On 23/07/2015 01:33, Alan VK2ZIW wrote:
Hi Richard and Bill,
Hi Alan,
I've read the docs:(http://doc.qt.io/qt-5/qaudioinput.html#setVolume).
But, how do you have two or more Input devices as in I and Q ?
Sound cards only have real amplitude magnitude samples (i.e. only the
magnitude part of a polar vector) at 0Hz base frequency and do not
capture Cartesian form sample values. Normally I & Q signals are passed
by using a stereo stream convention and the applications producing and
consuming the stream have to be aware of the usage and which stereo
channel is I and Q respectively. So this is not a Qt issue as such but
something that would be dealt with within the application(s).
(I'm not a Qt programmer)
Alan VK2ZIW
73
Bill
G4WJS.
*On Wed, 22 Jul 2015 20:48:47 +0100, Bill Somerville wrote*
> On 22/07/2015 20:33, Richard Shaw wrote:
>
> Hi Richard,
>
>
>
> Wed, Jul 22, 2015 at 2:14 PM, Bill Somerville
<[email protected] <mailto:[email protected]>> wrote:
>
>
> On 22/07/2015 18:45, Richard Shaw wrote:
>
> Hi Richard,
>
> I'm not sure who else has battled this but Qt in their
infinite wisdom decided they need a random name for the audio
connection to PulseAudio which means it will NEVER remember the
volume level you set.
>
What's wrong with setting the levels in WSJT-X, those settings
are saved between sessions?
>
> I have to cut it WAY down and I haven't done any actual analysis,
but the instructions indicate that that's not a good thing. Why amp
up your audio in software just to knock it back down in software? It
can't possibly help the audio quality.
You are right in principle but I would expect the default level to be
0dB i.e. what comes out of the A/D converter so doing all digital
attenuation in WSJT-X is optimal since it limits any accumulated
quantization error. The same would apply if the WSJT-X level were left
at 0dB (the middle I believe) and all digital adjustment were done in
the pulseaudio digital domain.
>
> The best solution is to set the analog level at he rig to a best
approximation and only do fine adjustments per band or to compensate
for AGC variations in the Rx.
>
>
> And either way, it's still just obnoxious on the Qt side. I wonder
how many presets PulseAudio has stored for device names it will never
see again?
Agreed that it would be nice to have a way of setting the stream name
on Linux, but there is an alternative where, like we do with Tx, the
stream level can be directly adjusted from the application using the
stream (http://doc.qt.io/qt-5/qaudioinput.html#setVolume). This allows
the application to store the level itself. Currently WSJT-X doesn't
use this, instead it has its own gain/attenuation stage further
downstream in the DSP chain. I believe the built in gain stage is both
historical from before we switched to Qt audio and also to provide
gain as well as attenuation (gain is not available on some platforms).
>
> I expect the availability of adjustment of the stream level via the
Qt API is the reason why our request to control the stream name has
been ignored by the Qt developers.
>
>
> Thanks,
> Richard
>
73
> Bill
> G4WJS.
Alan
Man's greatest waste of time: Worshipping the wrong God.
Consider Jesus.
---------------------------------------------------------------------------
Alan Beard Unix Support Technician from 1984 to today
70 Wedmore Rd. Sun Solaris, AIX, HP/UX, Linux, SCO, MIPS
Emu Heights N.S.W. 2750 Routers, terminal servers, printers,
terminals etc..
+61 2 47353013 (h) Support Programming, shell scripting, "C",
assembler
0414 353013 (mobile) After uni, electronics tech
------------------------------------------------------------------------------
_______________________________________________
wsjt-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/wsjt-devel
------------------------------------------------------------------------------
_______________________________________________
wsjt-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/wsjt-devel
