Posts: 36
Threads: 10
Joined: Oct 2025
Reputation:
0
I'm going to replace my Eversolo DMP-A8 with the T8 model.
I currently have the A8 connected to the Expert via RCA using a Wireworld Gold Starlight 8 digital coaxial RCA cable.
Since the Eversolo T8 has an AES/EBU output, I thought it might be an improvement to use an XLR cable to connect it to the Expert.
Before taking the plunge and spending more money (I'm not going to use a lower-quality cable), I'd like to hear the forum's opinion on this.
Thanks.
Carles (Valencia-Spain)
Expert 440 Pro Dual CI, Raidho XT2, Tellurium Q Ultra black; Eversolo DMP A8, Wireworld Gold Starlight 8 digital RCA; Mac Mini & JRiver Mediacenter, JCat isolator, Audioquest Carbon USB; Keces IQPR-3600.
Posts: 31
Threads: 1
Joined: Dec 2016
Reputation:
3
Hi, i also thought of switching from DMP A8 to T8, but did not, yet, as the A8 DAC is very helpful in feeding my headphone amp.
As a D400 user i found A8 -> USB -> Devialet as the best sounding link (there is no AES/EBU out at D8).
Nevertheless, this way Devialet gets the data packets as requested by itself, will store and reclock them.
In addition one can add an external device for galvanically isolating the A8 USB out from Devialet USB IN.
T8 offers ALL outputs galvanically isolated. So your choice is AES/EBU (Eversolo high precision clocks will be the timing master), USB (Devialet times data transmission) or ...
You should try any combination ... and trust your ears ...
Posts: 582
Threads: 27
Joined: Jul 2015
Reputation:
24
Out of all the digital inputs the Devialet offers, I've found that their AES/EBU to be the "most musical" one hence I use a Wattson Emerson Streamer powered by Uptone Audio's JS-4 Linear power supply and use the AES/EBU OUT of the Wattson to my Devialet's 440-Pro AES/EBU IN with an Audioquest WILD digital cable. I am extremely pleased with this configuration.
Posts: 171
Threads: 11
Joined: Dec 2014
Reputation:
3
According to my own experience the best digital input/signal path is through Ethernet/network using a high quality switch between your Devialet and your source. That way you end up with hands down the lowest distorsion possible.
TORUS POWER RM16 CE + NAGRA Streamer w FARAD Super 3 + D 250 Pro CI +
Magico S3 mkII + Wireworld cables = MUSIC 🔊🎶🤩
- SWEDEN -
Posts: 582
Threads: 27
Joined: Jul 2015
Reputation:
24
(03-Mar-2026, 21:54)MightyMart Wrote: According to my own experience the best digital input/signal path is through Ethernet/network using a high quality switch between your Devialet and your source. That way you end up with hands down the lowest distorsion possible.
Theoretically, I might agree BUT in practice, you use an External "Nagra Streamer" with sPDIF as input to your Devialet  and so do I to my AES/EBU of my Devialet.
Posts: 101
Threads: 2
Joined: Feb 2021
Reputation:
5
I connect my PC to a DDC and then connect DDC to Devialet via AES. My DDC is also externally clocked and the clock signal is passed to Devialet via AES. Devialet is a relatively old product, it may have a clock that is considered good at its time, but now almost all streamer products has better clock, especially cannot compare with external clocks. The Devialet network connection cannot compare with a clocked AES signal.
Posts: 2,493
Threads: 46
Joined: May 2014
Reputation:
100
04-Mar-2026, 15:40
(This post was last modified: 04-Mar-2026, 15:40 by thumb5.)
(04-Mar-2026, 10:24)ray1234 Wrote: I connect my PC to a DDC and then connect DDC to Devialet via AES. My DDC is also externally clocked and the clock signal is passed to Devialet via AES. Devialet is a relatively old product, it may have a clock that is considered good at its time, but now almost all streamer products has better clock, especially cannot compare with external clocks. The Devialet network connection cannot compare with a clocked AES signal.
