• 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!

Any Clock Sync Issue for using RME ADI-2/4 Pro SE as crossover ?

MANDEV1L

Member
Joined
Aug 10, 2024
Messages
21
Likes
1
Hi all,
Currently, I am using DDC > I2S DAC > Preamp > Subwoofer (as crossover) > Power Amplifer > Main Speakers

For optimize the signal path (reduce the cables length and minimize analog signal loss), I am planning to implement a digital crossover dsp.

In my seraching of a digital crossover,
I have looked into Minidsp SHD Studio and Motu 8D which both have dual AES outputs.
But their both internal sampling rates are limited to 24/96.
Then I looked into RME ADI-2/4 Pro SE, which seems suitable for my use.

Target:
Digital Preamp > DDC > AES to RME ADI-2/4 Pro SE (for crossover DSP, splitting high frequency signal and low frequency signal)

For High Frequency:
High passed AES digital signal > DAC 1 > power amplifer > main speakers

For Low Frequency:
Option 1: Low passed coaxial/toslink digital signal > DAC 2 > subwoofer, or
Option 2: Low passed XLR analog signal > subwoofer

Is RME ADI-2/4 Pro SE capable for this application ?

if Yes,
for Option 1, is there any clock non-sync issues between the clocks of RME, DAC 1 and DAC 2?
for Option 2, is there any clock non-sync issues between the clocks of RME and DAC1 ?
 
Last edited:
The Minidsp SHD Studio is made for this sort of application and then some. You are missing out on an excellent feature set focusing on the pretty pointless sample rate. 24/96 is better than needed for the task at hand.
 
The Minidsp SHD Studio is made for this sort of application and then some. You are missing out on an excellent feature set focusing on the pretty pointless sample rate. 24/96 is better than needed for the task at hand.
I know the SHD is excellent for my purpose.
But always wondering if there exists a better digtial Xover solution.

And the RME also provide me a choice for using Dirac/Sonarwork for room correction along with some kinds of Xover plugin to do my task.
Also, it provides a industrial proofed standard of SRC for the inputing signals.
 
Last edited:
And the RME also provide me a choice for using Dirac/Sonarwork for room correction along with some kinds of Xover plugin to do my task.
What makes you think so?
 
The RME does not provide any functionality for dirac/sonarworks. None.

Minidsp does dirac out of the box.
 
I would not use the RME ADI-2 Pro as a crossover. It has very limited PEQ functionality for this use case, and no delay.
I am using a Fireface UFX II which offers Room EQ (12 PEQ plus delay). Works great, however, is not the most convenient solution and requires knowledge to operate it correctly.
And - this might be important - the RME UFX II cannot switch sample rates without a very brief drop out upon switching - therefore I am feeding it with a constant 96/24 signal.
 
The Minidsp SHD Studio is made for this sort of application and then some. You are missing out on an excellent feature set focusing on the pretty pointless sample rate. 24/96 is better than needed for the task at hand.
PC runs around it in circles when it comes to DSP and simeles 64 bit FP with long effects chain and pretty much limited by user (and I mean educated well thought and long into it). If you want to use closed purpose box build your own as modular and compatible as you can. There are and efficient and economic enough embedded Intel based N100 (N300 lines) in naked (board only), with passive heat sink or industrial passive enclosure. Now rather popular for NAS builds in 2.5 Gbit X2 with WiFi 6E tho you need one with opposite approach and PCI-E x1 card slot (not as for second M2 SSD slot and or occupied by WiFi SoM extension). 24/96 for output is even overkill and certain enough even for extremes like noise (shaping) filters THX content and so on. As much as I know with Dirac it's limited to 48/24 again enough it's not DSD or BT lossy codec that it needs extensive noise filtering. Regarding DAC/ADC performance 100 and couple dB SINAD is enough tho you probably can get to 110+ regarding DAC without really breaking the bank. In the end it all depends on use case, requirements and money. If latency is not a problem on long distance networking is more than valid option.
Build example with handy old case (hard to find now days especially as mini with full high PCI card support and 5.25 rack on front) to be used with Asrock N100M alike Mini iTX board. For internal build with internal audio card unbalanced.
For multichannel external balanced interfaces build pretty much any N100 will do. And you go either balanced or unbalanced and the way regarding I/O. Up to 10 m in home conditions you shouldn't have problems with unbalanced cables as long as they follow proper AWG, you will get a ground loop (somewhere, sometime) eventually but that's life like. For more distance balanced is recommended from analog. Toslink is 5 to 10 m depending on quality of the cable and as rather old now with some limitations regarding bandwidth.
Sorry no time to elaborate more anyway that's enough to get some picture how things go and decide how much you want to persue it. And yes you will have to learn and work to get there.
 
The RME does not provide any functionality for dirac/sonarworks. None.

