NDX2

Ali, thanks.. though even the term bit perfect is kind of meaningless with out the time variable... a discrete stream when applying to PCM samples is about representation of a value at an infinitely small point in time .. therefore  the value is determined at a point in time.. so the clock time reference becomes an intrinsic part of the value.. so ‘bit perfect’ should really be ‘bit and timing perfect’ when referring to streams of data. Bit perfect makes sense when referring to static data like a file or a block of memory.

Hmack, the accuracy of the data with respect to the transport fron end kind of to my mind fits into the following.

1) obviously  data corruption... unlikely.

2) modifying or converting the data such as changing sample rate or sample word size... causing data rounding errors.

3) software codecs such as AAC, MQA, MP3 etc 

4) stream conversion .. ie converting say a USB, SPDIF, Ethernet RTP framing , to discrete sample stream or new framing transport stream such as SPDIF.

Now action 4) shouldn’t change any static values but it may impose variation in timing variables. These timing variables can be truncated to tighten the timing or even reclocked altogether. The digital switching and processing to do this creates noise on ground planes, powerline and even creates small electromagnetic fields.

These perturbations can and almost circertainly couple and modulate other functions, albeit manufacturers like Naim go to great effort to decouple this to very small levels, especially on their higher end products.

Now one such function being modulated could be an output data stream clock.. such as a SPDIF, USB transport framing clock or i2s stream to internal DSP and/or DAC. The levels of modulation would be minuscule, but are noticeable to many of us. Therefore in the limit unless idealised perfect decoupling isolation and decoupling is achieved.. a noisy data stream into a streamer transport or DAC is going to sound different from a clean stable transport stream.. and indeed most of us tend to notice this.

analogmusic posted:

I’m wondering the same. Is the output of various streamers bit perfect ?

for instance the auralic Aries has 4 filter settings so is the data changed by these 4 filters ?

chord do insist that no software should change the data to get the best out of the DACs

or at least the streamer should give the option of a bit perfect I.e no change to the original source data.

Simon I’m sorry but I’m not able to understand the advantages of ndx2 over nd5xs 2 when used purely as a transport ?

Once one as filtered the digital data stream, then the data values have been modified, and therefore clearly is not ‘bit perfect’ any more, it becomes ‘bit modified’.

Regarding the differences between ND5XS2, NDX2 and ND555 transports, is really cost and engineering effort. Although there are differences, the higher value products go to greater extent to decoupling through bigger and better PSU, spacing shielding and design.. and as per my post above.. hopefully you can now see the relative critical importance of that in a transport front end feeding an internal or external DAC.

But if it hurts your head.. don’t worry just trust what you hear and use that to determine what is best for you... which is what I certainly do.

But forgive me, I need to dash, I am buying my new NDX2 this afternoon.. getting ready to head to my dealer.

SongStream posted:

 

...Songstream, please see my earlier posts, I want the display.. the ND5XS2 has no display, so not sure how that could be an option for me. The internal PSU in the NDX2 is of a slightly higher quality/size too which clearly is important when trying to extract as much as you can from the architecture. The same held true in differences between NDX and ND5SX in transport mode... perhaps this is an area you have not really explored yet?

With regard to ‘redundancy’ in terms of weight and space and cost of raw componentry is arguably less in using an external DAC rather replacing the internal PSU...... but the key thing you might be missing is that these products are designed to operate in one of two modes..... which can be user configured ..  which is why I choose them ... so your point of so called redundancy is perplexing .. it’s how the products are designed!!.... but anyway.. isn’t that nice that Naim give you the options on how you can use their products... it’s certainly one of the reasons of why I buy them.

It's an area I've not explored at all, in terms of actual listening, hence the curiosity.  My system is....how do I put this....unconventional.  It has a Windows 10 PC as the digital transport.  There are many good points, and the sound quality vs anything else I've heard is unrivaled, but I'm not the kind of person that's out at dealers to audition stuff all the time.  What I am looking for is a digital transport with superb SPDIF output, to handle music from NAS and Qobuz.  In theory the perfect product is the Auralic Aries, but does it have the talent claimed?  Trusting in Naim as I do, I've been considering adding the ND5XS2, as I have no interest in a display on the device itself, and figure it should be able to deliver very close to the performance of the NDX2 as a transport.  Based on your experience it seems you disagree.

