DSD FME

[lwvmobile]

Trunking using the Icom is my interest. Is it or will it ever be a feature in FME?

Without actually owning the hardware, I can't really confirm that this is currently possible or not, or the steps that would be required to make it work. DSD-FME does indeed use RIGCTL as a tuning possibility, so if there is some interface that can use that to command the Icom to tune to frequencies and a method to tunnel the audio into DSD-FME (HDSDR and VBCables?), then it should work. How to get from point a to point b on this, I honestly can't say.

I went back and looked at the old answer I posted about this, and the answer may be in hamlib, which seems to be an interface that puts on an outward rigctl server, and connects to various hardware to control it.

FAQ

Ham radio control library for rigs, rotators, tuners, and amplifiers - Hamlib/Hamlib

github.com

Also, here is the vicinity of the previous reply I made on this, with support links to hamlib, so perhaps the answers you are looking for are somewhere in there, but its going to take some experimentation and trail and error.

DSD FME

dsd-fme-lite.exe -fi -i tcp -U 20022 -C examples\nxdn_chan_map.csv -T -N 2> nxdn_ran_26.ans I run with the above command but TCP and GPredict didn't start (uncheck), maybe a code 18 from me :unsure:

forums.radioreference.com

I think the general idea would be to do something like this:

Trunking: DSD-FME(rigctl client) --> hamlib (rigctl server) -->hamlib(interface to Icom) -->Icom
Audio: Icom-->HDSDR or SDR Console --> VBCables --> DSD-FME

 

[ArloG]

Without actually owning the hardware, I can't really confirm that this is currently possible or not, or the steps that would be required to make it work. DSD-FME does indeed use RIGCTL as a tuning possibility, so if there is some interface that can use that to command the Icom to tune to frequencies and a method to tunnel the audio into DSD-FME (HDSDR and VBCables?), then it should work. How to get from point a to point b on this, I honestly can't say.

I went back and looked at the old answer I posted about this, and the answer may be in hamlib, which seems to be an interface that puts on an outward rigctl server, and connects to various hardware to control it.

FAQ

Ham radio control library for rigs, rotators, tuners, and amplifiers - Hamlib/Hamlib

github.com

Also, here is the vicinity of the previous reply I made on this, with support links to hamlib, so perhaps the answers you are looking for are somewhere in there, but its going to take some experimentation and trail and error.

DSD FME

dsd-fme-lite.exe -fi -i tcp -U 20022 -C examples\nxdn_chan_map.csv -T -N 2> nxdn_ran_26.ans I run with the above command but TCP and GPredict didn't start (uncheck), maybe a code 18 from me :unsure:

forums.radioreference.com

I think the general idea would be to do something like this:

Trunking: DSD-FME(rigctl client) --> hamlib (rigctl server) -->hamlib(interface to Icom) -->Icom
Audio: Icom-->HDSDR or SDR Console --> VBCables --> DSD-FME

Thank you so much. This is the most detailed flowchart by far from anyone. And I have a fresh formatted pc to get things rolling.
I do apologize but once I reached a bottle neck I kind of put FMEl aside. Not being a smartass but kind of like 'oh crap another RTL chipset only application'. SDR Console frustrated me because although it seems that voice channel tracking is possible. It looks like RTL radios only work with it. And of course DSD+ has its limited radio specs to allow trunking.

Before I get to it. Looking at your flowchart. Can I assume that if I rest on a control channel with FME. And a voice channel that would tune another receiver normally to it is indicated. That rigcrtl/FME/HDSDR meld together to tell the R8600 to jump over to the voice channel.
And then when the VC clears the radio will jump back to the CC?
Man. I'm such a noob at this and the scanner guys are probably belly laughing.
I'll read back. Wondering if I need to make a .bat file to start FME up to see if the R8600 will play nice.
A bit more psyched just from you showing the logic of how things should work.
I've been able to always use rigctrl and cat (with USB as serial port....fuzzy on the Ethernet port) to control the radio. And of course the I/Q driver that works with HDSDR, SDR Console, etc. that can use an I/Q driver.

 

[ki4hyf]

I've been testing 'zDEV' off and on for a while, and I finally made it my default dsd-fme in anticipation of it becoming 'Main'. However, I noticed the 'zDEV' branch uses much more processor. Somewhere ~40% vs ~8%. This is using the "in-program" rtl for both instances. The command line is identical, I only changed the executable. It is very reproducible. BTW, I really like 'zDEV' and it's extra info.

 

[lwvmobile]

I've been testing 'zDEV' off and on for a while, and I finally made it my default dsd-fme in anticipation of it becoming 'Main'. However, I noticed the 'zDEV' branch uses much more processor. Somewhere ~40% vs ~8%. This is using the "in-program" rtl for both instances. The command line is identical, I only changed the executable. It is very reproducible. BTW, I really like 'zDEV' and it's extra info.

What's the build version on it? If its an earlier version that the current one, then it may be something that was fixed, I seem to recall something happening like a memory leak or similar that was addressed. Its also possible that it is using more CPU now due to the RMS function call or similar. I should also note, I changed the cmake back to build the zdev branch as dsd-fme now where for a while it was dsd-fme-zdev so I could get ready to just do a switcharoo on the main branch.

What hardware are you running on? Pi? Or a desktop or laptop? If its a low powered device, I can't say I know for sure how much CPU is being used, using higher end desktops and laptops primarily, its hard to gauge what 8% utilization is, and if its just a single thread, or across multiple threads.



