Minor updates on the AoIP front
Intel i210 NICs are the usual "network clock" tool for sub-ms precision with AoIP
Apparently not required, or maybe only on the Master Input node
but RPi5 built-in NIC may handle PTPv2 grandmastering on its own.
Might greatly simplify if using stock PCIe cards isn't required.
RPi4 "CM4" type hosts nodes are likely fine for 2-3 of the "Output Slave" nodes, that would save hundred$
...
Only the output node converters get sync'd via BNC Word Clock.
No, 8ch ADC on "Master Input" RPi node should also be BNC word clocked, in fact likely Master.
Thus ALL analog converter devices are part of the BNC WC sync.
Except maybe the mono trueSub group, a RPi5 + HiFiBerry 8ch DAC is IMO good enough since all stay under 80Hz, shoutout
@Keith_W
likely PipeWIre native IIR processing filters created by MSO I think, but maybe some FIR if needed as well convolved by CamillaDSP or Hang Loose.
Yes I know, I stated all DSP by "Master Input" node but I'm thinking this group is so separate, can be modular, easy to switch inputs or disable completely while experimenting with Stereo Bass and maybe even envelopment.
...
> So far pipewire-aes67 is leading contender, much cheaper than interface-based AVB even secondhand, trying to keep under $20-25 per analog port for the interfacii.
Since all the RPi nodes' realtime System Clocks will be kept in sync from the PTPv2 network grandmaster, together with near-100% coverage by BNC WC, it looks like AES67 would just be redundant, unnecessarily complex.
Both Gstreamer and PipeWire have native RTP modules, a much simpler, leaner & cheaper transport.
Favouring Gstreamer for now, since it's less likely to try to use ASRC resampling in response to each node's SysClock constant skew-adjusting from phc2sys PTPv2 disciplining.
Also GSASysCon is a tool I may want to test
@charlielaub
...
any constructive feedback welcomed
