Why is the nDAC so cheap?

quote:

Originally posted by pcstockton:
Really? OK.

13 pages of getting nowhere? Yeah thats a blast.

Enjoy.

Well.. that was 2 hours and 20 minutes you were out of the thread

quote:

Originally posted by JYOW:
To me, this means that there could still be differences between transports in how much RF noise they introduce to the DAC. But that, according to Naim, is also "significantly reduced".

But then this should be eliminated by optical and by placing the transport a distance away from the DAC.

quote:

Being a tin ear does have its advantage.

Or a sceptic with some understanding of what's going on

quote:

Originally posted by Andy S:
But then this should be eliminated by optical and by placing the transport a distance away from the DAC.

A pedantic correction. The split PCB/power supply will significantly reduce the RF between the DSP and the DAC compared with a shared PCB/power supply. This has nothing to do with the RF that might be induced by the transport.

Also note that the RF interaction will be further reduced by the use of an external power supply which powers the DAC sections independently of the DSP section.

quote:

Originally posted by pcstockton:
Really? OK.

13 pages of getting nowhere? Yeah thats a blast.

The point is that the thread pertains to a topic which is not in contridiction to the rules and regulations for using this forum.

Whether it is interesting or not [for others] is not for you to judge.

And it is interesting to some people.

Most of us follow most of Naim's "upgrade path" to a T.

We can choose to follow the gospel according to Naim blindly. Or we can choose to understand it more, after all we are paying good money for it.

As the digital world get more "modernized", a lot of the mystery shrouded behind the curtain is gone.

As the point of diminishing return is decreasing, it is increasingly difficult to justify the high price=high performance formula.

Which is exciting time for us natives - instead of blindly following that formula, we should try to equip ourselves to judge the price/performance gain.

quote:

Quote: Andy S

As to isolation, a single bitstream - especially if connected via optical - has minimal impact to the digital noise that could be present where a DSP exists.

quote:

Quote: Andy S

OK.. so how would it be correlated? I have an input buffer that is clocked by one clock, and the output buffer clocked by a different uncorrelated (if you are to believe Naims white paper) clock. Other than the clock selector switching around (which you'd expect to cause a hiccup in the output) how does input jitter produce output jitter - which Naim claim to remove....

OK, so the DSP produces significant noise. I don’t see why this ‘noise’ will be randomly distributed in time. What is the DSP doing when we are listening to music? Isn’t it mainly writing the incoming (S/PDIF) bits to the buffer memory and also extracting stored bits from the buffer memory? So the DSP is working to the ‘beat’ of the nDAC master clock to extract data from memory but is also working to the ‘beat’ of the S/PDIF clock to write data into memory. It is when the DSP performs an operation that (most?) noise is produced. So it will be producing significant noise to the 'beat' of both the nDAC master clock and the S/PDIF clock.

If the incoming signal has no jitter, then the (important) noise produced by the DSP will be synchronised with the outgoing signal. If any of this noise is present on the outgoing signal, it will be constant from bit to bit. If the incoming signal has jitter then some noise on the outgoing signal (from the DSP) will be correlated to the incoming jittered timing. I think you said that noise in a bitsream is simply jitter, so a ghost of the original jitter may remain.

An optical connection isn’t going to help here.

In this scenario it’s probably fair to say that all jitter present in the incoming S/PDIF signal has been removed (when it is in the buffer), however a smaller amount of jitter is added back by the nDAC processor noise and this is happens to be a function of the incoming signal.

Gavin.

quote:

if it really is a mass placebo effect...

You are under a massive placebo-effect.

quote:

Yes, I know that isn't a popular position around here, but I say what I find...

And I say what I find.

quote:

Originally posted by gav111n:

quote:

Quote: Andy S

As to isolation, a single bitstream - especially if connected via optical - has minimal impact to the digital noise that could be present where a DSP exists.

quote:

Quote: Andy S

OK.. so how would it be correlated? I have an input buffer that is clocked by one clock, and the output buffer clocked by a different uncorrelated (if you are to believe Naims white paper) clock. Other than the clock selector switching around (which you'd expect to cause a hiccup in the output) how does input jitter produce output jitter - which Naim claim to remove....

