I can’t take credit, that’s @GoldenSound’s ™️
2 Likes
You have to be careful with such statements - that those instances of galvanic isolation are not “perfect” does not mean that all aren’t (or at least, aren’t perfect for what they’re trying to address).
From the discussions I’ve seen/had, my understanding is that for the actual transmission frequencies the galvanic isolation there is perfect. Unfortunately, it was later discovered that RF from the FPGA could get past that isolation and modulate the ground plane in the Hugo TT2 or DAVE.
Had that been apparent during initial design/implementation, it could have been addressed and resulted in perfect-for-the-task isolation as opposed to isolated-for-the-primary-offender. Which is one reason why Hugo TT2 is a bit less prone to the issue than DAVE.
…
It’s curious to me as to why the M-Scaler uses BNC connections as the primary, Chord-to-Chord connection. Sure, you couldn’t use TOSLINK without having four such connections, but optical isn’t limited to TOSLINK and it would not have been a big deal to stick an ST-class (or other) optical link there to a) give MUCH higher bandwidth and b) avoid all these RF and coupling concerns entirely.
Of course, then you run into the people fussing over where to put their M-Scaler in relation to their DAC as they’re claiming that, even with all the isolation in the world on the connection (e.g. something like OptoDX, using optical networks, air-gaps, etc.) … now that naughty GHz RF is getting in to the DAC … through the air!
…
And then just for a bit of fun … I do a fair bit of extreme radio engineering (so all RF by definition) …
So, it’s constant source of amusement to me that we can reliably send very low-power signals, literally across our solar system, through an RF and EM environment that is so massively hostile it is hard to fathom, and get perfect data retrieval.
But … we can’t send very robust audio signals about a foot and a half without questioning everything about that connection (even when it is working perfectly) and, even, calling into question all of physics!
(Yeah, I know, it’s not quite the same thing … but it is no less amusing because of it - at least to those who work in RF, EM and digital transmission fields).
4 Likes
I’m also going to add this …
I am not saying that the RF/ground plane modulation with an M-Scaler and a Chord DAC isn’t real. Nor that there can’t be audible changes as a result of mitigating or addressing it.
I did some testing with an OptoDX, and also with the Wave cables.
OptoDX does indeed, at its output, deliver a cleaner signal than it receives from the output of the M-Scalers. What it did not do, in my testing, was make any measurable difference at the analog output of the DAC.
Indeed, the manufacturer acknowledges this (quoting from their white paper):
Can the benefit of an optical bridge be directly measured at the analog outputs of a DAC? In other words, is there a measurement such as Total Harmonic Distortion(THD) or Frequency Response that demonstrably shows the audible benefits obtained? In short, the answer is no.
They then go on to qualify that as “or maybe not yet” …
And I heard no difference with, or without the OptoDX in my system, blind or sighted. Which, again, doesn’t mean there isn’t one, just that I don’t hear it (plenty of people swear this thing gives night/day improvements).
…
It was the same case with the Wave cables … cleaner measured output, than input (that’s good) - but I found no measured nor audible change from the DAC (so no need for me to bother - YMMV).
The Wave cables are interesting. 20 tuned ferrites on a cable. That’ll result in 20x more RF attenuation than just using one such ferrite. Good so far. Nicely made. It’s a pretty design.
However, the actual RF attenuation performance is not as good as simply taking an equivalent ferrite ring (about $10), and then looping the BNC cable through it six times (which would give 32x the attenuation of a single loop). Each additional pass through the same ferrite doubles the RF attenuation.
So if you want to try this, and not drop between $800 and $1,500 or so, it’s easy and cheap to do so.
…
My favorite RF demo involves one of my little (10W) desktop/portable transceivers. If I key that up in the right frequency band, while just running a simple coax cable to an antenna (antenna in another room or outside), it’ll instantly cause every USB device in any room that cable passes through to disconnect.
Loop the source end of that same coax cable through a 2" ferrite ring 4x and the disconnects stop.
4 Likes
Just when I was about to consider a UnisonQuantum entanglement connection for my BiFrost 2/64 and whatever tube amp I may buy in the future. I figure if the Quantum connection can exceed the speed of light, then it can certainly connect to what I am considering buying now. All I’ll need to do is to follow the connection and see what my perfect choice was.
7 Likes
That’s funny!
[Now … putting my serious hat on …]
Depending on which interpretation of quantum mechanics you’re subscribing to, either the measurement effect occurs instantly or, in models that don’t recognize the wave-function collapse, there is no effect.
Either way, while all of the models do, I think, agree that you can exploit the mutual information between entangled particles, this still doesn’t allow you to transmit information at speeds > C.
[Serious hat still on, but perhaps cocked at a bit of an angle …]
Besides, the minute you start measuring your bits, you won’t know where they are anymore.
And in a quantum system, effects can precipitate their causes. Just to be extra weird, effects preceding causes and causes precipitating effects can both be true at the same time. Though this might well simply be due to an incomplete understanding of the concepts of “cause” and “effect” in a quantum system.
We don’t know.
And if we start trying to measure it again … all hell’s gonna break loose and there’ll be Schiit all over the place. I’m sure many of us remember what went down shortly after a one Gordon Freeman was seen doing this …
7 Likes
Don’t give Schiit any ideas, or they might go and develop a quantum DAC as a successor to Multibit… I guess as long as they can fit it on a set of Yggy upgrade cards then I won’t be too upset.
8 Likes
Sure, but can it play Crysis? 
4 Likes