Adding DAC to NDX. Go for used NDAC or new Chord Mojo?

I am using a Mojo with SN2/HiCap DR and I am simply impressed with this little DAC, I haven't heard the nDAC, but even if it is better than the mojo, that's not an issue given the price difference. With digital, the technology is moving so fast that we simply can't keep up with, and we shouldn't be surprized about dramatic changes in performance to price ratios. All good for us consumers, more pressure on the manufacturers to deliver more for less!

Antonio1 posted:

 

 

Hype in my opinion is also evident in lots of online reviews and that doesn't spare Naim too , hype is what leads former 202 owners to move to  nac272  thinking its an improvement, wrong assumption. 

 

You might take a bit of flack for making such a suggestion here.......although personally, I agree with you!!

ChrisSU posted:

Antonio1 posted:

 

 

Hype in my opinion is also evident in lots of online reviews and that doesn't spare Naim too , hype is what leads former 202 owners to move to  nac272  thinking its an improvement, wrong assumption. 

 

You might take a bit of flack for making such a suggestion here.......although personally, I agree with you!!

Insinuating that forum members can be given over to bouts of gushing hyperbole and exaggeration? Surely not.

Antonio1 posted:

Hype in my opinion is also evident in lots of online reviews and that doesn't spare Naim too , hype is what leads former 202 owners to move to  nac272  thinking its an improvement, wrong assumption. 

 

So 272 is not improvement over 202?

strange view, what is this based on?

Just stirring the pot I'd suggest, by bringing something totally unrelated to the thread into the mix. 

to put a different spin on Emre's point above -- I can readily believe that the NDAC does its thing better than the DAC in the NDX. what I find hard to fathom is how the NDX would supply the digital "ones and zeroes" to the DAC better than, say, a Sonos, which is far less expensive and has a great user interface. (full disclosure: I run a Sonos into the DAC in my SN1.)

joe9407 posted:

to put a different spin on Emre's point above -- I can readily believe that the NDAC does its thing better than the DAC in the NDX. what I find hard to fathom is how the NDX would supply the digital "ones and zeroes" to the DAC better than, say, a Sonos, which is far less expensive and has a great user interface. (full disclosure: I run a Sonos into the DAC in my SN1.)

...yet it does, by quite some margin ime.

G

I've seen the 1s and 0s argument put forward before.  It's very easy to prove to yourself that digital transports sound different, just plug your mobile phone, your sky box, and anything else with a digital output into the same dac and compare.  They all sound different.  Even a PC will sound different from a Mac Book Pro and a Mac Mini different to that.  I've no idea why buut it's easy to hear

Dayjay, are you able to  point me to a post or URL that describes this '1s and 0s' argument please. Because despite earning a relatively healthy living from being a computer/IT engineer  I have never understood this argument and so am genuinely curious to read it.. unless it is of course an erroneous oversimplification taken from an explanation of a concept? thanks

Simon-in-Suffolk posted:

Dayjay, are you able to  point me to a post or URL that describes this '1s and 0s' argument please. Because despite earning a relatively healthy living from being a computer/IT engineer  I have never understood this argument and so am genuinely curious to read it.. unless it is of course an erroneous oversimplification taken from an explanation of a concept? thanks

 

me too. I appreciate that in a binary system you are looking for an "on" or "off" state or possibly a "+" or "-" state or even a "1" or "0" state. But I have never imagined that a file of music on a CD, DVD, USB or any other device was as clear-cut as a string of 1's or 0's 

Mayor West posted:

Surprising but I'm sure Andrew Everard posted on here a while back saying Mac Mini/Mojo was giving NDS/555 a serious run for its money. 

Not quite what I said. I actually wrote that:

"FWIW I've just set up a Mac Mini (bought used off eBay for £250) for DAC reviewing, and I'm currently using it through a Chord Mojo into my system, controlled using screen sharing from my MacBook Air. It's surprisingly good, if not NDS/555 PS good"

Don Atkinson posted:

Simon-in-Suffolk posted:

Dayjay, are you able to  point me to a post or URL that describes this '1s and 0s' argument please. Because despite earning a relatively healthy living from being a computer/IT engineer  I have never understood this argument and so am genuinely curious to read it.. unless it is of course an erroneous oversimplification taken from an explanation of a concept? thanks

