Master/slave link

[Mikeeo]

> There has been no 'voicing' to the supplied interconnect. As a matter of fact, Devialet proprietary technologies make it rather insensitive which cables are used (be it power supply, speaker, or digital interconnects). We do not believe there will be 'flavour' differences between different digital interconnects between both units, and have not noticed any.

I must say I am rather intrigued as to why Devialet have chosen to include a £500+ digital cable with their 400 package. Also with my 500 I received what looked like a (white labeled) Transparent Premium digital cable which retails at £755.

If what they are saying is correct then what is the purpose of including expensive digital cables in their packages? Of course one can spend a lot more money on a digital cable but that's not the point.

Guillaume

500+ GBP is for the consumer I guess so I am pretty sure it is much less for the manufacturer and not to forget multiorder discounts. But as you say Dev seem not to believe in 'exclusive' cables which is contradictive and why didn't they supply a Mogami instead at a better value choice!?
I'll soon test another dig cable, just ordered the DIY parts.

/Mike

[GuillaumeB]

Without wishing to incur the scorn and obloquy of the cable sceptics among us, have any more of you carried out experiments with the digital link between master and companion?

Guillaume

[thumb5]

Without wishing to incur the scorn and obloquy of the cable sceptics among us...

Quite beautifully put - you're a master of words, Guillaume!

[Mikeeo]

Without wishing to incur the scorn and obloquy of the cable sceptics among us, have any more of you carried out experiments with the digital link between master and companion?

Guillaume

G, still waiting for my diy stuff that I need to assemble my fav dig cable.

In the middle of moving houses and today I disconnected the system and while doing this I noticed that one of the Devs rubber feet came lose. I watched it carefully and read "HQ ISOLATOR", hmm funny as it is just a thin rubber cap.

/Mike

[aerki]

I fully agree with you guillaume

This sync link is a serious issue for those having been investing a bunch of bucks, kiwis or whatever in the 2 amps.

I am currently busy evaluating vqrious cables.

Moreover i don't know so far what's the best option to go for: rca/rca or rca/aesebu.

The rca/aesebu is a tricky one, since on one side you have to comply with the 75 ohm impedance, versus 110ohm on the opposite side.
And when you know a bit of the basic physics for coax cable (impedance depending on diameter, thickness and type of the insulation material), this is simply not possible to obtain this just by soldering the wires on the right pins of a rca or xlr connector.
(As adviced on the devialet website )
To do it properly, it requires an impedance adaptation, with a large bandwidth

I will be testing this option next week. I ve ordered such a special cable from a french cable manufacturer, fyi see audioprana website. They call it symetriseur.

Canare is providing an impedance adapter for those who like diy...

[IanG-UK]

One of the reasons why Quad's business failed, when in UK hands, was that they never made much money in the good days, in part because Peter Walker was of the opinion that you should sell customers what they need rather than what they want. Hence a 24 year gap between loudspeaker releases and typically a 10 year gap between amplifier releases.

In the market of "big boys toys", that is generally a flawed approach - so we are inundated by multiple releases conferring allegedly significant improvements in cars, sporting equipment and audio/video equipment; and by questionable ancillaries. And by manufacturers aligning themselves with consumer flavours even if, privately, they think those alignments make no measurable or practical difference.

I'm sure I have spent silly money on cables and connectors and tweaks which I would claim are improvements (or did at the time of purchase) but which I could not prove through rigorous measurement and verifiable listening tests. I am quite happy that I have been "placeboed".

And this is the way it will remain, specifically on cables, in my opinion. I cannot remember any magazine testing of this general nature since the Quad test in, I guess, the early 70s when the listening panel (itself a flawed approach) could not distinguish between a Quad ll, a Quad 303 and a Quad 405. And reviewers best equipped to do such tests now, people like Colloms, Miller and Sircom, will never expose themselves to the risk of failure.

It would be far better if we adopted the expression "I think I can hear a difference" to "there is a difference" and then we would all be happier. Possibly.