OK, so the DSP produces significant noise. I don’t see why this ‘noise’ will be randomly distributed in time. What is the DSP doing when we are listening to music? Isn’t it mainly writing the incoming (S/PDIF) bits to the buffer memory and also extracting stored bits from the buffer memory?

No, it's doing the upsampling at least plus it will be monitoring what is going on in the system (buffer levels etc) i.e. controlling the way the DAC works. If it were as simple as you suggest, the DAC wouldn't need to have a DSP inside it

quote:

So the DSP is working to the ‘beat’ of the nDAC master clock to extract data from memory but is also working to the ‘beat’ of the S/PDIF clock to write data into memory. It is when the DSP performs an operation that (most?) noise is produced. So it will be producing significant noise to the 'beat' of both the nDAC master clock and the S/PDIF clock.

If the incoming signal has no jitter, then the (important) noise produced by the DSP will be synchronised with the outgoing signal. If any of this noise is present on the outgoing signal, it will be constant from bit to bit. If the incoming signal has jitter then some noise on the outgoing signal (from the DSP) will be correlated to the incoming jittered timing. I think you said that noise in a bitsream is simply jitter, so a ghost of the original jitter may remain.

If this were the case, then cables would sound different (some say they do, others - see other thread in hi-fi forum - say they don't) as they would alter the timing effects slightly. Having said that, unless the clocks were perfectly synchronised at either end (and this is highly unlikely due to miniscule variations in production), you will still get beating even if the reference clock is of very high quality as the frequencies will never be 100.000% the same and they will not have the same phase with relation to each other.

In addition to having difficulties explaining anything which could make different transports sound different, I'm also having difficulty understanding why more expensive transports sound better, and why people consistently score them better. If more expensive players/cables do sound different, it should be fairly straightforward to explain what is going on around here since the designers of the more expensive players will have to have made conscious decisions to minimise the effect of something...

Wouldn't it be nice to get two persons (one who he does hear a difference and one who says he doesn't hear a difference) in the same room and listen to the same music at the same time.

I wonder if either one would still hear differently from the other.

Just wondering how that would turn out

-
aleg

IMHO there are more important things in life and some people should take themselves less seriously.

quote:

Originally posted by Thorsten_Lux:
IMHO there are more important things in life and some people should take themselves less seriously.

Hi Thorsten,

Obviously, you hear differences in different transports used with the DAC, and you are not interested in why. That is fine, and no-one has criticized you for it. But why do you persist in making the same point over and over again? Don't you believe that the rest of us have a right to discuss this? Why does the discussion annoy you so much?

quote:

Originally posted by Andy S:

quote:

Originally posted by AMA:
Transports induce much more jitter into the bitstream than cables.
If nDAC can reject all jitter-related contamination from cables this does not mean it can do the same clean job on any transport.

Hi AMA,


Sorry, but yes it does (within the definition of a "reasonable" transport). Jitter is about presenting the same information at slightly different times. We are talking about the same clock (on average) just presented either slightly early or slightly late. Higher jitter transports (lets move them from 200ps -> 1000ps) just have a little more early and a little more late... The nDAC rejects this level of jitter - period (as you've found out with your cable trials).

I think people are missing the fact that jitter is both quicker and slower... If a clock is at 44.1kHz minus xHz, and it is ALWAYS at that frequency, it means it has zero jitter. A jittered clock is 44.1kHz+/-xHz. The average clock will be at 44.1kHz but can be up to 44.1+xHz and as slow as 44.1-xHz... It can change on a cycle by cycle basis..

Andy

This is a very simplistic and incorrect view.
Jitter is created in the transport by various means, power supply noise, motor noise, whatever, lets even say stray laser light. A closed eye pattern indicates the presence of jitter, however THE JITTER IS MERELY A VISIBLE AND MEASURABLE INDICATIOn OF THE PRESENCE OF NOISE. The more noise the more jitter. The jitter itself is not the problem, the problem is the NOISE.

The noise is transferred to the DAC, the fact that the master clock in the Niam DAC eliminates clock jitter will NOT remove this noise. The fact that it eliminates clock jitter caused by the transport clock and S/PDIF interface is very impressive, however THE DESIGN OF THE NAIM DAC ONLY RELIEVES SYMPTOMS, BUT DOESN’T EFFECT A CURE.

