Distributing MIDI Clock, Sync and Timing

[SteveECrane]

My apologies if this has already been covered - I couldn't find anything relevant - or if this isn't the correct forum; however...

The biggest issue I've got - and have always had - with my studio setup is distributing some form of time-based sync function across all my synths, keyboards, modular and effects, so that the arps, sequencers and pattern generators, et al, are all in time, always!

So, I am renewing my investigation into the best way to go about distributing MIDI Clock - or whatever timing protocol(s) I need to deploy - throughout my entire setup, hence this post!

The Details
I seldom - if ever - move my studio gear around (save for when I play a very simplified setup 'live'); however, I want to be able to be able to sit down at ANY device (MIDI controller keyboard, sequencer, pattern generator, arpeggiator or synth) and have everything else synchronised to them when I press 'play' (or start arpeggiating on a synth!).

That means either my Conductive Labs NDLR, Arturia's BeatStep Pro, Arturia's KeyStep 37, my Vermona Melodicer, Pamela's New Workout, my Doepfer Darktime, Korg SQ-1, Behringer 182 (via a Doepfer A190-4), or my (Cubase) DAW - when I'm using it - puts out appropriate clock/timing info (usually MIDI 24PPQN MIDI Clock) which is distributed appropriately whenever any one of these devices are actually running.

[Note: Unfortunately, the NDLR appears to put out MIDI Clock even when it's not actually playing and requires menu diving in order to turn it off and back on - something I'm loathe to do as I lose any creative spontaneity - whereas the Melodicer module is never, ever off, which is why I slave it to incoming CV timing info via Pamela's!]

I currently have everything (including Pamela's and my Arturia DrumBrute!!!) synchronised to my NDLR via an ageing (and faulty!) parallel port MotU MIDI Express XT, successfully operating in MIDI "Merge All" mode for ALL of its 8x8 MIDI ports ... but I have got to come up with a better solution in order to avoid inevitable MIDI loops and to reduce the sheer amount of MIDI sync timing data flowing around every, single device in my setup!

I'm seriously considering investing in a Conductive Labs MRCC - to replace my old MotU XT - simply so that I can create an appropriate distribution matrix which can be recalled at the flick of a switch.

There are currently no MIDI loops as every device is configured and connected either as a MIDI input devices or MIDI output device (essentially in a star fashion from the MotU Express XT) - with either a MIDI In or Out jack connected to it - with absolutely no device acting as both. No device acts as a MIDI thru - myriad Kenton boxes perform this function. A few CV-based devices are (re-)patched appropriately to send or receive their timing info.

There are no innate issues causing fundamental problems with this setup, apart from the following usability restriction and the sheer volume of MIDI data being distributed:-

1) The fact none of my synths (or MIDI controllers) currently receive any MIDI in data, so they cannot be sync'd to a single master clock without causing MIDI loops;
2) The amount of MIDI Clock timing info being sent to (and received by) all connected devices.

So, any ideas, folks?  Can the MRCC do all that I need? What are the issues / limitations, if any? 
Thoughts and constructive comments welcomed, encouraged and much appreciated.

[Additional Notes:-
MIDI gets into my Eurorack setup / modular environment via a Doepfer A192-2 and out of it via the aforementioned A190-4.
MIDI gets into and out of my DAW via a USB3-connected ESI Audio M8U eX, alongside the aforementioned MotU XT coupled to myriad Kenton Thru and Merge boxes which feed the M8U via a single port.
I try to only ever use MIDI over USB for firmware updates, initial setup/config matters, and sending SysEx patch info to individual devices, disconnecting them at source in order to avoid the inevitable ground loop and graphics whine issues, plus the ensuing confusion at the DAW end.]

By-the-way, apologies that this is so long but I realised most parts required a little bit more 'clarification' after I'd initially finished writing, solely in order to communicate exactly what I've got and what I'm trying to achieve. Sorry 'bout that!  Hopefully not a TL;DR post...!?

Cheers,

Steve.

PS Am I correct in my observation that the NDLR continues to generate MIDI Clock info even when it is not 'playing'? Disabling its Internal Clock setting (or simply pulling out the appropriate MIDI jack at the NDLR end!) removes this data stream, so it's definitely originating at the NDLR even when it's not running. Is this operation correct / normal and, if so, why and, if not, can it be configured differently?

PPS  This apparent clocking issue is why I sync my entire setup from my NDLR.

[Dark Waves]

Just one thought (and you might want to look over the manual for other aspects pertaining to your configuration) is that you could use presets on the MRCC to quickly reconfigure which source and destination combinations are active at any given time.  They're fairly simple to create and recall, with the caveat that you have to do all operations from the front panel, as there is no app to remote control the device.  Still, it's easy.  Also, depending on how many things you have to connect, it's possible to chain two MRCCs together.

[RadekPilich]

You might use presets and / or use MRCC as master and control its BPM / START/ STOP from some midi device via CCs / Prog changes. 

I wouldn't attempt this without controlling MRCC externally. 

Unless you use MRCC as master, you would always have to choose appropriate preset manually (via PCh) when you change your station.

[Darryl]

Regarding The NDLR clock, yes it's continually on, as it "should be" according to the MIDI spec. It's a choice, and Arturia for one, did it the other way, starting and stopping clock so we understand why people might think that's the way it should be. The intention from the MIDI spec perspective is that the clock is always running, and Start/Stop/Continue message control what's playing or not. These transport controls get things synced up, where not everything is necessarily going to sync on a clock start. Another benefit of leaving the clock running is when you have instruments and effects slaved to the clock, they can abruptly stop when the clock stops. For instance, you have a Dave Smith Instruments Evolver, it's synced to clock and playing along with it's sequenced effects. If you stop the clock, it abruptly stops the effect, when maybe all you wanted to do was stop your KeyStep sequence.
So, when the NDLR is slaved to an external clock, it's optimized for the "Arturia way". You arm the NDLR parts, and they start when the clock starts. We hope the choice provides a solution for both a continuously running clock, or a clock that start and stops. Having said that, it's a lot more fluid to operate The NDLR when it's using Internal clock.