 

me too. I appreciate that in a binary system you are looking for an "on" or "off" state or possibly a "+" or "-" state or even a "1" or "0" state. But I have never imagined that a file of music on a CD, DVD, USB or any other device was as clear-cut as a string of 1's or 0's 

Indeed - but the state is meaningless with out some concept of when or where, therefore I can imagine a model of 1,0,t or 1,0,z where t is a continuous variable of time or z is a discrete relative displacement as in a static file structure or data unit structure. And of course where t is a continuous variable it becomes analogue

Simon

there's no doubt whatsoever that i'm oversimplifying things. but to put my question in a different way: why would an NDX improve the sound, vs. a Sonos or similar device, when you put it in the signal chain between your router/NAS and your DAC? it just seems odd to me that a CD rip that "leaves" your NAS would be different when it "arrives" at your DAC, depending on whether there's an NDX or Sonos in the middle. 

To put it differently - the amount of things that can go wrong / interference is less with NDX.

I steongly encourage you to test different difital transports and hear for yourself. It's like with power cables - diffcult to appreciate until one hears it for himself.

joe9407 posted:

there's no doubt whatsoever that i'm oversimplifying things. but to put my question in a different way: why would an NDX improve the sound, vs. a Sonos or similar device, when you put it in the signal chain between your router/NAS and your DAC? it just seems odd to me that a CD rip that "leaves" your NAS would be different when it "arrives" at your DAC, depending on whether there's an NDX or Sonos in the middle. 

 

Fair question, but let's leave out conducted common mode noise.. and the Sonos is not that quiet, but the transport clock is important. Noise in the frequency domain in the clock will cause system cross talk intermodulation in the receiver. The tighter the clock the less cross talk artefacts or noise produced.

Simon

Don Atkinson posted:

I appreciate that in a binary system you are looking for an "on" or "off" state or possibly a "+" or "-" state or even a "1" or "0" state. But I have never imagined that a file of music on a CD, DVD, USB or any other device was as clear-cut as a string of 1's or 0's 

But actually that's exactly what represents the music, just ones and noughts, whether that be pits or absence of pits on a cd surface, or electrical signals that typically may be 5V (=on, representing 1), or 0V (=off, representing 0), or magnetic recording on a hard disk. 

the problems come when, for example, voltages are not exact, so instead it is taken that anything more than 2/3rds of the suppy votage (3.33V in this case) is a 1,  and 0 is less than 1/3rd (1.67V). Solved? But what if something interferes with the voltage, for example superimposed RF inferference or other electrical noise, modulating the digital voltage stream?Or if it stays on or off a touch too long due to timing innacuracy or interferences - it could get counted as a second 1 or 0 that didn't really exist, making the binary value completely different.

So what matters is preserving the 1s and 0 s without electrical interferences or timing variations (nothing to do with the T of PRaT!) - Simon's explanations say the same thing in more computer-related language.

Innocent Bystander posted:

Don Atkinson posted:

I appreciate that in a binary system you are looking for an "on" or "off" state or possibly a "+" or "-" state or even a "1" or "0" state. But I have never imagined that a file of music on a CD, DVD, USB or any other device was as clear-cut as a string of 1's or 0's 

But actually that's exactly what represents the music, just ones and noughts, whether that be pits or absence of pits on a cd surface, or electrical signals that typically may be 5V (=on, representing 1), or 0V (=off, representing 0), or magnetic recording on a hard disk. 

the problems come when, for example, voltages are not exact, so instead it is taken that anything more than 2/3rds of the suppy votage (3.33V in this case) is a 1,  and 0 is less than 1/3rd (1.67V). Solved? But what if something interferes with the voltage, for example superimposed RF inferference or other electrical noise, modulating the digital voltage stream?Or if it stays on or off a touch too long due to timing innacuracy or interferences - it could get counted as a second 1 or 0 that didn't really exist, making the binary value completely different.

So what matters is preserving the 1s and 0 s without electrical interferences or timing variations (nothing to do with the T of PRaT!) - Simon's explanations say the same thing in more computer-related language.

So, it's not as clear cut as a string of 1's and 0's, there are complications.

Some of which you have highlighted.

Some DACs are better than others at minimising these complications.

Seams like we are agreed. At least in principle.