This is the last update I made: v2.0.1-32-g20f0c85

 

[lwvmobile]

Not the same thing, but just for fun: did you know AnyTone has released a new firmware for their D878 to make them compatible with ARC4?

So...anybody got any samples of DMR AnyTone RC4 they want to submit?

 

[ki4hyf]

What's the build version on it?

v2.0.1-32-g20f0c85

What hardware are you running on?

Intel(R) 6 Core(TM) i7-10750H CPU @ 2.60GHz
64 gb RAM

MSI GF65 Thin – THE ULTIMATE SHOCKWAVE

The all new MSI GeForce RTX 30 Series gaming laptop – GF65 Thin brings thin and light to the 15.6” mainstream gaming laptop, featuring with up to 10th gen Intel Core i7 processor and latest GeForce RTX™ 30 Series graphics. GF65 Thin breaks the stereotype

us.msi.com

I have a lot of ram because I sometimes run several virtual machines. I wasn't running any virtual machines at the time I noticed this. I was also running two instances at that time.

I was looking at your reply and noticed the "Task Manager" picture. I've never used it, I never knew it was there. It reports the CPU usage at ~4%. Htop says ~40%. LOL! I'm not sure which one to believe. Here are some screenshots:

 

[lwvmobile]

I was looking at your reply and noticed the "Task Manager" picture. I've never used it, I never knew it was there. It reports the CPU usage at ~4%. Htop says ~40%. LOL! I'm not sure which one to believe. Here are some screenshots:

By default, htop reports as PERCENT_CPU, which is by thread or core, if you use F2 and browse to this location and change it to PERCENT_CPU_NORM, then it'll divide the usage by number of threads/cores available to you.



But at any rate, you still observe higher CPU usage, and while I was a bit brain dead in my last reply, there is a little more math being handled by the rtl input now, including low pass and rms, and frequent optimizations during each tune, so that probably will account for the higher usage.

 

[ki4hyf]

By default, htop reports as PERCENT_CPU, which is by thread or core, if you use F2 and browse to this location and change it to PERCENT_CPU_NORM, then it'll divide the usage by number of threads/cores available to you.

I learn something new every day.

But at any rate, you still observe higher CPU usage, and while I was a bit brain dead in my last reply, there is a little more math being handled by the rtl input now, including low pass and rms, and frequent optimizations during each tune, so that probably will account for the higher usage.

Gotcha. I do run dsd-fme on a pi4 4gb, but it's still on v2.0.0-100-gae58fab. I'll keep that version as a backup just in case. It's rock-solid.

 

[Forts]

So...anybody got any samples of DMR AnyTone RC4 they want to submit?

You still need? I can provide some if so.

 

[lwvmobile]

You still need? I can provide some if so.

Sure, if you want to send some, I'll take them.

 

[Forts]

Ok I should be able to whip some up this afternoon.

 

[lwvmobile]

Ok I should be able to whip some up this afternoon.

Sounds good to me.

 

[Forts]

Sent you a PM with a link to a wav sample.

Quick question.. is there a way to hold on a TG while it's active? I've been playing around today on my Cap+ system here at work and it's working very well, but I'm not sure if there is a way to hold on an active group.

Thanks for the great work!

 

[lwvmobile]

Quick question.. is there a way to hold on a TG while it's active? I've been playing around today on my Cap+ system here at work and it's working very well, but I'm not sure if there is a way to hold on an active group.

Not currently at any rate, I don't have any priority coded in for calls and groups, its just first come first serve. There is a white list mode, which will only tune to a channel if the group in the csv file is white listed with the 'A' designation, but in Cap+, calls will start on the rest channel anyways and there is no code to allow it to preempt the call to tune to the group in the white list. I might can make a change in the code (should just be a single line) that will allow the white listed TG to preempt the call in progress and be tuned, but if more than one TG is available in the white list, it will always favor the TG in the lowest LSN, even if you are listening to a whitelisted call from a TG in a higher LSN.

If I get a chance, I may code it and send you an exe file to test it with.

 

[Forts]

Great, thanks for the quick reply. I just wanted to make sure I wasn't missing something obvious.

 

[MTL_Emergencies]

New to DSD-FME
To monitor a Connect+/Cap+ system, do I need multiple dongles (with SDR++) and do I enter the LCN or LSN in the channels file?
Thanks

 

[Forts]

Just a single dongle. And yes, you need to have a channels.csv to properly steer things. Examples are in the, well, examples folder 😆

 

[MTL_Emergencies]

I did just that... DSD-FME detects activity but doesn't tune to voice frequency. I also doesn't see it try to tune to a vf.
Channel 1 is the CC


Here is the bat file I launch

cd dsd-fme
pulseaudio.exe --start --no-cpu-limit=TRUE --exit-idle-time=3600
ECHO The Pulse Audio Server should now be active!
@REM dsd-fme -fs -i tcp -U 4532 -T -C examples\ctm-site.csv -G examples\ctm-group.csv -N 2> log.ans
dsd-fme -fs -i tcp -U 4532 -T -C examples\ctm-site.csv -K examples\multi_key_hex.csv -N 2> log.ans

 

[Forts]

When you start FME does it show TCP audio and RIGCTL connected?

I haven't tried it with TIII stuff yet, just Cap+.

 

[MTL_Emergencies]

I see that RIGCTL is not connected. The port it's trying to connect is 7355 (on DSD-FME) but I specified port 4532 (on SDR++) and on the command line.