• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Mixing multiple consumer S/PDIF devices - Clock sync issues?

carat

Member
Joined
Dec 25, 2021
Messages
41
Likes
18
The MOTU LP32 is a digital mixer with 4x TOSLink inputs and 4x TOSLink outputs. All of these can be configured as stereo S/PDIF.
LP32.jpg

I would like to use it as a mixer for 4x stereo consumer S/PDIF devices (like PCs and gaming consoles) and output to 2x stereo consumer S/PDIF devices (a DAC and a Bluetooth transmitter).
map.PNG

Will this work, or am I going to have clock synchronization issues?

I don't really understand how clock synchronization works when multiple S/PDIF sources are mixing into one device. Does clock synchronization only matter between each source and input, or do all the sources need to be synchronized to one master clock (which will be impossible since these consumer source devices have no dedicated word clock inputs/outputs)? Also, will the source->mixer->output device sequence be a problem, since that's three S/PDIF devices daisy-chained?
 

LTig

Master Contributor
Forum Donor
Joined
Feb 27, 2019
Messages
5,760
Likes
9,442
Location
Europe
What does the manual say about this use case?

Edit: according to chapter Synchronisation on page 42ff you'll need word clock - as far as I understand it.
 
Last edited:

HCT-5808

Member
Joined
Feb 10, 2020
Messages
42
Likes
26
This switch should also do the job too, I think. Cheaper and I recall it was very highly rated by one of the RME folks.
 
OP
C

carat

Member
Joined
Dec 25, 2021
Messages
41
Likes
18
What does the manual say about this use case?

Edit: according to chapter Synchronisation on page 42ff you'll need word clock - as far as I understand it.
I read that and came away with the same impression. However, I'm so ignorant on this subject that I still wasn't confident either way, so I thought I'd ask here.
 
OP
C

carat

Member
Joined
Dec 25, 2021
Messages
41
Likes
18
This switch should also do the job too, I think. Cheaper and I recall it was very highly rated by one of the RME folks.
I need a mixer, not just a switch. Thanks for bringing that switch to my attention though, might come in handy in the future.
 

KSTR

Major Contributor
Joined
Sep 6, 2018
Messages
2,690
Likes
6,013
Location
Berlin, Germany
Mixing is not possible with this device as in your case all your sources will not be in sync (and cannot be made to be in sync as they don't have word-clock or SPDIF inputs). To be able to handle non-synced signals it would need to have an asynchronous resampler at each optical input to get them all in sync at the same clock.
 
OP
C

carat

Member
Joined
Dec 25, 2021
Messages
41
Likes
18
Mixing is not possible with this device as in your case all your sources will not be in sync (and cannot be made to be in sync as they don't have word-clock or SPDIF inputs). To be able to handle non-synced signals it would need to have an asynchronous resampler at each optical input to get them all in sync at the same clock.
Thanks for confirming that. I'll abandon the LP32 idea then.
I believe a product like the RME Digiface USB does what you want. It can sync to multiple spdif inputs. I've never used it, just browsed its literature. I think you need one of the PCs running the mixing software. You should talk w/ RME or a dealer or someone here who's actually used it. It will introduce some delay for the re-clocking, but I think the mixing is done in hardware controlled by the software, so it's not too bad (again, I've never used it, but that's my understanding from the literature).

I'm sure there's other products too, but I don't know them.
Thanks, I'll investigate the Digiface.
 

Atanasi

Addicted to Fun and Learning
Forum Donor
Joined
Jan 8, 2019
Messages
713
Likes
792
I think Digiface USB requires synchronized inputs. Mutec MC4, on the other hand, supports asynchronous resampling:
Integration of non-synchronizable devices, like musical instruments, DAT players, MD recorders or DAWs into digital studio environments
MC4 converts the inputs to ADAT, and you would need an audio interface to process that. It also uses AES or coaxial S/PDIF, not optical.
 
Last edited:

Morla

Senior Member
Forum Donor
Joined
Mar 15, 2020
Messages
325
Likes
302
Location
Europe/Germany
I use the LP32 for mixing multiple 44.1, 48 toslink inputs without problems.

The ADAT Ports are switched to simple Stereo. Clock set to internal.

Sources are A LG OLED TV, USB Interface attached to nvidia shield and tc visualizer.

Output is sent to 3 devices at the same time. Kii kontrol, dante bridge, minidsp shd studio
 
Last edited:

KSTR

Major Contributor
Joined
Sep 6, 2018
Messages
2,690
Likes
6,013
Location
Berlin, Germany
I use the LP32 for mixing multiple 44.1, 48 toslink inputs without problems.

The ADAT Ports are switched to simple Stereo. Clock set to internal.

Sources are A LG OLED TV, USB Interface attached to nvidia shield and tc visualizer.

Output is sent to 3 devices at the same time. Kii kontrol, dante bridge, minidsp shd studio
Probably you are just not noticing the existing problems as the manual (on pgs. 42 ff) is very clear about the requirement of all sources running on the exact same clock instance, quote "Your success in using the optical I/O ports on the LP32 depends almost entirely on proper synchronization."

I would suspect that a slight clock mismatch results in samples occasionally either being dropped or repeated, depending on the offset to the internal clock rate. This is a simple and effective strategy as those dropped or repeated samples are not readily audible unless it really occurs all the time (many many times per second).
But of course we don't know how the device handles non-synced clock rates.
 

dwkdnvr

Senior Member
Joined
Nov 2, 2018
Messages
418
Likes
698
I also find it odd that it would 'work' with unsynchronized consumer sources, particularly if you're mixing sample rates. At least based on the documentation which makes no mention of SRC capability.

The Motu 8D seems to be more directly targeted at that use-case, as it explicitly includes SRC on all inputs in order to allow running from a single internal master clock.
 

mansr

Major Contributor
Joined
Oct 5, 2018
Messages
4,685
Likes
10,700
Location
Hampshire
Many S/PDIF receivers can operate from a local master clock and drop/repeat samples as needed.
 
Top Bottom