DSD FME

lwvmobile

DSD-FME
Joined
Apr 26, 2020
Messages
1,297
Location
Lafayette County, FL
You have to manually click those (with DSD-FME not running) to enable the plugin first, then it'll go to a 'ready' or 'listening' state, then when you start DSD-FME, it'll connect to those. BTW, change the TCP server back to 7355, its the Gpredict that 'should' be 20022, or whatever it is in the box, you change the -U option to match it. Not the other way around.
 

mrscanner2008

Member
Joined
Jun 21, 2008
Messages
333
Location
Canada
With both TCP or virtual cable, nothing is received. see attach DSD-FME and DSD+FL screenshot for the same system .
 

Attachments

  • DSD-FME TCP.png
    DSD-FME TCP.png
    357.8 KB · Views: 55
  • DSD-FL Mt carmel.png
    DSD-FL Mt carmel.png
    402.7 KB · Views: 52
  • nxdn_ran_26.zip
    18.9 KB · Views: 11

lwvmobile

DSD-FME
Joined
Apr 26, 2020
Messages
1,297
Location
Lafayette County, FL
What port number is listed in the GPredictConnector plugin, in the screenshot, its showing you are connected on port 4532, but that isn't the default for that plugin, that's the default inside of SDR++ however, by all means, it should be tuning from what I'm seeing in the screenshot, but in the log, its not showing a RIGCTL connection, so not quite sure why that setup won't tune to the VCs for. Could be that I still need to tweak a couple things, either that, or GPredict isn't responding or it tunes there, but can't get a frame sync. How is the system signal on those VC's, strong, or kind of weak? Might be worth narrowing your VFO inside of SDR# a little bit, I find 4000-5000Khz good for NXDN48, having the VFO larger than the actual signal can lead to some bad decodes sometimes, or frame sync issues.
 

lwvmobile

DSD-FME
Joined
Apr 26, 2020
Messages
1,297
Location
Lafayette County, FL
Also, just worth mentioning, make sure you have SDR# in the correct tuning mode, I don't think Sticky tuning works correctly with the gpredict plugin, it'll change the frequency numbers, but not actually change the tuning. Seems like I had issues with that in my VM as well, sometimes you launch SDR#, and it refuses to change its tuning even with new numbers.
 

ArloG

Member
Joined
Feb 14, 2020
Messages
346
This link to the current Windows binary is here.




Depends on what all you want to do with it. The easier thing might be to just start it up and route the audio into it with the VB cable like how you have DSD+ listening to it.

Does the SDR Application handle tuning your Icom? if so, which SDR Application and/or plugin is it? Is it compatible with Rigctl?
The number one thing that you mentioned previously is the ability to access menus as opposed to the light edition.
Much like DSD+ free vs Fastalne does. That caused confusion for me when trying to help DSD+ users in the past (and I'm STILL not a pro!!)
I had Fastlane since being mistakenly lead to believe it would do more with an I/Q input (passive digital monitor). I bought it for life.

And don't really intend on getting an RTL radio. Although I do have an RSP2 that does work with DSD+. But that's DSD+.....right?
My intention is to not start an antenna farm or buy another antenna switch with more selections.
I live in a rural area but in direct LOS of an antenna site with P25, NXDN, DMR,etc.. DSD+ just hasn't "tripped my trigger" and voice on CC is spotty and fragmented at best. When I get any voice at all.

As for your question. HDSDR is my main SDR program. SDR Console, second. And an older ver of SDR# with the usrp plugin to allow me to use my radio with I/Q controls. So yes. Tuning, mode, rf gain, mode, if and audio bandwidths are all controllable by them.
I know that setting up the .bat files is mandatory. And definitely could setup a dedicated Linux box. Or...in a VM???? maybe?
I guess just having access to drop down menus would get me started and then learning the in's and out's of the .bat files.
Something the compiled ready-to-go version for Windows doesn't have. Am I correct?

