PC and MAC Rips: PC Superior?

Hi Simon -

I am not so sure of that!   AndyS was adamant in his belief that there was no way, given the architecture described in the DAC white paper, for sources for the DAC to sound different in and of themselves.

After months of back and forth, a consensus was reached:  assuming that two sources were delivering the same, bit-perfect stream, if one sounded different, then it had to be somewhat "broken".   It had to be emitting enough EMI, RFI or some other noise or vibration to create an "environmental effect" (my words, not his) that had an impact beyond the buffer, on the DAC's output stage or something else beyond it in the playback chain.  Otherwise, AndyS took on and defeated a number of good debaters.  He convinced me and many others that the DAC was able to remove all S/PDIF-related jitter, just like it says in the white paper.   Of course, there is the possibility, as absurd as it may be, that Naim has not correctly described the DAC's capabilities in the white paper.  In any event, to the best of my knowledge, Naim to this day has remained silent on this topic, preferring us to use our ears and decide for ourselves.

If you go back and read:

this thread

and its successors, you will see that the discussion went on for quite a long time.   Many of us, including myself, have never been able to hear differences among sources for the DAC.  Perhaps we were influenced by this thread, or perhaps the differences are so small that they can only be heard in a top-drawer, reference-level setup.  Don't know.  

To add to the confusion, AllenB, whose opinion I respect, has changed his view on sources for the DAC, and now reports that can hear differences where before he could.  (It is a bit more complicated than that, but you can read for yourself Allen's thoughts in his new thread about his loaner NDX).

Anyway, I believe, as Likesmusic says above, that AndyS would have argued that the artifacts you described are not significant enough to prevent the DAC from eliminating S/PDIF-related jitter.  Unfortunately, I am not an engineer, and cannot make those arguments on his behalf.   If you are not up for reading those old threads and commenting (they do make for excellent technical reading IMO), then the best I could offer would be to cut/paste some of AndyS's more cogent comments from those old threads for you to respond to.  If you are up for it, I would be glad to do so.   I would probably start a new thread, and begin it with a big warning for folks not interested in technical minutia to simply ignore it.   I would also probably receive a lot of criticism and complaining for opening up this old can of worms, but, to be honest, the issue was never resolved in my opinion.   It remains, in my view...debatable.

Also, for reasons I am still not sure of, the topic produced some of the most heated discussions I've ever seen on this forum.   I hesitate to rehash the subject for concerns over stirring up these old feelings, but I am also fascinated by the idea of you having an opportunity to refute his arguments, even if it is done "in absentia".  Anyway, am equally glad to drop the topic, or explore it further.   Totally your call Simon, so let me know what you think.  

Would also welcome other people's thoughts on whether this is worth pursuing in a new (technical!) thread or not.

Thanks.

Hook

Hook

Thanks, from an engineering perspective everything is ultimately a compromise (one thing we are always taught). So algorithmically or functionally you can design a robust approach that is 'perfect', however in implementation  everything has a side affect or unintended affect. The challange of engineering is to try and make these side effects and compromises fall within design tolernces and ultimately irrelevant, in this case to the listener - but as you can see a judgement has been made and is non absolute and is ultimately a question of balacing the law of dimishing returns - another engineering design pricniple. 

A non truly perioidc load (switching a square wave with jitter) is going to produce side effects that correllate to the load - for this example lets chose the device powerlines/supplies and logic circuites do produce switching noise hence why designs pay close attention to decoupling to minimise this. The powerlines ultimately are shared between the reception clocking circuitry and the clock circuity of the DAC. Therefore the aim is to make this affect fall within the design tolerences of the device. As there is no such thing as 100% reliability/perfection - inevitably there will be side effects. A top class design from a manufacturer such as Naim is going to / should make this insiginficant - but ultimately can't remove it together. Whether an individual can hear these compromises is where all fun begins :-)  but those compromises are definitely there.

To be honest as this is not an engineering forum or technical forum - there probably is not much mileage in discussing further here  and for it to be meaningful we would need to get into the design of Naim components, layouts and design tolerences at the low level and that isn't going to happen as clearly that is commercially sensitive to Naim.

So I am quite happy to go along with  people to report on thier subjective assessments, as ultimately I guess as a consumer device that is what counts, and let each person make thier own mind up rather than rely on others to tell them possible falshoods or pseudo science as justification for what they are supposed to hear.

Simon

Well said Simon, and probably a good place to end this discussion.  

Funny, I keep hearing the Buffalo Springfield-era Stephen Stills in the back of my brain...

"There's something happening here
What it is ain't exactly clear
...
I think it's time we stop, children, what's that sound
Everybody look what's going down"

And I wonder how he will sound streamed via an NDX/DAC/555PS.

Hook

"perfection is a purely human aspiration and laughably implausible to the rest of the cosmos"

Tog

Hook,

I thought that Hjalmar Nilsson, the code designer for the nDAC, sort of revealed naim's position during the chat event.

asenna04 asked:

OK, I thought you would say that. But the nDAC utilizes beffuring to eliminate jitter. Why then do different transports that feed the same file into the nDAC would sound different?

HN answered:

There are a multitude of reasons for that, and it's a bit much to take here since this is mainly a discussion regarding the FLAC release. However, RF interferance is a popular explanation. Having a poorly designed switch mode power supply, commonly found in cheaper sources, will also affect the sound.

Gav

Hi Gav -

Thanks for posting that.   Would have been nice to hear what some of those other reasons are.

I have no problem believing that a cheap source with a horrible SMPS spraying the entire setup with RFI could sound different, since it would negatively effect the playback logic beyond the DAC's buffer. 

The mystery to me has always been how sources of "acceptable quality" could sound different, since at least in theory, both should be able to deliver a bit perfect data stream "well enough" for the DAC to be able to buffer and re-clock.

For now, I think I will accept, but continue to ponder, Simon's explanation.   It has almost a Zen-like feel to it.   Maybe the DAC has to work harder for some sources than others to achieve its goal of zero S/PDIF-related jitter.  Perhaps by working harder, there is a negative effect on its output stage....or something like that.  

Hook