Simon-in-Suffolk posted:

joe9407 posted:

there's no doubt whatsoever that i'm oversimplifying things. but to put my question in a different way: why would an NDX improve the sound, vs. a Sonos or similar device, when you put it in the signal chain between your router/NAS and your DAC? it just seems odd to me that a CD rip that "leaves" your NAS would be different when it "arrives" at your DAC, depending on whether there's an NDX or Sonos in the middle. 

 

Fair question, but let's leave out conducted common mode noise.. and the Sonos is not that quiet, but the transport clock is important. Noise in the frequency domain in the clock will cause system cross talk intermodulation in the receiver. The tighter the clock the less cross talk artefacts or noise produced.

Simon

That might be correct. Who knows?

 But, I agree with Dayjay, even though the 1 and zeros are all present and correct, ultimately the sound quality is related to the quality of the equipment delivering them.

Simon-in-Suffolk posted:

Dayjay, are you able to  point me to a post or URL that describes this '1s and 0s' argument please. Because despite earning a relatively healthy living from being a computer/IT engineer  I have never understood this argument and so am genuinely curious to read it.. unless it is of course an erroneous oversimplification taken from an explanation of a concept? thanks

 

Probably not Simon, I've been in many a forum where there have been people prepared to swear blind that 1s are 1s and 0s are 0s.  Either they reach the dac or they don't is the argument and it irritates the life out of me because it's almost as easy to disprove as the all amps sound the same argument.  If you want a laugh just google it and I'm sure it will crop up

Cheers DayJay, I guess mass web usage and simplified wikis might be to blame.. It's good in one way as perhaps to many there is less 'magic' around, but bad in another as it has bred ignorance and bigotry.

The funny thing is a 1 or 0 is an abstract term, and with Manchester encodings which are common in bit streams such as SPDIF and Ethernet transport a '1' value can be a '0' and a '0' value can be a '1' value  as it depends what the value of its predesccor was in time ... Love to see those suffering from oversimplifitus explain that one    But that tends to be ignored by  marketing overviews as its 'detail' and 'technical'.

Simon-in-Suffolk posted:

joe9407 posted:

there's no doubt whatsoever that i'm oversimplifying things. but to put my question in a different way: why would an NDX improve the sound, vs. a Sonos or similar device, when you put it in the signal chain between your router/NAS and your DAC? it just seems odd to me that a CD rip that "leaves" your NAS would be different when it "arrives" at your DAC, depending on whether there's an NDX or Sonos in the middle. 

 

Fair question, but let's leave out conducted common mode noise.. and the Sonos is not that quiet, but the transport clock is important. Noise in the frequency domain in the clock will cause system cross talk intermodulation in the receiver. The tighter the clock the less cross talk artefacts or noise produced.

Simon

Thanks, Simon. That makes sense -- or at least as much sense as this stuff can make to a non-technically-inclined person such as myself. Thinking some research might be in order, I dug up an old Stereophile review of the Sonos. The measurements section of the review mostly dealt with the Sonos's DAC, but I thought the following excerpt might be relevant:

"I recorded the ZP80's digital output on my laptop, using a Metric Halo FireWire interface to turn the S/PDIF data into SD2 files. These were then translated to AIF files, which were compared, bit for bit, with the original AIFs that had been used to generate the ALC versions. There were no differences, showing that Apple Lossless, the Sonosnet encoding and encryption, and the ZP80's data processing are all transparent to the original data."

Does this mean that there were no issues to speak of regarding the transport clock, or could performance still be improved by a better one, such as the one the NDX presumably uses?

Thinking about this a bit further (no pun intended), suppose you had a high-resolution camera, took a raw image with it, then connected the camera to, and displayed the image on, your laptop. Is there a box you could insert between camera and laptop that could make the image more aesthetically pleasing, even if the data that goes into and out of the box remains the same?

It would be fun to experiment with this stuff sometime. Our hobby seems totally insensible to 99% of the population, and I too have heard differences that some find implausible. I bought some cryo-treated "Tibia" power cords with beefy-looking plugs from AV Options here in the States, and I thought they sounded better, though many might say that a meter of power cable can't improve upon what's transmitted by miles of cable from the power plant to one's home.

Simon-in-Suffolk posted:

Cheers DayJay, I guess mass web usage and simplified wikis might be to blame.. It's good in one way as perhaps to many there is less 'magic' around, but bad in another as it has bred ignorance and bigotry.