Oh. Been off on assignment in a land far away. I believe HDSDR has CAT control as does Console. Need to sit down and refresh myself.
Thanks! Great freakin' work!!
 

lwvmobile

DSD-FME
Joined
Apr 26, 2020
Messages
1,297
Location
Lafayette County, FL
Thanks! Great freakin' work!!
Thanks, I appreciate hearing that, even if you haven't tried it yet.

I know that setting up the .bat files is mandatory. And definitely could setup a dedicated Linux box. Or...in a VM???? maybe?
I guess just having access to drop down menus would get me started and then learning the in's and out's of the .bat files.
Something the compiled ready-to-go version for Windows doesn't have. Am I correct?
Dedictated VM is up to you if you want to go that way, but I was able to package enough of the Cygwin environment together in a zip file with the bat files and exe files, the current Windows 'Lite' RC versions maintain all the functionality of the main branch with ncurses terminal, menu, etc. When I gave it that name this time, I did so before realizing I wouldn't need to cut features, just needed to find a way to make it work prepackaged.

Oh. Been off on assignment in a land far away. I believe HDSDR has CAT control as does Console. Need to sit down and refresh myself.
Thanks! Great freakin' work!!

I wonder, have you ever heard of or tried our hamlib before? Its a protocol that can run that can remote control your transceiver. The hamlib radio includes a protocol called rigctl, which is the bases of the rigctl support in dsd-fme and also the ones used by the gpredict plugin for SDR# and the built in rigctl clients in both GQRX and SDR++, among other places. If you could set this library up on your machine and configured for you ICOM, and get the rigctld process running, it is feasible that DSD-FME should be able to control the device directly in a tuning/trunking sense.


As far as what comes out of the radio itself, does it have any sort of disc tap output, or is it just the IQ that can be send over network or USB or something that gets read in by HDSDR, etc? I'm honestly not very familiar (not familiar at all) with how your device works with the software and how the audio is passed from point a to point b to point c. I might have to resort to youtube videos to see if I can put it all together :ROFLMAO:

DSD+ just hasn't "tripped my trigger" and voice on CC is spotty and fragmented at best. When I get any voice at all.

I wonder if there is any sort of audio filtering being applied to prevent a good decode. If signal is good, and its fed properly, should be okay of voice and cc stuff for the most part, assuming its not LSM/Simulcast related.
 

millam

Old Radio Guy
Premium Subscriber
Joined
Jan 18, 2005
Messages
722
What is the purpose of "-N 2> log.ans"? Seems to log the screen, I am using a CMD window to
start the program.

Mil
 

ArloG

Member
Joined
Feb 14, 2020
Messages
346
Thanks, I appreciate hearing that, even if you haven't tried it yet.

Actually it is rc2 installed. Just more reding and trying is needed. Audio in/out routing stopped me short of going further. It was late.

Dedictated VM is up to you if you want to go that way, but I was able to package enough of the Cygwin environment together in a zip file with the bat files and exe files, the current Windows 'Lite' RC versions maintain all the functionality of the main branch with ncurses terminal, menu, etc. When I gave it that name this time, I did so before realizing I wouldn't need to cut features, just needed to find a way to make it work prepackaged.



I wonder, have you ever heard of or tried our hamlib before? Its a protocol that can run that can remote control your transceiver. The hamlib radio includes a protocol called rigctl, which is the bases of the rigctl support in dsd-fme and also the ones used by the gpredict plugin for SDR# and the built in rigctl clients in both GQRX and SDR++, among other places. If you could set this library up on your machine and configured for you ICOM, and get the rigctld process running, it is feasible that DSD-FME should be able to control the device directly in a tuning/trunking sense.

