OP25 OP25 (boatbod) broadcasts too short

Status
Not open for further replies.

SFusion

Newbie
Joined
Jan 25, 2017
Messages
4
@boatbod - I apologize if this is the wrong place to ask this, I looked for a related thread but I don't know what words to search.

I have rx.py for P25 Phase II simulcast system running on my RPI 3b+

Here is my config:

Code:
./rx.py -l http:192.168.1.9:8080 --nocrypt --args "rtl" --gains 'lna:36' -S 2400000 -X -q 0 -v 1 -2 -V -U -T trunk.tsv 2> stderr.2

My problem is the broadcasts sound like they are getting cut short - from the beginning and the end.

They sound like they start too late and end too early.

The person is already speaking by the time the audio plays, and it seems to cut off the broadcast before the speaker is quite finished.

It is maybe 0.25 to 0.5 seconds on each end, but enough to be hard to listen too.

I have no idea what setting to adjust?

Thanks!
 

wgbecks

Active Member
Joined
Jan 17, 2005
Messages
949
Location
NE Wisconsin
Your audio is directed to go direct to the default Linux sound system that may be using ALSA or Pulse Audio. It would be helpful to post the Linux OS in use and the hardware platform you're running it on.
 

SFusion

Newbie
Joined
Jan 25, 2017
Messages
4
I am running DietPi (stripped down Debian) on a RPI 3b+

RTL-SDR dongle

Audio is all ALSA as far as I know.
 

noamlivne

Member
Joined
Sep 7, 2012
Messages
182
Also, becasue you are using Raspberry Pi which can be weak, change the value of -S to something much lower such as 1000000 or even much less.
 

boatbod

Member
Joined
Mar 3, 2007
Messages
3,371
Location
Talbot Co, MD
Also, becasue you are using Raspberry Pi which can be weak, change the value of -S to something much lower such as 1000000 or even much less.
For a RPi3b the biggest things you can do is "-S 1000000" to reduce the wasted cpu cycles for unnecessary usb data transfer/filtering and make sure you are not running any of the gnuplots.
 

SFusion

Newbie
Joined
Jan 25, 2017
Messages
4
For a RPi3b the biggest things you can do is "-S 1000000" to reduce the wasted cpu cycles for unnecessary usb data transfer/filtering and make sure you are not running any of the gnuplots.

Great thank you - I will turn it down. I didn't think it was an issue because CPU is pretty consistently only 20% or even less when running headless. But I didn't think about usb data.

On the gnuplots - I run the web interface, are those plots always running in the background - even if I don't open them? Or is there some way to disable them?

Thanks!
 

boatbod

Member
Joined
Mar 3, 2007
Messages
3,371
Location
Talbot Co, MD
Great thank you - I will turn it down. I didn't think it was an issue because CPU is pretty consistently only 20% or even less when running headless. But I didn't think about usb data.

On the gnuplots - I run the web interface, are those plots always running in the background - even if I don't open them? Or is there some way to disable them?

Thanks!
On the boatbod repo the gnuplots are only active if you call them up from the command line or turn them on from the terminal. If the web terminal closes the plots will be stopped too.
 
Status
Not open for further replies.
Top