This is why transports do matter.

quote:

Originally posted by fatcat:

This is a very simplistic and incorrect view.
Jitter is created in the transport by various means, power supply noise, motor noise, whatever, lets even say stray laser light. A closed eye pattern indicates the presence of jitter, however THE JITTER IS MERELY A VISIBLE AND MEASURABLE INDICATIOn OF THE PRESENCE OF NOISE. The more noise the more jitter. The jitter itself is not the problem, the problem is the NOISE.

The noise is transferred to the DAC, the fact that the master clock in the Niam DAC eliminates clock jitter will NOT remove this noise. The fact that it eliminates clock jitter caused by the transport clock and S/PDIF interface is very impressive, however THE DESIGN OF THE NAIM DAC ONLY RELIEVES SYMPTOMS, BUT DOESN’T EFFECT A CURE.

This is why transports do matter.

Hi Fatcat,

My (very limited) understanding is however it is introduced to the system (motor, power supply, whatever) jitter manifests itself only as timing errors in the digital bitstream. What other "noise" can there be in the digital domain, apart from making a "1" a "0" or vice-versa? (and if this happened to any extent, you'd likely get no music at all.)

The DAC reclocks the bitstream and so removes the jitter introduced between the reading of the disk and it's arrival in the DAC buffer, however it was caused - the DAC can't "know" how the jitter was introduced, still less care.

Darrell

quote:

Originally posted by DarrellK:
Hi Fatcat,

My (very limited) understanding is however it is introduced to the system (motor, power supply, whatever) jitter manifests itself only as timing errors in the digital bitstream. What other "noise" can there be in the digital domain, apart from making a "1" a "0" or vice-versa? (and if this happened to any extent, you'd likely get no music at all.)

The DAC reclocks the bitstream and so removes the jitter introduced between the reading of the disk and it's arrival in the DAC buffer, however it was caused - the DAC can't "know" how the jitter was introduced, still less care.

Darrell

Wot he said....

quote:

Originally posted by Andy S:

quote:

Originally posted by DarrellK:
Hi Fatcat,

My (very limited) understanding is however it is introduced to the system (motor, power supply, whatever) jitter manifests itself only as timing errors in the digital bitstream. What other "noise" can there be in the digital domain, apart from making a "1" a "0" or vice-versa? (and if this happened to any extent, you'd likely get no music at all.)

The DAC reclocks the bitstream and so removes the jitter introduced between the reading of the disk and it's arrival in the DAC buffer, however it was caused - the DAC can't "know" how the jitter was introduced, still less care.

Darrell

Wot he said....

Definatley, Wot he said


Manifest : To show or demonstrate plainly; reveal:

Jitter reveals the presence of noise.

quote:

Originally posted by fatcat:

quote:

Originally posted by Andy S:

quote:

Originally posted by DarrellK:
Hi Fatcat,

My (very limited) understanding is however it is introduced to the system (motor, power supply, whatever) jitter manifests itself only as timing errors in the digital bitstream. What other "noise" can there be in the digital domain, apart from making a "1" a "0" or vice-versa? (and if this happened to any extent, you'd likely get no music at all.)

The DAC reclocks the bitstream and so removes the jitter introduced between the reading of the disk and it's arrival in the DAC buffer, however it was caused - the DAC can't "know" how the jitter was introduced, still less care.

Darrell

Wot he said....

Definatley, Wot he said


Manifest : To show or demonstrate plainly; reveal:

Jitter reveals the presence of noise.

Wot I really said:

"jitter manifests itself only as timing errors..."

In other words (taking your definition of "manifest"): timing errors (in the bitstream) reveal the presence of jitter.

I suppose it would be more accurate if I had said: "jitter *causes* (only) timing errors in the bitstream presented to the DAC."

Anyway, semantics aside, If I can be so bold as to sum up the thread so far:

Assuming (a) the nDAC removes all (post production) jitter, and (b) its DAC and analogue sections are not compromised in some way by the s/pdif signal which is introduced to the buffer, the source of the s/pdif signal (aka "the transport") *cannot* make a difference.

