Well I don’t think it’s a placebo effect as I was surprised by the result, it went against my expectation. Also specs aren’t the issue here as all media converters must conform to the same spec - it’s an international standard. It’s how it has been implemented that may effect SQ.
If we look at photos of the Ring DAC board and the new APEX version - they do look pretty similar. A cynic might say - it’s just about more profit. But we know that’s not true, there’s an accumulation of many small tweaks and updates that might not be obvious on initial visual inspection.
All I can say is Qobuz streaming SQ is now superb, seriously challenging CD and SACD replay which it didn’t before. It may be that there are other cheap MCs out there that are as good as the ADOT. But you know, life’s too short…
Where do you use the ADOT? I’ve got mine between my home router and Etherregen and then my Melco server and Rossini both connected via standard Ethernet cables to the Etherregen, the idea being that everything connected to the Etherregen should benefit from the ADOT.
I have the ADOT MC in my listening room connected to the dCS via a short CAT cable. The ADOT then connects optically to a Ubiquity Edge router in another room via an S100 (which has SFP connectors). There’s 15m of optical fibre cable. It sounds best that way round, I don’t know why.
The Melco S100 was obtained a while ago in a fit of zeal when ‘audiophile’ switches were all the rage. I now feel the total galvanic isolation provided by an optical link is probably all you need. So today I would not buy a fancy audiophile switch.
FMC is just an off-the-shelf box—with a ~80-cent crystal and nothing at all in the way of high-quality voltage regulation, with an different brand label just.
A while back I took a stab at optical and set up a dual Cisco 2960 system with Cisco cages and an optical cable from Amazon (thanks to some guidance from @Anupc). I felt there was a definite improvement in performance of my system.
When my APEX upgrade was performed this summer my dealer brought my box back to the house however we had difficulty connecting it to the second Cisco 2960 (running off of the first Cisco via optical). Anxious to test out the APEX we bypassed the second router and simply connected directly without the benefit of optical. I did not have the chance to hear what optical did with an APEX’d Rossini.
Fast forward 7 months later and I decided to have a go at optical a second time. I discovered the problem was due to the optical cage not fully sitting inside the optical slot in the router. Once that was fixed I sat down to repeat the same demo from 7 months earlier - ie it would be the first time listening to an APEX’d machine fed by an optically isolated router.
My word - there is definitely a difference. A noisy haze removed resulting in a blacker background and increased dynamics. I am rather taken by it all.
The added advantage is the optical system is relatively cheap as the Cisco components were all sourced for ridiculous prices on Ebay.
The Ediscreation Fibrebox struck me as a better way to do this than using a standard fibre media convertor (with a greater focus on clocking and a better power supply).
Having compared a TP-Link media convertor with the Fibrebox II, there was no contest.
The Fibrebox looks like an elegant one box reasonably priced solution. How does one obtain it (I am Stateside) - direct from the manufacturer? I assume 110v is an option
Yeah, it makes digital more intriguing. But, get one and hear it working.
I am no expert but doesn’t clock crystals generate noise too? So, most probably, they are not reclocking the data for reclocking but are forced to reclock in order to use better clock component.
I am a convinced and happy user of both Innuos PhoenixNet and Ediscreation’s Silent switch.
I don’t think we can ever say that something in the audio chain doesn’t/can’t have an impact on sound quality, only that we don’t think it is likely - a few years ago, I felt the same way about power cables.
I could hypothesise that a PLL tracking a signal with jitter could introduce more noise into the system, but equally I have no way of proving whether that is the case
Well, in this case, since the ethernet transmission is asynchronous and buffered at the dac side, it’s just logic.
Which doesn’t mean that noise can’t enter the scene through a shielded ethernet cable, just to give an example. So the difference some people hear with different switches (I can’t hear that difference tbh) or cable (a cat7 stp cable was more noisy than a cat5 utp one in my experience) may come from that.
But jitter, please, no.
@Zapp My argument was more that one can’t simply rule something out happening because you can’t think of a reason why it might happen.
That said, your answer suggests you misunderstood my side comment. I was not saying it was jitter in the data stream that could be the issue in the example I gave, but that the PLL for the ethernet receiver inside the DAC tracking the difference in clock signals could create noise.
Are you saying that such circuits inside a DAC can’t create noise?
I’m afraid Zapp is right and our friends at dCS have stated same. This is asynchronous transmission. Once the streamer starts doing its magic and converting data frames then clocks kick in, but not before. Ethernet simply doesn’t work like that. I’m not generally a guy who talks in absolutes, trust me!
What I clearly failed to convey very well is that the topic I’m talking about has nothing to do with jitter on the audio data stream, but the fact the demodulation process of the ethernet signal almost certainly creates noise on adjacent circuits (which is one reason why, I suspect, dCS puts this component in a separate box in the Vivaldi series).
My hypothesis was that a Phase Locked Loop, inherent in the design of an ethernet circuit, could incrementally add to this noise when tracking a clock that is less stable.
As it happens, I believe dCS do an extremely good job of isolating ethernet noise compared to other manufacturers.
My main point was to say that ‘I don’t think we can ever say that something in the audio chain doesn’t/can’t have an impact on sound quality, only that we don’t think it is likely’.
Still puzzled. Granted, any circuit activity is capable of generating some noise I suppose. And I’m a guy who uses fiber to isolate my system as much as possible from whatever noise there might be (good hygiene from my POV). But how does the Fibrebox—or any other fiber solution—reduce this “PLL-induced noise”? Whether there is a Fibrebox or not, the Ethernet stream goes into an Ethernet transceiver. How does the Fibrebox reduce noise at the transceiver of the DAC or Upsampler or other “network device”?
