DSDPlus DSDPlus 2.165 bug?

[amcferrin90]

I didn't see an overall bug list. I updated to 2.165 yesterday and ran the app overnight on my work computer. When I left, it showed the system properly, decoding channels, group calls, etc. When I got in this morning it was still decoding however the system type changed from a P25 to a D-Star. The site ID was a long number ending in -1.16 which was also the P25 site number. I exited and restarted DSD+ and the data showed correctly. See pictures.

This is the Ohio MARCSip system Columbus node 1.16.

What info do you need to look into this?

Also a noob question: Is there a link somewhere that goes over decoding and following audio streams with a second SDR radio?

 

[mtindor]

Are you using FMP24 or FMPA TCP-linked to DSDPLus? Or are you using SDRSharp, SDRUno or some other SDR software for the signal source? I ask because Site 1.16 is a simulcast and you need to be using FMPA or FMP24 TCP Linked with DSDPlus in order to handle the simulcast. Otherwise, you are prone to errors. I would not call this a bug at all, but rather a situation where at some point there was poor decode.

Also, keep in mind that if you are monitoring a specific system long term, it is best to force only that decode option so that there is no chance that it would even detect P25 as D-Star or any other protocol. I would add the -f1 switch and remove -fa switch (if you have it in your commandline). That'll force P25-only decoding and will minimize any possibility of it falsely detecting some other sync pattern when the signal is bad.

 

[amcferrin90]

I was using just dsdplus.exe with the cc.bat command. I never gave the simulcast thing a thought. I'm using a SDR tuned with SDR# piping audio through VAC. I haven't gotten into audio decoding at all yet hence the second question. I tried using the TCP linking but get some error and just never fooled with that mode since. With gain the signal is -20 which may be too strong. I'll give it a shot with FMP and see what that does.

 

[mtindor]

I was using just dsdplus.exe with the cc.bat command. I never gave the simulcast thing a thought. I'm using a SDR tuned with SDR# piping audio through VAC. I haven't gotten into audio decoding at all yet hence the second question. I tried using the TCP linking but get some error and just never fooled with that mode since. With gain the signal is -20 which may be too strong. I'll give it a shot with FMP and see what that does.

Any time you are monitoring a simulcast you want to use FMP24 (if you have a dongle) or FMPA (if you have an Airspy) TCP-linked to DSDPlus. That is the only way to effectively decode simulcast. With that said, it is entirely possible that you are in an area where you get a good decode of the simulcast (because you aren't hearing competing tower sites) . So just using SDRSharp + VB Cable + DSDPlus might be okay. But the only "supported" method of copying LSM is by using FMP24 / FMPA TCP-linked to DSDPlus. Either way, if you are monitoring that site long term, you should consider adding -f1 on the commandline (and removing any other -f# parameter on the commandline) so that it will force decoding of only P25.

Are you using Fast Lane version ? Or are you using the ancient free version 1.01 from the dsdplus.com website ?

 

[amcferrin90]

I'm using the fastlane version. I signed up as a lifer which was a great deal.

When I use TCP, I get "TCP Connect error = 10061" with "no connection could be made because the target machine actively refused it". I've cleared the apps through the firewall but it seems something else is messing with it?

So I just am importing from SDR#2 because it has a better antenna, outputting now to VAC#2 then on dsdplus input from VAC#2. I'm getting data decoding without errors so I'm ok. I did add the -f1 to the command line.

Can you point me to where I can read about holding on specific groups and anything I need to do to decode audio? I would imagine something has to tell the second SDR dongle where to tune for the active frequency for the selected group if that frequency is outside the bandwidth of the decoding SDR dongle.

Thank you.

 

[mtindor]

You need to have a reference in FMP24 such as -o20001 and then a reference in DSDPlus such as -i20001 . -o for FMP24|A, and -i for DSDPlus. Same port must be referenced on both.

If you fire up FMP24|A and never fire up DSDPlus, or vice versa, then connections over TCP cannot be made. Fire them both up, referencing the same port, and it should work.

