Why is the nDAC so cheap?
Posted by: Andy S on 04 May 2010
Serious question.
Have Naim scored an own goal? Using a cheap PC and optical to DAC on it's own is such a massive boost over my old CDS1 it just isn't funny and a mate is selling his CDS3 head end as the PC/DAC/XPS is as close as you could get to a CDS3. Not only that, I can connect up a number of sources and get benefit - the TV sounds SO much better through it.
Don't get me wrong, I'm not complaining since I've just bought one (the demo only lasted 5 minutes in all honesty - the distance was that big), just curious...
Have Naim scored an own goal? Using a cheap PC and optical to DAC on it's own is such a massive boost over my old CDS1 it just isn't funny and a mate is selling his CDS3 head end as the PC/DAC/XPS is as close as you could get to a CDS3. Not only that, I can connect up a number of sources and get benefit - the TV sounds SO much better through it.
Don't get me wrong, I'm not complaining since I've just bought one (the demo only lasted 5 minutes in all honesty - the distance was that big), just curious...
Because the SPDIF framing is done in hardware and if it isn't perfect the DAC shouldn't sync.quote:Originally posted by pcstockton:
This is a HUUUUUGE assumption. Why would it always be correct? How could it EVER be correct if ASIO/Foobar isn't bit perfect.
quote:Every time someone brings up a completely valid reason you spin it around and say "well they should be using toslink", or "it must not be bit perfect."
Have you taken one second to think that maybe things arent always bit perfect (especially through CDPs), and people are connecting mostly through BNC and coax?
Yup. If the source isn't bit perfect, all bets are off (I've said it a number of times above!!). It could be that lesser CD transports are less capable of reading the data correctly but given the rips I've been doing I've found my cheapo £20 DVD drive rips accurately on 95% of discs at around 3x speed.
The other 5% are damaged in some way, so you'd expect them to fail.
quote:Have you considered for a moment that the "Chassis/Floating" switch does not cure EVERY single kind of ground issue ESPECIALLY when using coax or BNC?
All the chassis/floating will do is avoid ground loops - which will present itself as a humming on the speakers. It should have nothing to do with data reception or isolating the DAC from the transport. Having said that, the design of the DAC isolates the transport from the DAC as the transfer
of data between theDSP and DAC is done optically.
I don't think I am. The only ones that we can come up with so far are:quote:You are starting to develop a nice long list of exceptions to your dogma. A list that is quickly and easily punching holes in your theory.
- Not all transports can read 100% reliably
- There could be a possible RF issue described above by Gavin (but I must admit, I'm not convinced of that one yet)
Perhaps you could suggest what other exceptions I have come up with
quote:Yes you have clearly shown that in a vacuum all sources will sound the same due to an elimination of timing issues. But microphony, RF, ground loops, lack of bit perfection etc, all can play a part in degrading sound.
Please dont explain this away as avoidable with switches and cables. It is really happening in this real world.
Who told you that? How does it inject noise? Does the sound degradation (which you described previously as a click) happen when you are doing other things - like dragging a window around? My graphics card is doing a sound visualisation. I'll try turning it off and seeing if that improves my sound (in fact, it is the graphics card doing the SPDIF output for me).quote:My Foobar sounds great and I believe it to be bit perfect, not only with ASIO/Transit, but also ASIO or Wasapi on the Juli@. Yet I can hear a difference in sound when doing things like minimizing the Foobar window. I was told the graphics card is injecting a little noise into the mix.
I'm totally confused what you are saying here. Either it does sound different with different sources - in which case, lets find what's causing the differences, or it doesn't sound different in which case it's source independent.quote:While it does not change the overall SQ of the source, it does definitively point to ways in which sources could sound different given a handful of other things than jitter.
The Naim DAC is great, and for me is basically source independent. But there is no way it is a magical piece of kit that CANNOT possibly sound different given other systems and methods.
-patrick
LOL... I'm not the one getting heated under the collar - and if someone could come up with a valid reason, I'm perfectly happy to learnquote:Originally posted by pcstockton:quote:Originally posted by Andy S:
Not saying that it doesn't make a difference in any circumstance, just wondering if it is still valid in todays designs...
Yes, that must be it. The engineers at Naim are hopelessly stuck in the 90's and have yet to get to your level of knowledge on the spdif interface, transports and DACing in general.
How many DACs have you made Andy?
Just for ONCE in your life, maybe it is best that you step back and realize you could possibly be wrong. It is the simplest answer.
Naim didn't put SPDIF on their players or make DACs for a number of years - things seem to have changed, just asking the question - the (excellent) Naim designers aren't Gods you know (they just make exceedingly good hi-fi 
)As to how many DACs, none...
nor am I likely to! But I do build boxes that have HF tuner, HDMI, analogue video, analogue audio, SPDIF etc inputs and outputs with processors running at a couple of hundred MHz and memory being clocked at the near GHz frequency together with several million lines of software all in small boxes that have to work flawlessly - so I think I understand the principles involved here... All I'm doing is, based on my knowledge, trying to understand what could cause differences.ANY question is a valid question, whether it is from you to me, or from me to Naim.
