• Registered: 2021-07-18
  • Posts: 3

Topic: 12 mic - MADI Daisy Chain or Star

I recently had a recording where I needed 2 12mics. I tried both scenarios, which both worked. RME clearly recommends daisy-chaining, and I wonder why is that?

1. Star-Configuration:
12mic-1 was connected via MADI-Coax to 12mic-2 (both ways) -- 12 mic-2 was connected to 12mic-1 as stated and to MADIface Pro via MADI-optical. Both 12-mics had an additional Wordclock between each other, with the 12mic-2 being the master.
Benefits: clear clockmaster for the optical MADI-Stream, only one generated MADI-Stream. If 12mic-1 fails, I have all the signals from 12mic-2 and no interruption of the recording, just loss of some channels.
Disadvantages: no MIDI-over-MADI for 12mic-1 (why is that? Are SysEx-messages not transferred between OPT and Coax?). Not recommended by RME (and also supposedly impossible, if I read the manual correctly?)

2. Daisy-Chain (recommended by RME)
MADI-loop of the 3 devices. 12mic-1 and 12mic-2 still connected via Wordclock, but the 12mic which is closest to the MADIface Pro as Clock-Master.
Benefits: AVB-remote worked for both devices
Disadvantages: routing is a little bit more clunky, as I had to make sure which channels to "pass-through" and which to occupy and couldn't just connect MADI 1-24 of the last 12mic to the MADIface. Also when one device fails, the whole loop breaks, leading to a total loss.

Why is configuration 1 not recommended by RME? IMO, this is effectively a 2-point connection from the perspective of the MADIface Pro. Will it be possible to pass through MIDIoMADI between the connections in the future?

2 Reply by ramses (edited by ramses 2023-07-27 16:11:55)

  • ramses
  • DAW Mastermind
  • Offline
  • From: UTC+1
  • Registered: 2012-10-14
  • Posts: 13,229

Re: 12 mic - MADI Daisy Chain or Star

RME approach (case #2) is more efficient, as it has fewer hops/delay over MADI.
There is a delay of 3 samples for routing packets through a device via MADI at single speed, see manual.

Your use case #1 results in 3+3+3 = 9 samples delay

    BBF Pro->12Mic#2->12Mic#1->12Mic#2->BBF Pro
                          3               3               3

RME's recommendation #2 results in only 3+3 = 6 samples delay

   BBF Pro->12Mic#1->12Mic#2->BBF Pro
                       3                3

You could organize your setup this way:

clock master          clock slave                  clock slave
BBF Pro ------>------12Mic#1-------->--------12Mic#2
    |                                                                 |
    +-----------------------<--------------------------+


12Mic with SteadyClock FS are the perfect clock slaves and will remove jitter from the incoming clock signal from MADI efficiently. This might also work for changes between 88.2 and 96 kHz at double speed, but not between single/double speed (higher bandwidths/sample rates based on port multiplexing cannot be recognized).

With the BBF Pro as clock master you can use the application to change between e.g. 44.1/48 kHz and the 12Mic will follow the sample rate.

Word clock will bring no benefit over the clock distribution via MADI, Steadyclock FS will remove any clock jitter efficiently. You can save the money for this additional cabling.

The routing is fully normal .. channels that only need to pass through the device need to be routed from input to output thats all.

Other advantages:
- galvanic isolation by using only MADI optical and
- as you said, could be the case that MIDI over MADI only works this way but no bridged between optical and BNC
  (never tried that).

BR Ramses - UFX III, 12Mic, XTC, ADI-2 Pro FS R BE, RayDAT, X10SRi-F, E5-1680v4, Win10Pro22H2, Cub14