The funny thing is a 1 or 0 is an abstract term, and with Manchester encodings which are common in bit streams such as SPDIF and Ethernet transport a '1' value can be a '0' and a '0' value can be a '1' value  as it depends what the value of its predesccor was in time ... Love to see those suffering from oversimplifitus explain that one    But that tends to be ignored by  marketing overviews as its 'detail' and 'technical'.

Hi Simon, You know I always tune in when you are speaking, and we kind of communicate without direct dialogue. Not to much of a technical guy, but a Math Major, so I guess I tend to think in a logical manner!  So on one end I'm a sucker for the relationship between  the price and quality thing, you know,  Ndac vs Hugo.  But deep down inside, I really admire the gent who points out that CDs are not "watered down", and the human ear can only hear sounds up to about 20kHz. So I really tend to gravitate towards the argument that sample rates higher than the 44.1kHz can't be deciphered by the human ear! 

Hence, I'm just now considering  entering the world of Streaming on the Cheap!  2nd hand Ndac $1800 USD

Audirvana Plus > MM > hiFace Two converter > Naim DC 1 Digital Connector > Ndac / Cdx2 > 282 > SCDR > 3 x 250.2 > Briks

I love you Engineers! 

X  raise to the (n + 1)/ (n + 1) 

Out! 

Allante93! 

Hi Allante93, interesting comments. With regard to you sample rate bandwidth and pitch comment... You are absolutely right regarding pitch, and of course our pitch bandwidth declines as we get older as well, however we hear more than pitch, we hear the timing between sounds, this resolution exceeds our pitch bandwidth by a fair degree. Thetefore a bandwidth of upto 20kHz and with hidef much higher when used in a sampling context can be of use in portraying natural sounds way beyond our ability to hear the equivalent pitches. Our brain and ears are fantastic organisms.

Simon

Years ago, (when I was learning a bit about GPS, and more importantly, how to teach it,) it struck me that the encoded signal received from each satellite could be likened to a stream of music. Say, “The Star-spangled Banner” from satellite No1 and “God Save The Queen” from satellite No2 etc.

GPS receivers know the code (or “Tune”) transmitted by each satellite. When the code from a satellite is received it has to be compared with the code stored in the receiver. Assume (for the moment) that the tune in the satellite and the tune in the receiver start at the same time. By progressively delaying the version stored in the receiver until it is synchronised with the version that was transmitted by the satellite, you can measure the travel-time and hence the distance between satellite and receiver.

Now, it isn’t that simple, of course. For one thing, the clock in the receiver is naff compared to the four atomic clocks in each satellite. But with four (or more) satellite signals and a bit of arithmetic, you can work out the x, y, z co-ordinates of the receiver and the precise atomic time as well.

I visualized each satellite signal as a “Bar-code” – a bit like the sort you see on stuff in the grocery store that go “Bleep !” at the check-out. Synchronising two of these looks easy. But is it ?

You can line up the black bars (ie the 1’s) and the white bars (ie the 0’s) but……..there is an awful lot of slack.

Even though there is little doubt that the Bar-codes are synchronised, the leading edges and hence the precise timing can be mis-aligned. In GPS terms this is significant. Just as significant is the fact that the naff clock in GPS receiver has to be re-synchronised with the satellite clocks many times each second in order to get meaningful position and time results. The old arithmetic has to be done pretty quick!!

In music terms, I guess that “clocking” and “arithmetic” are just as important as in GPS. ie you know it’s a 1 or a 0, but just exactly “when” do you release it ?

Perhaps some music systems are better at working out “when” to release a sound and “when” to suppress a sound than others ?

And I bet it’s a lot more complicated than this analogy !

Joe947, I think in your description you were comparing relative sample values in files, IE a big matrix values. If bit 'bit perfect' is maintained then the matrices should be the same or equivalent. If you turn that matrix into a discrete series of values at a point in time IE a data stream then time becomes a key attribute to. Now we could call discrete time interval variations of this  data stream transport jitter. Many  digital audio devices have effectively removed this transport jitter from causing sample rate jitter. However this variation in discrete timing intervals will in the real world cause crosstalk in the receiver circuitry as it receives and processes this data. This crosstalk can cause EM interference, ground plane and voltage bus modulation. At an abstract level this is systems engineering. IE an input produces an output plus error, an input plus error, produces a square and sum of the input and output errors plus the output signal... Ok simplified but hopefully by not too much.

Simon