 CHANNEL 0 MYCALL N0CALL MODEM 300 1000 1200 AGWPORT 8008 KISSPORT 8009
Uhmmmmm port 8001. I don’t understand that yet, in the config it really is port 8009.
Now I haven’t fully read up on PulseAudio and pactl and the virtual cable/cards. Perhaps my wording and references are not quite correct.
Now I have the opportunity to test…..
3 thoughts on “Direwolf Vs QtSoundModem (part 1)”
it’s not exactly like that
pulse audio should not be used as it generates
more use of the cpu and processes the audio in an indirect way
the ideal is to use alsa giving all the priority to the soundmodem
another thing that direwolf does not behave well in packet nor is it intended to be used in this ax25 modem
in buenos aires we are 5 bbs in 145010 and direwolf generates a loop since it cannot resolve the order of the packets
example when 3 packets are sent and there is an error in one… direwolf sends all 3 again and the correspondent waits for 1 and asks again and this is how this error occurs… that is why those of us who do packet by rf continue to use the old soundmodem from linux
the configuration that you propose works but it is not the correct one so that it works in rf with several bbs sending fwd
I always read your blog. Too bad I don’t speak English. I have a lot of information that I would like to share.
always supporting the packet your effort is appreciated
Hi Eduardo (Edward),
Thanks for the feedback. I’ve read that PulseAudio uses a bit more CPU. This is not really important for the test. I don’t read anywhere that the audio is processed by default, but that it is possible is correct.
I’ve been really busy getting it done with Alsa. Unfortunately, that didn’t work out.
I haven’t heard/read anything about the behavior you describe. Of course I could have missed that.
The test I’m doing is mainly about the receive side of Direwolf and Qtsm.
If you’d like to share something, I’d be very happy. You can also send it in Spanish and I will try to translate it.
73 Niels PD9Q
could you make your post in text mode
and send them as a bulletin on the packet network