Hi, I demo’d the NDX 2 & ND5XS2 as digital transports into my Hugo TT and personally found the difference in performance to be very small. My original plan was to upgrade my NDX to the NDX 2 but after hearing both the new streamers side by side through the Hugo TT I decided to go with the ND5XS2. I have to say I was surprised as I was fully bought into buying the NDX 2 prior to the demo but in the end for me the difference in performance didn’t justify the extra cost of the NDX 2, also from a purely aesthetic view, I prefer not to have a display, so an added bonus.

analogmusic posted:

I’m wondering the same. Is the output of various streamers bit perfect ?

for instance the auralic Aries has 4 filter settings so is the data changed by these 4 filters ?

chord do insist that no software should change the data to get the best out of the DACs

or at least the streamer should give the option of a bit perfect I.e no change to the original source data.

Simon I’m sorry but I’m not able to understand the advantages of ndx2 over nd5xs 2 when used purely as a transport ?

The Manchester encoding that governs digital signals relies entireky on time. Each bit is represented as change over a defined time period, not as a discrete signal denoting 1 or 0. As a result, the degree of drift between any two endpoints that can cause jitter decreases as the data rate increases at the transport layer. High data rates therefore, can be thrown even by length and quality of cable. 

This also means capturing a reclocking the stream fron a buffer is not a silver bullet. This method does a lot stop the compounding of errors going further locally over (what we hope) is a short distance of cm with a common clock. But this happens after data has arrived from a more distant endpoint and however that happens will rely on good clocking and cabling.

Audio applications aside, these problems actually get more frequent lately with low cost devices pushing massive amounts of data these days. The data might originate bit perfect  but whether the receiving endpoint interprets it correctly or even avoids signal sync drops due to cheap and nasty clocks and cabling is another matter. Just see what happens when you push 4K 60fps video down a 5m long cheapo hdmi cable. 

"Bits are bits" is really meaningless. It is about producing a source data stream that is controlled and steady enough to not be misinterpreted by the receiving endpoint. Even in IT (where too much if this ill informed "bits are bits" mantra comes from) considerations of clocking and cabling do come into data centre architecturing where massive throughput is concerned. At least it does in ones run properly.

analogmusic posted:

 I have ordered a Allo digione signature and and should receive it very soon.

Still though, with for example the uptone Linear power supplies one for the clean and one for the dirty side we are looking at an additional 400 dollars for the clean side and an additional 925 dollars for the dirty side. It seems the LPS 1.2  doesn't have enough current capability to power the dirty side, maybe as it seems to be a bank of capacitors running side by side.

Yes I am aware batteries can be used, but the point is the power supply does make a difference even in a streamer, and although I cannot afford the 5000 GBP retail of an NDX2, the ND5XS 2 has exactly the same streaming board, and has a more affordable cost, closer to the auralic aries G2.

I had a look at pictures of the internals of the DCS network bridge, which shows that it has a made in UK Linear toroidal power supply, but when compared in visual size to the one inside the NDX2, it seems to  look  a lot smaller but otherwise the streaming board of the DCS is massive compared to the one inside the NDX2. 

All options like Auralic, Aurender, DCS and Naim still expensive though, which is why for the time being I am trying the allo digione and saving up for the Hugo Mscaler which makes a considerable improvement to my Chord Dave.

As for USB based solutions, for a Chord DAC at least, I like (to my ears) a SPDIF connection more compared to a noisy laptop (even if on battery) into USB.

As for NDX2 sounding better than NDX, if bits are bits and the power supply almost the same, for a Chord DAC it should make little difference (as it reclocks the bits anyway and has excellent jitter rejection)... but the NDX2 does seem to turn off the DAC section, and DAC Chips are noisy, so maybe that could be one reason why the NDX2 sounds a lot better?

Maybe the NDX2 network card is less noisy than the one inside the NDX... 

It seems to be quite complicated to a non-engineer like me, so once I purchase the Mscaler Hugo for my Dave, the Naim ND5XS2 seems to be an attractive solution.....

I have measured the power required by the dirty side of the DigiOne Signature: for my system it is between 500 and 800mA while transcoding 24bits/192kHz .flac fiels to .wav.

Thus, a LPS-1.2 should be enough to power the dirty side of the DS. It is more than enough to power the clean side that requires less than 100mA.