Good conversation.
The receiver as many others as far as I know have CI-V, CAT control over USB using a driver which creates a virtual com port. Also over TCP/IP.
Will definitely know more very soon after some reading and trying. If this is a hint that using hamlib to control the radio through your program would work. Awesome! The bandstand for DSD+ is to support trunking with what was 2 RTL based dongles and then 1. Same for the SDRPlay RSP radios. I know using Commander allowed radio control over usb with a virtual com port.
Also I investigated hamlib of some flavor once but it had no R8600 support. Programmers speak so fluidly of things that are a bit difficult to grasp.
I often wondered why a build to suit, insert your own ASCII, Hex, BCD codes rig control program wasn't readily available. You know. Like x00ac for step up in freq., x0312 tune to, and all that stuff. Way above me. Guys are building external encoders that take care of the inability of the receivers native freq. step inability to tune in 1Hz increments. Not important in VHF and most HF. At least not to me.



As far as what comes out of the radio itself, does it have any sort of disc tap output, or is it just the IQ that can be send over network or USB or something that gets read in by HDSDR, etc? I'm honestly not very familiar (not familiar at all) with how your device works with the software and how the audio is passed from point a to point b to point c. I might have to resort to youtube videos to see if I can put it all together :ROFLMAO:

Out of the radio. A USB audio port. No good for DSD. Works with fldigi and sorcerer, etc soundcard apps. without needing sdr software.
No native disc. out like my IC-R8500 has.
Also 10.7MHz if out through usb and a rear RCA port.
In DSD+ the path would be like this:
Radio I/Q USB > PC > SDR Application > SDR app. Audio Out piped to > VB Cable (or Virtual Audio Cable if you please) > DSD+
And DSD+ would then be run in Passive Digital Monitor mode. Meaning no rig control. Just an unfiltered (just like disc. in) audio signal to feed DSD+.
That's all I got on that. Hey man. I'm a Floral City boy in the great white north now.


I wonder if there is any sort of audio filtering being applied to prevent a good decode. If signal is good, and its fed properly, should be okay of voice and cc stuff for the most part, assuming its not LSM/Simulcast related.

As mentioned. Once I get audio routing aced in FME. It will be basically the same as DSD+ for now at least.
DSD+ gives a very good data stream showing pretty much what your screen shots show. CC, VC, If the stream is P25 p1 or 2, DMR, NXDN 4800 or 9600. In cases where there may be voice on the CC. It was 'garbled' and spotty. If I parked the radio on a VC. Good luck! I get data streaming but no voice like the 'tube vids show.
It peed me off that the Fastlane sub didn't 'in your face' let you know that with radios like mine, you're not really getting anything except a menu bar that the free version didn't have. Functionality wise.

Please dont get side tracked on your project. Here on my rant at least. There is a learning curve.
Just opening the "0 - start-dsd-fme-no-pulse.bat" file.....I mean. I'm running Windows, so no Pulse Audio, right? Kinda sent me into the ethers.
 

millam

Old Radio Guy
Premium Subscriber
Joined
Jan 18, 2005
Messages
722
Is there anyway to sync this so its easier to see?

Mil
 

Attachments

  • p1_1.JPG
    p1_1.JPG
    74.5 KB · Views: 48

EI9BAB

Member
Joined
Sep 3, 2021
Messages
57
I am using multiple versions of DSD and DSD-FME. I have everything set up in a windows command shell for DSD-FME-LITE to decode live audio and the audio comes out fine when I CTRL-C that and re-run it with the "... -r filename.amb" there is unfortunately no audio heard. (I've tried this with and without Pulse Audio but I think the problem is the actual decoding, although I do see plenty of info when the "... -Z" option is included. Also audio errors: 0 / Total header errors:0 so it does seem to be readable.

I have also tried the same decode with output to a WAV file but that also just gives silence for the duration of the sample. The same .amb file will play fine on other versions of DSD so again the input file seems fine and I'm now at a bit of dead-end. Any ideas please?
 

lwvmobile

DSD-FME
Joined
Apr 26, 2020
Messages
1,297
Location
Lafayette County, FL
I have also tried the same decode with output to a WAV file but that also just gives silence for the duration of the sample. The same .amb file will play fine on other versions of DSD so again the input file seems fine and I'm now at a bit of dead-end. Any ideas please?

