<html><head></head><body><div style="font-family: Verdana;font-size: 12.0px;"><div>
<div>
<div>Hello,</div>
<div> </div>
<div> thank you,<br/>
your answers made up my mind<br/>
and currently I am in the process of reading in depth about your suggestions.<br/>
</div>
<div>
<p><span style="color:#000080;">>> What is the use case for many sound cards ?</span></p>
<div><strong>There are two use cases: </strong><br/>
<br/>
- The Madi-card should run HQ-mastering grade converters (around 16 i/o).<br/>
<br/>
- The Expert Sleepers converters will interface with an Eurorack modular-synthesizer.<br/>
They are the only ones who have DC-coupled i/o with 20 Vpp,<br/>
for control-voltages and audio. </div>
<div><br/>
( Any DIY solution is out of question for now,<br/>
as they have a decade of experience and their gear is running smoothly )</div>
<div> </div>
<div><br/>
<span style="color:#000080;">>> Use 48000 and be happy, 96000 is<br/>
>> only good for recording bats </span><br/>
<br/>
This could lead towards an endless debate, - which I am not up for in this place,<br/>
but some thoughts on-topic:<br/>
<br/>
I am in a love-hate relation with digital audio processing,<br/>
never experienced a converter in person that is comparable to an all-analog chain.<br/>
Speacially the very highs and 3Dness. <br/>
<br/>
Now I decided for a no-compromise chain (that's why I choose Linux)<br/>
- and I am aware that there are different scientific statements about 48 vs 96k.<br/>
<br/>
( some:<br/>
hearing range, overhead, commercials, other factors for quality as jitter, whole chain, room ect. ...<br/>
vs.<br/>
latency, plugin-qualities and -precision, non-ideal niquist-filters, downsampling to 44.1/16, ... )</div>
<div> </div>
<div>And I guess there is no mastering-studio running 48k in 2020, no offence intended.<br/>
<br/>
I am a PD-programmer and do have a background in electronics and acoustics,<br/>
so I guess that I know what I am talking about (as everyone does ;)</div>
<div>
<div> </div>
<div><br/>
But,<br/>
if I would use 48k only,<br/>
I could run the Expert-Sleepers converters via ADAT<br/>
and yes, ideed - things would be easier.</div>
<div> </div>
<div> </div>
<div><span style="color:#000080;">>> It looks like those ES-8's have ADAT I/O. Could you sync their internal clocks by daisy chaining them via ADAT I/O off of the RME's ADAT output? </span><br/>
</div>
<div>Not at 96k (and yes I know s/mux, but that's not possible here). </div>
<div><br/>
<br/>
<strong>Now I'd like to elaborate the possiblities for running the Madi-chain at 96k, at least? </strong><br/>
<br/>
<span style="color:#000080;">>> you will be using USB hubs which have been known<br/>
>> to cause trouble with audio devices. so be prepared for xruns at any<br/>
>> buffer size less than 1024 (maybe even there). </span></div>
<div> </div>
<div>Yes, I have read about those troubles. But now I found positive experiences, too (including the Jack-website).<br/>
<br/>
It is not resolved why there have been issues with multiple USB-devices<br/>
(might just be clonflicts in shedueling, clocks, bandwith?),<br/>
<br/>
but<strong> if I would use more than one USB 3.2 - PCIe slots in a desktop-computer,<br/>
the bandwith would be more than sufficient!??? </strong></div>
<div> </div>
<div>Of course, faster than 1024 buffer size would be cool.<br/>
<br/>
<span style="color:#000080;">>> Each card have its own hardware clock </span><br/>
<br/>
<span style="color:#000080;">>> You would use an instance of zita_a2j to connect each "secondary" card to the JACK server which is using the "master" card. zita_a2j will resample as needed to keep things in sync. </span></div>
<div><br/>
https://kokkinizita.linuxaudio.org/linuxaudio/<br/>
<br/>
Very cool!<br/>
I am having a close reading on this now.</div>
<div><br/>
It seems like this is taking care of the drifting clocks with a buffer and alignment?<br/>
<strong>The RME as master? Does that mean the hardware-clock of the RME would define the whole DSP-chain? Somewhere I read that RME-cards can only run as slave in Linux, but maybe this is outdated?</strong><br/>
There might be some other down-sides? Phase issues ect? ... I will read more ...</div>
<div><br/>
Zita seems to make up many possibilities ...</div>
<div> </div>
<div> </div>
<div><br/>
<strong>For now I see the following routes:</strong></div>
<div> </div>
<div><strong>- 96k for everything</strong>, RME-HDSPe-Madi and multiple USB-Expert Sleepers<br/>
via zita-ajbridge ?</div>
<div> </div>
<div>https://www.expert-sleepers.co.uk/moduleoverview.html<br/>
https://www.rme-audio.de/de_hdspe-madi.html</div>
<div> </div>
<div> </div>
<div><strong>- 48k for everything</strong>, run the Expert Sleepers i/o with Adat via RME-MadiFx (with Adrian Knoths Alsa-driver) <> RME-Adi648 or multiple RME-RayDat<br/>
<br/>
https://www.expert-sleepers.co.uk/moduleoverview.html<br/>
https://www.rme-audio.de/de_hdspe-madi-fx.html<br/>
https://github.com/adiknoth/madifx<br/>
https://www.rme-audio.de/de_adi-648.html</div>
<div>https://www.rme-audio.de/de_hdspe-raydat.html</div>
<div><br/>
<br/>
<br/>
<br/>
<strong>- 96k for the Madi-Chain and 48k for the ADAT Expert Sleepers chain. </strong></div>
<div><br/>
Maybe this could be achieved<br/>
- with zita-ajbridge by re-sampling?<br/>
- Or with zita-njbridge in a network with one Rasperry-Pi for each USB-connection and re-sampling?<br/>
With this I might get rid of USB-conflicts, too. Running into more possible failures?<br/>
<br/>
Is running different samplerates a good idea?</div>
<div> </div>
<div> </div>
<div>Bests, Manuel</div>
<div> </div>
<div> </div>
<div> </div>
</div>
</div>
</div>
<div> </div>
</div></div></body></html>