 

[amcferrin90]

So I just am importing from SDR#2 because it has a better antenna, outputting now to VAC#2 then on dsdplus input from VAC#2. I'm getting data decoding without errors so I'm ok. I did add the -f1 to the command line.

Can you point me to where I can read about holding on specific groups and anything I need to do to decode audio? I would imagine something has to tell the second SDR dongle where to tune for the active frequency for the selected group if that frequency is outside the bandwidth of the decoding SDR dongle.

Thank you.

 

[mtindor]

To hold on talkgroups, raise or lower priority, or lock them out, you basically right-click or left-click on the talkgroups in the Channel Activity window. You have to play around with left-clicking and right-clicking. And if you accidentally lock something out or set it to a priority you don't like, you can always go into the DSDPlus.groups file and edit it.

As far as your other questions about tuning, please lay out specifically (in a well explained fashion), how many copies of SDR# you have running, how many copies of FMP24 you have running, and how many copies of DSDPlus you have running.

You make sure and specify -rc on the FMP24|A and DSDPlus commandlines for the control channel monitoring. Then you make sure and specify -rv on the FMP24|A and DSDPlus commandlines for the voice channel following. If you don't use FMP24|A and use SDR# instead, then you will just be making sure those are set on the DSDPlus commandline for each.

FMP24|A.exe
-rc Role is trunk control/rest channel monitor
-rv Role is trunk voice channel monitor

DSDPlus.exe
-rc role is control/rest channel decoder
-rv role is voice channel decoder

fmp24 -h / fmpa -h and dsdplus -h are helpful. So are ? in the FMP and DSDPlus windows main windows.

 

[mtindor]

So I just am importing from SDR#2 because it has a better antenna, outputting now to VAC#2 then on dsdplus input from VAC#2. I'm getting data decoding without errors so I'm ok. I did add the -f1 to the command line.

Can you point me to where I can read about holding on specific groups and anything I need to do to decode audio? I would imagine something has to tell the second SDR dongle where to tune for the active frequency for the selected group if that frequency is outside the bandwidth of the decoding SDR dongle.

Thank you.

Per DSDplus.txt:

Channel Activity Window:
Left click on priority Increase traffic priority
Right click on priority Decrease traffic priority
Left click on target Increase priority override
Right click on target Decrease priority override
Left click on headings Clear all lockouts
Right click on headings Clear hold

Also, you can use the Spacebar to toggle Hold / Unhold on a talkgroup that is active.

 

[amcferrin90]

My input/outputs from DMP/DSDPlus are:
i1 = 'Line 2 Virtual Audio Cable' [input from FMP]
i2 = 'Line 1 Virtual Audio Cable'
o1 = 'Speakers'
o2 = 'Line 2 Virtual Audio Cable' [output from FMP to DSD+]
o3 = 'Line 1 Virtual Audio Cable'

I have two SDR tuners so using -i2 on FMP selects the second tuner. I read that -i# selects a SDR tuner not an audio pipe.

This is my FMP string: "FMP -i2 -o2 -P-56.0 -b12.5 -f774.19450" I should add -rc to that. By doing this it's actually pulling straight from the SDR_Tuner2 without SDR# at all. I had to offset the frequency to get the actual control channel center, I'll need to play with the correction settings to make it right but it tunes 774.28125 dead center so I'm happy with it.

This is my DSDPLus string: "DSDPlus -f1 -rc -i1 -o1 -v1 -wsl400.2 -wss100.200 -wel0.1 -wcl0.445 -wes400.520 >>CC.log" which has the correct role. I added the -o1 so I would hope decoded audio goes out to the speaker. I don't hear anything. I don't see anything to indicate which group would be coming out of the speakers either.

 

[dave3825]

FMP -i2 -o2 -P-56.0 -b12.5 -f774.19450

DSDPlus -f1 -rc -i1 -o1 -v1 -wsl400.2 -wss100.200 -wel0.1 -wcl0.445 -wes400.520 >>CC.log"

A very basic batch file for tcp. This eliminates the need for any vb cable or virtual audio cable.