I've just tested it and confirmed the same behavior. Its not a feature I use anymore, see below. Let me look at it and see why it isn't producing any audio. I can see its decoding the .amb/.imb files with the -Z payload option turned on, but no audio coming out. I'll try to get it patched and working again. Might not be immediately though.

Meanwhile, if you'd like to capture and playback, you can use the symbol capture bin method, in addition to just capturing the voice ambe/imbe frames, it captures the entire signal and can replay it. If its just data, it'll blow through it pretty fast and then sllowdown to playback audio in real time. Depending on how you like to use dsd and dsd-fme, this can be done in dsd-fme with

-c capture.bin to capture a system signal as a symbol dump

-i capture.bin to play it back

Also, if you use the ncurses terminal, you can create per call wav files as another option.
 

EI9BAB

Member
Joined
Sep 3, 2021
Messages
57
I've just tested it and confirmed the same behavior. Its not a feature I use anymore, see below. Let me look at it and see why it isn't producing any audio. I can see its decoding the .amb/.imb files with the -Z payload option turned on, but no audio coming out. I'll try to get it patched and working again. Might not be immediately though.

Meanwhile, if you'd like to capture and playback, you can use the symbol capture bin method, in addition to just capturing the voice ambe/imbe frames, it captures the entire signal and can replay it. If its just data, it'll blow through it pretty fast and then sllowdown to playback audio in real time. Depending on how you like to use dsd and dsd-fme, this can be done in dsd-fme with

-c capture.bin to capture a system signal as a symbol dump

-i capture.bin to play it back

Also, if you use the ncurses terminal, you can create per call wav files as another option.
Thanks for the speedy response. I'm glad it wasn't just me being stupid! Those instructions are useful but I have a couple of channels where there is not a lot of activity and it would probably create quite a big file to monitor for 24 hours. I already have these samples so I'll wait for a fix - but no pressure. I can code a little bit so if there is something I can do myself or you want any help testing or whatever then let me know. Thanks for all the amazing work on this seriously cool version!
 

EI9BAB

Member
Joined
Sep 3, 2021
Messages
57
On a separate question - I presume that the TCP link in DSD Fastlane's FMP24.exe/DSDPlus.exe is not compatible in any way with the TCP link in DSD-FME? I couldn't get them to work together but as FMP does great Audio pre-processing for DSD, I am just using the audio output option with Virtual Cable to route the audio into DSD-FME for the moment. Just thought it worth asking anyway!
 

lwvmobile

DSD-FME
Joined
Apr 26, 2020
Messages
1,297
Location
Lafayette County, FL
Thanks for the speedy response. I'm glad it wasn't just me being stupid! Those instructions are useful but I have a couple of channels where there is not a lot of activity and it would probably create quite a big file to monitor for 24 hours. I already have these samples so I'll wait for a fix - but no pressure. I can code a little bit so if there is something I can do myself or you want any help testing or whatever then let me know. Thanks for all the amazing work on this seriously cool version!

Looking through the code now, and I still can't figure out why they suddenly aren't playing, the trace shows it should be playing, but probably some arbitrary change I made along the way has them not playing for whatever reason. I'll do some more digging. Seems to affect both the 'main' and the 'lite' branches.

On a separate question - I presume that the TCP link in DSD Fastlane's FMP24.exe/DSDPlus.exe is not compatible in any way with the TCP link in DSD-FME?

