P25RX-II P25RX-II Firmware Updates / Testing

[maus92]

Are there any secrets to getting the BT adapter to connect to car audio systems? Thus far, my Tesla MY sees the P25RX2 and seems to pair with it, but then the connection process stalls on the car side. This is with the 11-28 firmware. Just a note that I can connect BT audio when I use a laptop computer as an intermediary device (the car completes pairing successfully with the Dell) - something I want to avoid.

Note that it doesn't pair with my old OontZ Angle 3 Plus BT speaker either, but the 3.5mm audio sounds good.

 

[freqseeker]

I know BT with cars varies from brand to brand. My Ford Sync 3 doesn't work. My Oontz Angle works great with it.

 

[btt]

New version available. 11-29_1827. Starting with this version there is a new feature called "RX Model". There are currently 2 models. More will be added in future versions. RX Model refers to the DSP / bit-decision code and not the hardware itself. RX Model 0 (default) is very close to the code in the previous release, but there is an improvement. RX Model 1 may be a better choice for some systems. I'll attach an image showing the symbol plot with RX Model 1 selected while monitoring a local system. I'm thinking that the default value for P1 sync threshold (value of 4) may be too high for a default value. 2 may be a better default for non-VHF systems. It is a value you may want to make 0 if you have a good signal and you are not mobile
.

 

[Vermonster]

Confirming the prior 2226 was excellent on all 3 Con+ systems I tried it on.
Only the 900mhz system in a fringe reception area required manually tweaking the LNA gain for optimal reception.
A note to others playing with Con+, you want to enter each frequency on the DMR tab 1 tab in the correct order ONE TIME for each frequency. You do not want to enter a frequency twice (e.g. once per slot) or skip numbering (as in DSD+FL Frequencies file)
###
Regarding new 2021-11-29 20:50.
Hung on update with fast/slow lights; resolved with device unplug, hold button on replug then release after 10 seconds
After install, fully lit RF Sig Level graph issue identified; fixed with restart

Model selection clearly impacts Signal Insights view on at least one Con+ test system. I will test several this morning.

Personal request to implement save/backup for DMR systems when you get a moment. It would make testing on multiple systems just a bit easier.

Thanks

 

[btt]

Confirming the prior 2226 was excellent on all 3 Con+ systems I tried it on.
Only the 900mhz system in a fringe reception area required manually tweaking the LNA gain for optimal reception.
A note to others playing with Con+, you want to enter each frequency on the DMR tab 1 tab in the correct order ONE TIME for each frequency. You do not want to enter a frequency twice (e.g. once per slot) or skip numbering (as in DSD+FL Frequencies file)
###
Regarding new 2021-11-29 20:50.
Hung on update with fast/slow lights; resolved with device unplug, hold button on replug then release after 10 seconds
After install, fully lit RF Sig Level graph issue identified; fixed with restart

Model selection clearly impacts Signal Insights view on at least one Con+ test system. I will test several this morning.

Personal request to implement save/backup for DMR systems when you get a moment. It would make testing on multiple systems just a bit easier.

Thanks

DMR Backup/Restore has been added to version 11-30_0458.

 

[pcman67]

New version available. 11-29_1827. Starting with this version there is a new feature called "RX Model". There are currently 2 models. More will be added in future versions. RX Model refers to the DSP / bit-decision code and not the hardware itself. RX Model 0 (default) is very close to the code in the previous release, but there is an improvement. RX Model 1 may be a better choice for some systems. I'll attach an image showing the symbol plot with RX Model 1 selected while monitoring a local system. I'm thinking that the default value for P1 sync threshold (value of 4) may be too high for a default value. 2 may be a better default for non-VHF systems. It is a value you may want to make 0 if you have a good signal and you are not mobile
.View attachment 112931

So, the RX models is sort of like a profile that you can select (or create and select) for different environments (e.g. one for mobile scanning and one for stationary scanning (home) or perhaps profiles for different systems?

If so, I like that idea! Looking forward to downloading and testing it later this morning when I get a chance.

 

[btt]

So, the RX models is sort of like a profile that you can select (or create and select) for different environments (e.g. one for mobile scanning and one for stationary scanning (home) or perhaps profiles for different systems?

If so, I like that idea! Looking forward to downloading and testing it later this morning when I get a chance.