Minidsp does dirac out of the box.
Actually not the truth...
I am planning to use M4 Pro Mac Mini with RME ADI-2/4 Pro SE to run DIRAC/Sonarwork VST plugins, and fabfilter pro Q plugin for crossover

I have already tried using fabfilter pro Q for Room EQ correction purpose in a Focusrite Audio Interface, it does a fabulous job...
with 24 customized PEQ bands !
 
Last edited:
The OctoDAC also might be worth a look. Offers XO, PEQ and delay functionality with it's latest firmware, however, requires configuration per config file (no easy to use GUI).
 
Actually not the truth...
I am planning to use M4 Pro Mac Mini with RME ADI-2/4 Pro SE to run DIRAC/Sonarwork VST plugins, and fabfilter pro Q plugin for crossover

I have already tried using fabfilter pro Q for Room EQ correction purpose in a Focusrite Audio Interface, it does a fabulous job...
with 24 customized PEQ bands !
What I actually planning to do is using plugins for room correction and crossover (not using the ADI its own filters), as I have tested in focusrtie audio interface. The result is positive (though I haven't added a delay for the sub phase adjustment yet) :

ee7c28.png


2bnm5t.png


9f72tm.png


Also the latency is low enough for video playback, gaming and recording

3v5k2s.png


I can even replace the Room EQ Correction VST on the above design with Dirac/Sonarworks plugins to see which one can provide the best result.
What I really concern is if there is any technical clocking issue for the connections mentioned in thread #1
 
Last edited:
I am planning to use M4 Pro Mac Mini with RME ADI-2/4 Pro SE to run DIRAC/Sonarwork VST plugins, and fabfilter pro Q plugin for crossover
Your original statement did not mention this. I wish you well.
 
I agree with at least one of the earlier post. Do all processing in the digital domain, then only one final DA conversion step using a multi-channel DAC (like Okto8 or Topping DM7) to feed your mains & sub.

If you use dual stereo DACs that are not clock synced, there will be wierd phasing issues at the crossover region. I encountered this problem, that’s why I moved to DM7 multi-channel DAC to overcome the clocking issue.

No point going through MiniDSP, or multiple DA-AD-DA conversion.
 
I agree with at least one of the earlier post. Do all processing in the digital domain, then only one final DA conversion step using a multi-channel DAC (like Okto8 or Topping DM7) to feed your mains & sub.

If you use dual stereo DACs that are not clock synced, there will be wierd phasing issues at the crossover region. I encountered this problem, that’s why I moved to DM7 multi-channel DAC to overcome the clocking issue.

No point going through MiniDSP, or multiple DA-AD-DA conversion.
Can I add an adjustable delay plugin in the LPF path for my above design to handle the subwoofer's phase and dual dac clocking issues by calibration (by measuring the delay using the dedicated crossover frequency tone signal) ?

In AES connection, which device has the master clock ?
Are high frequecy and low frequecy signals non-sync to a master clock in my above design ?

btw, the minidsp SHD does doing the process in digital domain
https://www.minidsp.com/products/streaming-hd-series/shd-studio
 
Last edited:
Can I add an adjustable delay plugin in the LPF path for my above design to handle the subwoofer's phase and dual dac clocking issues by calibration (by measuring the delay using the dedicated crossover frequency tone signal) ?

I faced the clocking issue when using dual usb DAC. USB is asynchronous therefore it’s almost impossible to clock sync “normal” USB DAC. IIRC there are USB DAC with word clock input but those are expensive, and requires an external clock device.

I do all my digital processing in foobar2000. I used the FIR filter to implement delay by inserting the required number of zero samples to the desired channel. (For example in my case 1323 samples for 7.5ms delay at 176400Hz.)

In AES connection, which device has the master clock ?
Are high frequecy and low frequecy signals non-sync to a master clock in my above design ?

I have not tried AES, therefore unable to advise you on this.

btw, the minidsp SHD does doing the process in digital domain
https://www.minidsp.com/products/streaming-hd-series/shd-studio

Thanks. I learn something new.
 
Hi all,
Currently, I am using DDC > I2S DAC > Preamp > Subwoofer (as crossover) > Power Amplifer > Main Speakers

For optimize the signal path (reduce the cables length and minimize analog signal loss), I am planning to implement a digital crossover dsp.

In my seraching of a digital crossover,
I have looked into Minidsp SHD Studio and Motu 8D which both have dual AES outputs.
But their both internal sampling rates are limited to 24/96.
Then I looked into RME ADI-2/4 Pro SE, which seems suitable for my use.