I tried to make it work with both FMP24 in fast lane, and the older packaged in FMP.exe in v1.101, and while i can technically connect to the TCP socket in FMP, I can't get anything usable out of it. I have no idea how the two communicate, or if the older FMP.exe even sends audio out TCP, or if that is just for remote control. I honestly didn't invest a lot of time trying to figure it out, either, by the way, just did it more or less to see what would happen. FMP24 connects, reports an error, and then disconnects.
 

mrscanner2008

Member
Joined
Jun 21, 2008
Messages
333
Location
Canada
What port number is listed in the GPredictConnector plugin, in the screenshot, its showing you are connected on port 4532, but that isn't the default for that plugin, that's the default inside of SDR++ however, by all means, it should be tuning from what I'm seeing in the screenshot, but in the log, its not showing a RIGCTL connection, so not quite sure why that setup won't tune to the VCs for. Could be that I still need to tweak a couple things, either that, or GPredict isn't responding or it tunes there, but can't get a frame sync. How is the system signal on those VC's, strong, or kind of weak? Might be worth narrowing your VFO inside of SDR# a little bit, I find 4000-5000Khz good for NXDN48, having the VFO larger than the actual signal can lead to some bad decodes sometimes, or frame sync issues.
I tried different port combinations: TCP 7355 or 20022 and GPredict 4532 or 20022, same result. With the combination TCP 7355 and GPredict 4532 I noticed that DSD-FME receives some conversation but it misses a lot. log included.
I even tried with another version of SDRsharp but without success. I will try the other modes (DMR, CON+, CAP+) to see if I have the same problem.
 

Attachments

  • nxdn_ran_26 (3).zip
    96.8 KB · Views: 6

lwvmobile

DSD-FME
Joined
Apr 26, 2020
Messages
1,297
Location
Lafayette County, FL
I already have these samples so I'll wait for a fix - but no pressure.

Okay, I figured out something, its related to the auto gain. If you use the -g option and pick -1 it will disable auto gain, or you can use something like -g 1 or -g 2 then it'll play them again. I'll have to review the auto gain function again, seems like I did some things to it when working on the crackling/auto smoothing stuff a little while back.

This command worked for me, but you'll need to adjust it based on your own usage.
dsd-fme -g -1 -r mbefiles/*

I noticed that DSD-FME receives some conversation but it misses a lot. log included.

I'll have to review your logs some more, but did you do any tweaking on your VFO inside of SDR#, need to bring that bandwidth down some (4000ish), tweak the gain input levels, turn on the squelch, and also disable any audio processing/filtering if its enabled. Also, what command did you end up using in your startup. Make sure it has the -T option for trunking in there.

If its still not working after all that, could be some code on my end that still needs cleaning up. Its hard to test this one, since I don't have trunking NXDN anywhere nearby, and not many people are willing to let me remote in and control their stuff (understandably so). Testing solely on wav files isn't the greatest for trunking because observing the behavior in real time and making adjustments is necesary to getting it to work properly without playing the guessing game.
 

mrscanner2008

Member
Joined
Jun 21, 2008
Messages
333
Location
Canada
Okay, I figured out something, its related to the auto gain. If you use the -g option and pick -1 it will disable auto gain, or you can use something like -g 1 or -g 2 then it'll play them again. I'll have to review the auto gain function again, seems like I did some things to it when working on the crackling/auto smoothing stuff a little while back.

This command worked for me, but you'll need to adjust it based on your own usage.
dsd-fme -g -1 -r mbefiles/*



I'll have to review your logs some more, but did you do any tweaking on your VFO inside of SDR#, need to bring that bandwidth down some (4000ish), tweak the gain input levels, turn on the squelch, and also disable any audio processing/filtering if its enabled. Also, what command did you end up using in your startup. Make sure it has the -T option for trunking in there.

If its still not working after all that, could be some code on my end that still needs cleaning up. Its hard to test this one, since I don't have trunking NXDN anywhere nearby, and not many people are willing to let me remote in and control their stuff (understandably so). Testing solely on wav files isn't the greatest for trunking because observing the behavior in real time and making adjustments is necesary to getting it to work properly without playing the guessing game.
I tried all possible settings in SDR# without success. I will try with SDR++ sometimes... I will look to see if I could not provide you with a remote with N48 trunking system.
 
Top