OK, so things are pretty much as I thought they were. Just wanted to make sure I wasn't missing anything should I decide to redesign this system. I'll probably keep things as-is for now, then add another private node for P25.
toggle quoted messageShow quoted text
On 11/25/2020 7:51 PM, Steve N4IRS wrote:
you have D-Star, DMR and AllStar.
If you used AllStar to mix you could keep the number of vocoders to two and only have one transcoder. For the audio, you would have 2 private nodes. The first node would be DMR and pointed at TGIF. The second node would be D-Star and pointed at XLX. As you said in your first message, you issue would then be metadata.
On 11/25/20 7:32 PM, Patrick Perdue wrote:
I'm not great at writing out flows.
My setup currently doesn't use ircDDBGateway, since that's all handled at the XLX reflector end. So, it goes like this.
Allstar <-> AB <-> HBLink <-> MB <-> XLXD (DMR, since XLX is also handling a BM interlink and YSF traffic) <-> ambed (connected to the XLX reflector) <-> TGIF.
I'm sure I screwed that up.
So basically, native AMBE traffic is transcoded to analog using the MD380-emu vocoder back to Allstar. DStar is transcoded through the pair of ThumnDV's on the reflector, which is then sent to MD380emu and then Allstar. Between DMR or YSF and DStar, this is fine, but from Allstar to DStar, it's going through another AMBE transcode before A_B sees it, so more latency and lower quality audio.
If I connect to the reflector directly as DStar, then use ASL to connect to the other modes, perhaps on another module of the reflector and HBLink, then DMR-YSF would only see the call specified in A_B, correct?
On 11/25/2020 7:18 PM, Steve N4IRS wrote:
I don't see how there are 2 vocoders to get from AllStar to D-Star. The flow would be:
AllStar <-> AB <-> MB <-> ircDDBGateway <-> XLX for D-Star.
You could drop the ircDDBGateway and go to XLX DMR.
On 11/25/20 7:14 PM, Patrick Perdue wrote:
Sure, but I'm currently going through two of them to get from Allstar to DStar.
I'm just wondering if there's a way to drop one of them out and still maintain metadata. I could directly connect A_B and MMDVM_Bridge to the XLX reflector and only direct DStar traffic to the module, and audio between DStar and the other digital modes should be about the same as it is now, but audio from Allstar to DStar has one less generation of vocoding, but then I lose metadata between the digital modes. At least, this is how I understand it.
On 11/25/2020 6:55 PM, Steve N4IRS wrote:
Either way, you have to go through a vocoder to go from analog to D-Star Same for P25, you still have to go through a vocoder.
On 11/25/20 6:28 PM, Patrick Perdue wrote:
I have a system that incorporates ASL, XLX, HBLink and DVSwitch. I'm toying with the idea of adding P25.
Currently, DStar is handled by XLX and a pair of transcoders connected to the XLX reflector, not DVSwitch. I would like to connect ASL directly to the XLX reflector in DStar mode, so that the audio to and from DStar to ASL is not transcoded through AMBE, but still pass metadata to DStar from the other digital modes. Is this at all possible? Similar question in regards to P25.
Basically, I want to do the least amount of transcoding possible to and from analog, while maintaining the greatest compatibility across digital modes. What I have now works, but I'd love to get rid of that extra layer of transcoding between analog and DStar without sacrificing metadata.