:FMP24-CC.bat:
FMP24 -i2 -o20001 -v3 -f852.7375

:CC.bat:
DSDPlus -rc -i20001 -E -v3 >>CCsuffp25.log

Then

:FMP24-VC.bat:
FMP24 -i1 -v4 -o20002 -P0.0 -rv

:VC.bat:
DSDPlus -rv -Pwav -i20002 -v3 >>VCsuffp25.log

The -o20001 in first fmp24 outputs to tcp port 20001.
The -i20001 in first dsdplus opens tcp port 20001

The second fmp24 outputs to port 20002
The second dsdplus opens tcp port 20002

 

[amcferrin90]

Ok So something else I missed and learned in the threads is that I should be using FMP24 not FMP. Now FMP has been working ok. When I send the same string using FMP24, decoding slows to a crawl, the gain on the SDR tuner is 30 dB higher, in the decoding window is a LOT of CRC errors. I'm using an i7-3770 8-core machine with 8 GB of RAM, 50% available. The only difference I saw was that FMP24 wants to use 2.4 Mhz of bandwidth while the default for FMP is 1.0 MHz. When I run FMP24 my overall CPU usage goes from 4% on 1 core to 9% on 4 cores. Thoughts?

 

[mtindor]

Ok So something else I missed and learned in the threads is that I should be using FMP24 not FMP. Now FMP has been working ok. When I send the same string using FMP24, decoding slows to a crawl, the gain on the SDR tuner is 30 dB higher, in the decoding window is a LOT of CRC errors. I'm using an i7-3770 8-core machine with 8 GB of RAM, 50% available. The only difference I saw was that FMP24 wants to use 2.4 Mhz of bandwidth while the default for FMP is 1.0 MHz. When I run FMP24 my overall CPU usage goes from 4% on 1 core to 9% on 4 cores. Thoughts?

Your machine is beefy enough. You really should benefit from using FMP24. Yes, because it's dealing with significantly more spectrum (approx 2.4 times as much) it would make sense to me that it uses more CPU. I never gauged the CPU usage between the two. It may using the amount of CPU that the author expects, or maybe not. But it sounds reasonable to me.

Because it is working with a larger slice of spectrum, there is also a risk of having [unwanted] signals of much higher strength within that passband. And any such signals would cause the noise floor to change. It could be noisy with FPM24 because of strong signals within that 2.4 mhz passband that weren't present in the 1 mhz passband.

Use 'g / G' in FMP24 to adjust your gain up and down until the waterfall looks clean. It's very easy to tell when things start to get dirty (front end starts to overload, images start to appear, etc). Lower the gain enough so that doesn't happen, and you should still have enough gain on a signal to work with it. If in your particular case you don't have enough gain to copy the wanted signal well, you might have to switch back to FMP. but do also make sure you are using a proper bandwidth setting ( press "b" in FMP24 to toggle between them) for the signal you are monitoring. P25 = 9.5 (or at most 12.5). DMR / NXDN96 = 7.6. NXDN8 = 4.0. For analog, use whatever sounds best.

 

[slicerwizard]

Ok So something else I missed and learned in the threads is that I should be using FMP24 not FMP. Now FMP has been working ok. When I send the same string using FMP24, decoding slows to a crawl, the gain on the SDR tuner is 30 dB higher, in the decoding window is a LOT of CRC errors. I'm using an i7-3770 8-core machine with 8 GB of RAM, 50% available. The only difference I saw was that FMP24 wants to use 2.4 Mhz of bandwidth while the default for FMP is 1.0 MHz. When I run FMP24 my overall CPU usage goes from 4% on 1 core to 9% on 4 cores. Thoughts?

Your CPU load is negligible, don't worry about it. You must use FMP24 (not FMP) TCP linked to FL DSD+. FMP24 must be using 9.5 or 12.5 kHz bandwidth (press B to adjust) when monitoring P25 signals, especially simulcast/PSK modulated signals. FMP24 RF gain should be as high as possible without clipping, i.e. no carriers reaching the very top of the spectrum window and no extra artifacts/spikes rising from the noise floor. Press g and G to experiment with RF gain levels.

That's how you start with a solid foundation. I assume you have two dongles? If so, one is for control channel decoding, the other for voice channel decoding. Each dongle needs a running copy of FMP24 and DSD+.