Target:
Digital Preamp > DDC > AES to RME ADI-2/4 Pro SE (for crossover DSP, splitting high frequency signal and low frequency signal)

For High Frequency:
High passed AES digital signal > DAC 1 > power amplifer > main speakers

For Low Frequency:
Option 1: Low passed coaxial/toslink digital signal > DAC 2 > subwoofer, or
Option 2: Low passed XLR analog signal > subwoofer

Is RME ADI-2/4 Pro SE capable for this application ?

if Yes,
for Option 1, is there any clock non-sync issues between the clocks of RME, DAC 1 and DAC 2?
for Option 2, is there any clock non-sync issues between the clocks of RME and DAC1 ?
Personally, I'd never trust a sub to be a great preamp. i'd much rather have a good preamp provide a sub output (with configurable xover) and of course an output to the poweramp that feeds my main speakers. i don't like the idea of putting my sub between a competent source and the amp powering my speakers. the fewer devices in the signal path, the better. but that's just me.
 
Personally, I'd never trust a sub to be a great preamp. i'd much rather have a good preamp provide a sub output (with configurable xover) and of course an output to the poweramp that feeds my main speakers. i don't like the idea of putting my sub between a competent source and the amp powering my speakers. the fewer devices in the signal path, the better. but that's just me.
Yes. I agree with you.
I have already placed an analog preamp, SMSL SP400 THXAAA 888 balanced headphone amp/preamp between the DAC and subwoofer.
I even suspect if there is AD DA convert in the subwoofer, though the audio quality is quite good currently.
But I still want to optimize the signal path.
 
Last edited:
Yes. I agree with you.
I have already placed an analog preamp, SMSL SP400 THXAAA 888 balanced headphone amp/preamp between the DAC and subwoofer.
I even suspect if there is AD DA convert in the subwoofer, though the audio quality is quite good currently.
That's why I am optimize the signal path right now.
I have been *tempted* to do it because it seems such an easy solution. However I have never seen anyone publish measurements on subwoofer *outputs*. Some even try to power your main speakers directly. I do use a sub because it is a great way to make your system have a much easier time delivering more balanced/controlled sound, but our measurement protocols for now don't fully embrace the potential of subs.
 
I faced the clocking issue when using dual usb DAC. USB is asynchronous therefore it’s almost impossible to clock sync “normal” USB DAC. IIRC there are USB DAC with word clock input but those are expensive, and requires an external clock device.

I do all my digital processing in foobar2000. I used the FIR filter to implement delay by inserting the required number of zero samples to the desired channel. (For example in my case 1323 samples for 7.5ms delay at 176400Hz.)



I have not tried AES, therefore unable to advise you on this.



Thanks. I learn something new.
I think "clocking issue" will occur for using 2 USB DAC, as both DACs are working asynchronously.

But for AES/coaxial/Toslink connection, I'm not sure.
In my design, AES source signal will be input to the RME ADI ADI 24 SE.
After DSP, AES output goes into DAC 1.
Coaxial/Toslink output goes into DAC 2, or XLR to subwoofer.

There should be only 1 clock working as master (I think).
I have asked ChatGPT for more information, wishing anyone here can justify it:

"I connect the aes output of Matrix Audio X-SPDIF3 to the AES input of RME ADI-2/4 Pro SE for digital signal processing.After DSP, the AES output of RME ADI-2/4 Pro SE is connected to DAC 1 and Coaxial/Toslink output to DAC 2 for audio.Is there clock sync issue between them ?Which device act as a mater clock ?"

GPT answers:

" In your setup, the RME ADI-2/4 Pro SE is the central device for signal processing and will likely act as the master clock for both downstream DACs. Here’s how the clocking works in your configuration:

Clock Sync Between Matrix Audio X-SPDIF3 and RME ADI-2/4 Pro SE:

The RME ADI-2/4 Pro SE is likely using the clock embedded in the AES signal from the X-SPDIF3 as its reference. Alternatively, you can configure the RME to use its internal clock and allow it to reclock the incoming AES signal for better jitter performance.
Clock Sync Between RME ADI-2/4 Pro SE and the DACs:

The AES output to DAC 1 and the Coaxial/Toslink output to DAC 2 will carry the audio signal along with the clock derived from the RME ADI-2/4 Pro SE. Both DACs will sync to the clock in the incoming signal.
Potential Issues:
Clock Mismatch: If the DACs are unable to lock to the clock in their respective input signals, you might experience clicks, pops, or distortions. This is rare with high-quality DACs.
Jitter: Each additional step in the signal chain can introduce jitter, which could potentially degrade audio quality. However, modern equipment like the RME ADI-2/4 Pro SE and high-end DACs usually handle this well.
Recommendations:
Master Clock Configuration: Let the RME ADI-2/4 Pro SE serve as the master clock by using its internal clock as the reference. This ensures that all downstream signals are based on a single, stable clock.
Sync Settings: Check the sync settings on all devices to ensure they are configured to accept an external clock (in the case of the DACs) or to use the internal clock (for the RME ADI-2/4 Pro SE, if preferred)."
 
Back
Top Bottom