[thumb5]

It would be far better if we adopted the expression "I think I can hear a difference" to "there is a difference" and then we would all be happier. Possibly.

Ian, for what it's worth, I agree with your sentiments on this general topic. But it seems to me that there are some unknowns with respect to the Devialet master/slave link that complicate that pragmatic view.

If the cable is carrying high-speed digital data with fast rise/fall times, and there's no error detection/correction - not that I know either way - I think I'd also be quite worried that there is no obvious "standard" to which the cable should conform. Potential impedance mismatches could be a significant worry.

One would hope that Devialet have designed the link protocol so that it either works perfectly or doesn't work at all. If that's the case then the choice of cable should be ... (choosing my words carefully!) ... less significant.

Ian

[GuillaumeB]

It would be far better if we adopted the expression "I think I can hear a difference" to "there is a difference" and then we would all be happier. Possibly.

Ian, for what it's worth, I agree with your sentiments on this general topic. But it seems to me that there are some unknowns with respect to the Devialet master/slave link that complicate that pragmatic view.

If the cable is carrying high-speed digital data with fast rise/fall times, and there's no error detection/correction - not that I know either way - I think I'd also be quite worried that there is no obvious "standard" to which the cable should conform. Potential impedance mismatches could be a significant worry.

One would hope that Devialet have designed the link protocol so that it either works perfectly or doesn't work at all. If that's the case then the choice of cable should be ... (choosing my words carefully!) ... less significant.

Ian

Great post Ian. This is exactly why I'm interested in hearing from other dual mono owners. My own experience consisted of swapping the stock cable (a rebadged TA premium S/PDIF) with the TA XL.

I also tried a Chord Sarum TA and a Chord Signature Digital, but the TA XL was much better - to my ears!

In fact I would say that the upgrade from stock to TA XL was far greater than upgrading the link from streamer to master. Considerably so I might add. It was such a big upgrade that it made me really wonder what on earth was going on here.

If others find no difference that's fine by me too! I'm not trying to convince others of anything, I'm just interested in hearing from others.

I was also intrigued to hear that in the latest firmware Devialet had supposedly fixed a jitter issue between Master and Slave/Companion... and then there have been these reports from others on here about a lack of homogeneous sound...

Guillaume

[IanG-UK]

Thanks for your replies Guillaume and Ian. I know this is straying from topic but ...

I'm a bit surprised that Devialet are not more forthcoming on issues like these. Their kit is pretty innovative (and patent protected, for what that might be worth) so they must have tons of data on lots of things. One particular example would be them saying exactly what new releases are fixing. There must have been 20 or so releases by now so it would make interesting reading.

I know it will be sensitive if those things reveal shortcomings which they have yet to optimise (for example I guess there have been significant modifications to AIR since its first release) but there are presumably some things that they could(and should) comment on, most pertinently the master/slave link (which is obligatory) and the mains supply or cable (which is obligatory too). In a sense they have already done this with loudspeakers by creating SAM, presumably using some undisclosed speaker cable of some undisclosed length. So I think they should say on all these things whether it matters or not, purely from a measurement perspective - and then whether there is any scientific evidence that this matters in terms of listening.

In fact, if Devialet have a weakness it is in their general communication and the change in their website from a technical bias to a marketing one. Probably a good move commercially but there are enough eggheads out there to warrant lots of technical data with the proviso that we cannot expect Devialet to respond to challenges about it.

I know doing all this might oblige them to then provide appropriate mains cable and digital cable (which might cost £1 or could cost £1,000) - if they do not do so already - but at least it would then satisfy those who believe everything can be measured and leave the others to choose or try other things.

[Confused]

Some very good points above. It would appear that no one (out side of the Devialet inner circle) actually knows how the slave master link actually functions. Plus the cable will only influence sound quality in one speaker, although I guess this is fairly significant if you consider stereo imagery etc.

Does anyone know how exactly the slave master link works?