From a technical point of view, most of what you said there makes no real sense to me, though I am not arguing with your preferences and what you hear. There is no separate clock signal in an AES3 connection, which means that the Devialet has to recover the clock from the data (audio samples). Typically this is done using a phase-locked loop which is subject to timing variations, so the accuracy of the recovered clock still depends on how well the clock recovery mechanism in the Devialet works. By contrast, if you deliver audio samples via the network, the Devialet is in complete control of the clocking and this generally provides higher timing accuracy (the same is true when using the asynchronous audio interface).
Again, I'm not arguing with what you hear, but it's unlikely that when you use AES you have a higher-accuracy clocking system end-to-end.
Roon (Mac Mini), Wilson Benesch Full Circle, Expert 1000 Pro CI, Kaiser Chiara
Warwickshire, UK
Posts: 171
Threads: 11
Joined: Dec 2014
Reputation:
3
06-Mar-2026, 18:22
(This post was last modified: 08-Mar-2026, 00:04 by MightyMart.)
(04-Mar-2026, 00:29)audio_engr Wrote: (03-Mar-2026, 21:54)MightyMart Wrote: According to my own experience the best digital input/signal path is through Ethernet/network using a high quality switch between your Devialet and your source. That way you end up with hands down the lowest distorsion possible.
Theoretically, I might agree BUT in practice, you use an External "Nagra Streamer" with sPDIF as input to your Devialet and so do I to my AES/EBU of my Devialet.
What matters even more than the type of connection used is the quality of the source used to stream. The NAGRA Streamer is a superior streamer compared to many other streamers both with and without a feature that allows for streaming via the network.
And theoretically the NAGRA Streamer would probably be even more superior if that feature had been available. 🙃
TORUS POWER RM16 CE + NAGRA Streamer w FARAD Super 3 + D 250 Pro CI +
Magico S3 mkII + Wireworld cables = MUSIC 🔊🎶🤩
- SWEDEN -
Posts: 101
Threads: 2
Joined: Feb 2021
Reputation:
5
(04-Mar-2026, 15:40)thumb5 Wrote: From a technical point of view, most of what you said there makes no real sense to me, though I am not arguing with your preferences and what you hear. There is no separate clock signal in an AES3 connection, which means that the Devialet has to recover the clock from the data (audio samples). Typically this is done using a phase-locked loop which is subject to timing variations, so the accuracy of the recovered clock still depends on how well the clock recovery mechanism in the Devialet works. By contrast, if you deliver audio samples via the network, the Devialet is in complete control of the clocking and this generally provides higher timing accuracy (the same is true when using the asynchronous audio interface).
Again, I'm not arguing with what you hear, but it's unlikely that when you use AES you have a higher-accuracy clocking system end-to-end.
You are correct, Devialet has to recover the clock signal from AES. From my experience, a good clock signal fed via AES is still better than using the internal clock of Devialet. Anyway, that could be particular just to my system setup, it is just a reference for anyone else.
Posts: 255
Threads: 10
Joined: Oct 2017
Reputation:
30
If I remember correctly, after Devialet brought the CI, the ports on the CI (USB, Ethernet etc.) were handled by the clock on the CI, while for ports like COAX and AES/EBU that remained on the old platform, Devialet used a different clock. If I remember this information correctly, this could be the reason for the sound difference that occurred. I don't know if anyone else on the forum remembers anything else about this topic.
Audioquest Thunder | High (Variable) Current Power Cable - Audioquest Niagara 3000 - Apple MacMini (2018) - Cardas Clear USB/ Blue Jeans Ethernet Cable - Devialet Expert 440 Pro on IsoAcoustics Orea Bronze - Van Den Hul The Breeze Hybrid Speaker Cables - Focal Scala Utopia Evo on IsoAcoustics Gaia-TITAN Theis
|