I plan on adding profiles in future versions. The "RX Model" selection selects what type of pre / post symbol-synchronization code is run prior to decoding the demodulated symbols. For example, try selecting "RX Model 1". In this RX model, only the phase of the incoming symbols are used in the bit decisions. Symbol magnitude is always on the unit circle. This will be obvious when viewing the symbol plot on the signal insights tab. The reason for adding the "RX Model" option is for maintaining consistency. Multiple people asked me not to change the DSP code from the alt-testing version (because it was working very well for them). Adding the idea of RX Models allows running different DSP code for different systems / situations. Eventually, the RX Model selection could be associated with a scanning-configuration profile.

 

[kruser]

Hi Todd,

Are new updates for the original P25RX using BTConfig now a dead thing?
I've not seen any updates for the original model since the 16th of this month but have seen several in this thread for the P25RX-II.
It seems the original P25RX has been abandoned although I did see mention of it in one of the recent P25RX-II firmware notes.

Thanks

 

[maus92]

Hi Todd,

Are new updates for the original P25RX using BTConfig now a dead thing?
I've not seen any updates for the original model since the 16th of this month but have seen several in this thread for the P25RX-II.
It seems the original P25RX has been abandoned although I did see mention of it in one of the recent P25RX-II firmware notes.

Thanks

I would say that Todd is developing the firmware to dial in the performance for the P25RX-II hardware design, rather than neglecting the former. He probably doesn't want to produce more new devices until the P25RX-II is performing satisfactorily for most users.

 

[btt]

Hi Todd,

Are new updates for the original P25RX using BTConfig now a dead thing?
I've not seen any updates for the original model since the 16th of this month but have seen several in this thread for the P25RX-II.
It seems the original P25RX has been abandoned although I did see mention of it in one of the recent P25RX-II firmware notes.

Thanks

No, the P25RX updates are not dead! Sorry for the delay, but the next P25RX update is going to be a massive update. It will take me some time to complete. I'm trying to maintain things the best I can in order to survive. I only sold 4 receivers in the past month (if that). As you might expect, I need to sell more than that to survive financially. I am currently trying to get the firmware for the model that I'm currently selling to acceptable performance for everyone. I believe it is there now with the exception of 2 people that are still having issues. I'm planning to resume selling P25RX-II hardware this week. After I feel confident that I can keep paying the bills, I will circle back and incorporate updates from the P25RX-II into the P25RX. I am not going to abandon it.

 

[kruser]

No, the P25RX updates are not dead!

Thanks for the update! I've really enjoyed using the original P25RX.
Recent sudden issues on my end have prevented me from being able to participate much with the original model plus put a halt on my plan to purchase the P25RX-II for the time being. Hopefully things will turn around soon.

Hang in there as you have a great product!

 

[Vermonster]

DMR Backup/Restore has been added to version 11-30_0458.

When exactly do you sleep?
Testing now.

Sync indicator is fixed.

With the caveat that reception in the 900mhz band has been horrible this morning on RX-II, DSD+FL dongle and scanner; somewhat improved now (maybe some frost melted):
400mhz Con+ systems working very well and any difference between model 0 and 1 is incredibly subtle; probably very slight edge to model 0.
On 900mhz Con+: When a voice call is decoded, the quality is excellent. But, occasionally sync is lost for a few seconds. I *think* this is more common on model 0. The loss seems to always follow a voice call (but most voice calls-->CC are fine).

Screenshots during CC and VC attached.
Relevant portion of console log showing where "No Control Channel on LCN1" is show, but the CC is actually on LCN1, and when the system retunes to LCN1 it successfully identifies it:

->(VOICE), freq: 939.462500 sig 1 rssi: -108 dBm Slot: 1 TGroup: 240, Desc: (unknown tgroup), $
(voice paused)
rssi: -107 sa 1 ue 1 freq 938.387500 $
rssi: -108 sa 1 ue 1 freq 938.387500 $
rssi: -107 sa 1 ue 1 freq 938.387500 $
rssi: -107 sa 1 ue 1 freq 938.387500 $
rssi: -108 sa 1 ue 1 freq 938.387500 $
rssi: -108 sa 1 ue 1 freq 938.387500 $
rssi: -107 sa 1 ue 1 freq 938.387500 $
rssi: -108 sa 1 ue 1 freq 938.387500 $
rssi: -108 sa 1 ue 1 freq 938.387500 $
rssi: -108 sa 1 ue 1 freq 938.387500 $
rssi: -108 sa 1 ue 1 freq 938.387500 $
rssi: -107 sa 1 ue 1 freq 938.387500 $
rssi: -107 sa 1 ue 1 freq 938.387500 $
rssi: -108 sa 1 ue 1 freq 938.387500 $
rssi: -107 sa 1 ue 1 freq 938.387500 $
rssi: -108 sa 1 ue 1 freq 938.387500 $
set freq: return cc
setting freq to 938387500
sample rate set to 24000.0
DMR, sys_id 0x008b freq: 938.387500, sig 1 rssi: -109 dBm $
DMR: No Control Channel on LCN1.
DMR: Trying LCN1.
setting freq to 938387500
sample rate set to 24000.0
set freq: cc change
DMR, sys_id 0x008b freq: 938.387500, sig 1 rssi: -110 dBm $
DMR 2021-11-30-08:10:25 DATA_SYNC: tsbk_ps 13, sys_id 0x008b freq: 938.387500, sig 1 rssi: -109 dBm $

More conlog: https://pastebin.com/raw/XjQvufd2

 

[maus92]

Reporting back about my Bluetooth issue with the Ooantz Angle 3 Plus - I was able to get the devices to pair. I had to disable all the other Bluetooth devices / on the computers in my office and reset the Ooantz device (the Angle 3 can be finicky.)

 

[btt]

I took the latest version for a test drive this morning (11-30_0458). After a reset-to-defaults, the only change made was to enable Bluetooth, roaming mode, and change P1 sync threshold to 2. I can definitely say that RX Model 0 (the default) is working as good or better than any previous version in a mobile simulcast environment. Note this release does not reset the configuration to defaults, so you may want to do that manually if you have changed any of the advanced settings in previous versions.

 

[shirsch101]

Todd,
I am switching between RX0 and RX1, this is awesome, either model is far way better than anything else!!!!!

 

[maus92]

I'm not sure if there were any changes in the firmware between 11-28 and 11-30 wrt Bluetooth, but I was also able to pair with the Tesla today as well as the Ooantz mentioned earlier. But it sounds better coming through the Ooantz than the factory audio system, so I will do that. I do think I need to boost the Bluetooth audio out level a bit - I think there is a command for that.

 

[btt]

I'm not sure if there were any changes in the firmware between 11-28 and 11-30 wrt Bluetooth, but I was also able to pair with the Tesla today as well as the Ooantz mentioned earlier. But it sounds better coming through the Ooantz than the factory audio system, so I will do that. I do think I need to boost the Bluetooth audio out level a bit - I think there is a command for that.

There were no changes related to Bluetooth. Sometimes pairing multiple times can help. Yes, you can change the Bluetooth audio gain manually with the following commands:
$ bt_gain 1.3 (default value 0.8)
$ save (save to flash)

I will get this added to the GUI. Thanks for the update.

 

[W0RS]

Has there been any testing done on the VHF DMR Cap+ system? The local PD and Fire just switched over to a Mototrbo Cap+ system....

 

[Vermonster]

Has there been any testing done on the VHF DMR Cap+ system? The local PD and Fire just switched over to a Mototrbo Cap+ system....

I have tested prior firmware on Cap+ VHF (150mhz) and UHF (450-470 and 900 mhz) with generally good success. I will try to test the current firmware this morning and report back. Many school buses here use Cap+ so I should have good traffic.

 

[btt]

There is a new version 12-05_1612 available. RX Model 3 has some major changes in this version. If you are one of the remaining people having issues with P25 reception, please give this version a try with default settings first. (manually reset to defaults). You can set the BB VGA gain to a fixed value with this version (try 30-36 dB). This sets the threshold level for when the LNA gain start to be adjusted. Probably best to leave the LNA and Mixer gains to automatic for testing. In a future version, I will add something to aid in determining how to set these values. For reference, with the AGC Target set to 60dB, the LNA and Mixer gains set to auto, and an input signal of -95 dBm, try to adjust the BB VGA gain such that the LNA gain is around 20dB. The other major changes for rx-model 3 are related to frequecy/phase error correction. Hopefully this is the improvement that is going to resolve the remaining issues (I have good reason to believe this). I was able to play back some of the worst IQ files with much more decoded audio. Because the IQ files I currently have already were frequency corrected with prior methods, the files are not a complete indication of how this version will operate. Mobile and fixed testing was done with this version. Fixed testing was done with multiple receivers. It is working excellent here.