[arm-allstar] Linear vs. Log audio plots

[David McGough]

Hi Everyone,

To give a little more background, by default, the "linear" TX level
calculation is used only for the original CM108 (PID: 0x000C, original 
URI, some clones). Otherwise, the "log" calculation method is used.

Several weeks ago a ham was having problems fine-tuning the TX audio 
at very low levels. He had a chip where the "log" level adjustment was 
getting applied, so, I forced the driver to use the "linear" method. And, 
then I forgot and left that code in there (oops!).

For this application (i.e.: setting the deviation level of an FM 
transmitter), I think it makes more sense to leave the adjustment method 
"linear," however, that obviously will disturb audio levels for existing 
installations, where the driver originally selected the "log" method.

BTW, if this application was to adjust the volume driving a speaker, the 
log level method (AKA: "audio taper") is preferred, since the human ear is 
a logarithmic device!

Anyhow, now you can selectively force the simpleusb driver to use the
method you prefer, rather than just accepting the defaults. This setting
may also ease transition from original CM108 chips to other varieties,
like the CM119A found in the URIx and Repeater-Builder RIM products.


73, David KB4FXC


On Mon, 2 Oct 2017, "Doug Crompton via arm-allstar" wrote:

> Dave did some plotting of the TX audio level changes using the linear and
log methods. As expected they are quite different. The linear has better
spread in the high and low extremes and the log in the center. I might
point out that many of us are using the newer URI which by default would
use the log method and would have little adjustment at the ends of the
1-999 range. The linear may be preferred but since this hack has been in
the code for so long we are leaving it in as default but giving you the
ability to select linear or log manually. Here are his comments and graphs.

----------------------------------------------------------------------------
Here are GNUplot graphs showing the linear and log modes of TX audio level
adjustment. The log mode seems whacky to me, but, I verified that is the
correct formula used in the channel driver!

NOTE that the CM1xy mixer range is 0 to 152.

Linear calculation (old URI cards and tx_audio_level_method == 1):

https://hamvoip.org/photos/Gnuplot-TX-method-1-linear.png


Log calculation (any non-cm108 adapter):

https://hamvoip.org/photos/Gnuplot-TX-method-2-log.png

73, David KB4FXC
_______________________________________________

arm-allstar mailing list
arm-allstar at hamvoip.org
http://lists.hamvoip.org/cgi-bin/mailman/listinfo/arm-allstar

Visit the BBB and RPi2/3 web page - http://hamvoip.org