Yup. 100% correct!quote:Originally posted by gav111n:
Hi Andy,
A minor point and it's probably irrelevant to the discussion but you said:
'Having said that, the design of the DAC isolates the transport from the DAC as the transfer of data between the DSP and DAC is done optically.'
From what I looked at, the iCoupler (isolator) is described as a chip-scale micro-transformer.
It shouldn't do as they are transmitting a digital signal - so it should be immune to noise. I quite liked this from the datasheet:quote:I don't suppose this makes any differences in terms of letting noise through to the DAC itself.
quote:Analog Devices:
As seen, the ADuM120x are extremely immune and
can be affected only by extremely large currents operating very
close to the component at a high frequency. For the 1 MHz
example, a 0.5 kA current would have to be placed 5 mm away
from the ADuM120x to affect the operation of the component.
500A 5mm from the chip and it still works to spec. Nice
Ugghh.. You got me... Right through the heart......quote:Originally posted by fatcat:
That was the intention
If someone (of God like status here) makes a claim, then it's either right or wrong. I'm perfectly happy to be told I'm wrong, but I'd like some explanation behind it please - thank you
Seriously, if you read my earlier postsquote:Originally posted by fatcat:quote:White paper
All S/PDIF induced jitter coming in to the Naim DAC is removed.
Notice how it only claims to remove jitter induced by S/PDIF
if it removes the SPDIF jitter, it will remove ALL jitter between A->D conversion and the point of capture at the SPDIF interface on the nDAC. There is NO POSSIBLE WAY for it to determine where the jitter comes from and so eliminate ONLY the SPDIF jitter. If it eliminates SPDIF jitter, it eliminates all jitter produced by the transport by definition....Jitter is difference in TIMING. Nothing more, nothing less. I can delay half the track by 17 days but as long as I buffer things correctly and present the whole thing to the DAC at a monotonic rate, the delay doesn't matter. If the DAC really eliminates SPDIF jitter, it eliminates all jitter period.
Absolutely no change in standpoint at all. Reread what I've said - only 3 posts above yoursquote:Originally posted by fatcat:
That’s a very interesting change in standpoint.
Just coming back on this - does the "avoid jitter completely" refer to the jitter present at source a la Naim white paper (techies tend to like being a...ly retentive on details like this - see the above thread for proofquote:Originally posted by Richard Dane:
Just to throw something into the discussion here. In an email conversation with one Naim's digital engineers, he confirmed that the DAC does remove "all" (his parenthesis) spdif related jitter, and that this can easily be measured. However, he went on to say that nobody has yet figured out a way to avoid jitter completely...
)? I really can't see how you can differentiate between jitter added by the SPDIF vs jitter from the transport reading the data.Again, coming back on this - what you are saying (now I've re-read this) is putting an aerial (which is what a terminated coax cable is) close to the DAC upsets it... Hmm... no, I'm not surprised to be honest. If you're saying RF upsets the sound, then an easy experiment (especially if you have lots of transports) would be to take two transports - one that sounds "bad" the other that sounds "good" and connect them both at the same time, via electrical, both playing music. If the good transport now sounds as bad as the bad transport then you've probably got to a major cause of difference. Having said that, I wouldn't expect cost to be a sole differentiator in RFI performance.quote:Also, what may be of more blame for performance differences is the different levels of RFI that different sources emit. His comments; "You can actually easily hear if you just put a terminated (resistor in the end of the cable) spdif cable in the box, without electrically connecting it at all to the DAC".
So, a bit-perfect streaming player connected via toslink should sound as USB sticks. Ahh.. that's why I can't tell a difference thenquote:Originally posted by pcstockton:quote:Originally posted by Andy S:
Which should mean that connecting it via optical should remove this... (there is no concept of adding a resistor to an optical cable)...
Right, but connecting via toslink might be the least popular. Especially when you consider most people assume that BNC is optimal if you have a way to use it. Hence it is possible that many sources sound different to people due to a handful of reason besides timing errors.
Basically Andy, your argument is reduced to the following.
All sources will sound the same if they present themselves identically (in terms of RFI and other "noise") and bit perfectly, to the DAC and are connected solely through toslink. I would agree with that.
. If we assume that RFI is the only other means of interference (and, by the way, the full metal enclosure makes a good screen against RFI from outside), then all transports (that are bit perfect) will sound the same via toslink. And because spdif is bit perfect, any cheap toslink cable will do. Agree?I'll agree with what you've said, if you agree a possibility is that the differences are so small, it is the placebo effect in action.quote:I would also assume these are VERY subtle differences and maybe only resolvable on an extremely revealing kit with very discriminating ears.
The Naim DAC surely seems to level the playing field futher than any other DAC around right now.
But it is clearly not out of the question for different sources to sound different.
Nice out of context quote Adamquote:Originally posted by Adam Meredith:
How wrong you were.
Did you stand for Parliament last Thursday - you'd make a good politician Yes - the optical gets transformed into electrical but what about the large DSP chugging away next to it at a few tens of MHz. This will totally dominate any RF from the digital side of things.quote:Originally posted by gav111n:
I am not sure that an optical input gets round this does it? Does the optical S/PDIF signal not get transformed into an electrical signal before the DSP? So you still end up with S/PDIF generated RF in the box.
Oh - I do hope soquote:Do you think the naim experts have a chuckle about this thread over their sandwiches?