No sustainable argument against the above statement has been made. So we are left with two possibilities:

1. at least one of the two assumptions above is false, or
2. any differences heard between different transports are due to placebo or a badly-executed comparison.

For some reason Sherlock Holmes has come into my head: "when you have eliminated the impossible, whatever remains, *however improbable* must be the truth"

Anyway, let's just assume

1. That after all the *transport induced* jitter is eliminated.
2. There are still electrical noise introduced by the transport via the coaxial cable

The Naim nDAC white paper states:

"How to overcome the problem of S/PDIF noise entering a DAC
The Naim DAC’s high-speed DSP (digital signal processor) front-end is electrically isolated from its high-resolution DAC and analogue circuits. Also, the two sections are run from separate power supplies. Together these measures *significantly reduce* the digtal RF noise which could affect the analogue stage."

Since problem (1) is solved. And for problem (2) the above statement states that the Naim DAC DSP is "electrically isolated" from the analogue circuit and these meashres "significantly reduced the digital noise".

My question then is that since Naim seem to have gone into greatly length to reduce digital noise, shouldn't the differences between transports, assuming that they exist, be vanishingly small?

And are these differences worrisome enough to buy a new CDX2S for instance?

quote:

Originally posted by fatcat:
Jitter reveals the presence of noise.

Nope. Jitter produces distortion when you convert from digital to analogue. That's all it does. Lower jitter -> lower distortion -> cleaner reproduction of original sound.

As I've said before (and Darrell said too), there is no way of determining where the jitter comes from when you receive the data stream at the SPDIF interface - if you eliminate jitter at the SPDIF interface, you have eliminated all transport induced jitter period.

quote:

Hi Thorsten,

Obviously, you hear differences in different transports used with the DAC, and you are not interested in why. That is fine, and no-one has criticized you for it. But why do you persist in making the same point over and over again? Don't you believe that the rest of us have a right to discuss this? Why does the discussion annoy you so much?

Easy.

A:

"I hear a difference when I use different sources on the nDAC."

B:

"This can´t be. You are under the influence of a mass-placebo-effect."

Extremely scientific.

quote:

Quote: Andy S:

No, it's doing the upsampling at least plus it will be monitoring what is going on in the system (buffer levels etc) i.e. controlling the way the DAC works. If it were as simple as you suggest, the DAC wouldn't need to have a DSP inside it

No Andy, I don't think this covers it. The upsampling operation is tied to the nDAC master clock, so the noise produced from the DSP during this process is clocked to the nDAC master clock. On the point about monitoring what is going on in the buffer memory, the white paper says ‘The SHARC DSP monitors the rate at which the RAM buffer is either filling or emptying’. This operation again seems to be tied to either the nDAC master clock or the S/PDIF clock. As long as a DSP operation is tied to either clock then the point I raised still holds. What other operations is the DSP doing (when playing music) that is not tied to either clock so that the ‘jitter ghost’ in the output signal is masked with noise?

I totally believe that everyone has a right to such a discussion, if it really IS a discussion.

quote:

Originally posted by gav111n:
No Andy, I don't think this covers it. The upsampling operation is tied to the nDAC master clock, so the noise produced from the DSP during this process is clocked to the nDAC master clock.

The DSP is driven by a separate clock to the one delivering the data to the DAC (it's even in the nDAC architecture document on p4). The two are decoupled and will be all over the place with respect to each other. The bit noise that you are talking about (if it exists) will be lost in the noise (pun intended ) of all the other digital signals being switched on the DSP board.

The DAC master clocks are only used to pull the data across the isolators - everything else the DSP does will be decoupled from the master clocks.

quote:

Since problem (1) is solved. And for problem (2) the above statement states that the Naim DAC DSP is "electrically isolated" from the analogue circuit and these meashres "significantly reduced the digital noise".

My question then is that since Naim seem to have gone into greatly length to reduce digital noise, shouldn't the differences between transports, assuming that they exist, be vanishingly small?

Why is NAIM still not answering these even to me ( ) very interesting questions?

Andy S.,
please don´t say: They want to make more money.

OK, OK, Andy, you win 'DAC'.

Your prize is a CDS3!

It should make a beautiful plinth for your laptop.

Gav.