That said, I do not think that it is worth investing too much for powering the dirty side of the DS. An ifi iPower 5V/2.5A would be more than enough. But for the clean side I would recommend a good PSU and the LPS-1.2 is certainly one.

Brand new NDX2 now running in..... playing the PJ Harvey John Peel sessions via Roon... rather good..

Simon-in-Suffolk posted:

Brand new NDX2 now running in.....

Congratulations Simon....enjoy

Simon-in-Suffolk posted:

Brand new NDX2 now running in..... playing the PJ Harvey John Peel sessions via Roon... rather good..

Great news Simon! Congratulations! Enjoy 

Simon-in-Suffolk posted:

Brand new NDX2 now running in..... playing the PJ Harvey John Peel sessions via Roon... rather good..

Congratulations, enjoy the NDX2 and ... allow its DAC also to run in!

ChrisDH posted:

 

Hi, I demo’d the NDX 2 & ND5XS2 as digital transports into my Hugo TT and personally found the difference in performance to be very small. My original plan was to upgrade my NDX to the NDX 2 but after hearing both the new streamers side by side through the Hugo TT I decided to go with the ND5XS2. I have to say I was surprised as I was fully bought into buying the NDX 2 prior to the demo but in the end for me the difference in performance didn’t justify the extra cost of the NDX 2, also from a purely aesthetic view, I prefer not to have a display, so an added bonus.

Interesting, when you say the difference was very small, did you actually hear a difference at all?

i'm just asking as sometimes we are careful not to make statements that will create controversy and so call no difference a small difference. Of course one man's small difference is another man's "night and day", "jaw dropped" etc. moment.

.sjb

nbpf posted:

Simon-in-Suffolk posted:

Brand new NDX2 now running in..... playing the PJ Harvey John Peel sessions via Roon... rather good..

Congratulations, enjoy the NDX2 and ... allow its DAC also to run in!

Indeed.. it’s DAC has been running constantly and is sounding very enjoyable indeed.. so far so good.. 

Congratulations Simon 

Sloop John B posted:

ChrisDH posted:

 

Hi, I demo’d the NDX 2 & ND5XS2 as digital transports into my Hugo TT and personally found the difference in performance to be very small. My original plan was to upgrade my NDX to the NDX 2 but after hearing both the new streamers side by side through the Hugo TT I decided to go with the ND5XS2. I have to say I was surprised as I was fully bought into buying the NDX 2 prior to the demo but in the end for me the difference in performance didn’t justify the extra cost of the NDX 2, also from a purely aesthetic view, I prefer not to have a display, so an added bonus.

Interesting, when you say the difference was very small, did you actually hear a difference at all?

i'm just asking as sometimes we are careful not to make statements that will create controversy and so call no difference a small difference. Of course one man's small difference is another man's "night and day", "jaw dropped" etc. moment.

.sjb

Hi, I don’t think I said anything controversial, just in my personal opinion when being used as a transport the difference to me wasn’t big enough to justify going with the NDX 2. Yes there was a difference but just not enough for me.

ChrisDH posted:

Sloop John B posted:

ChrisDH posted:

 

Hi, I demo’d the NDX 2 & ND5XS2 as digital transports into my Hugo TT and personally found the difference in performance to be very small. My original plan was to upgrade my NDX to the NDX 2 but after hearing both the new streamers side by side through the Hugo TT I decided to go with the ND5XS2. I have to say I was surprised as I was fully bought into buying the NDX 2 prior to the demo but in the end for me the difference in performance didn’t justify the extra cost of the NDX 2, also from a purely aesthetic view, I prefer not to have a display, so an added bonus.

Interesting, when you say the difference was very small, did you actually hear a difference at all?

i'm just asking as sometimes we are careful not to make statements that will create controversy and so call no difference a small difference. Of course one man's small difference is another man's "night and day", "jaw dropped" etc. moment.

.sjb

Hi, I don’t think I said anything controversial, just in my personal opinion when being used as a transport the difference to me wasn’t big enough to justify going with the NDX 2. Yes there was a difference but just not enough for me.

Perhaps I phrased myself clearly. I’m wondering the size of the difference perceived. Would you clearly recognise it in a blinded comparison do you think?

thanks

John. 

A question if I may. There is talk earlier in the thread about bit perfect being a not particularly useful term and also that the transport is critical ito music recovery in a digital system.

If data is taken from a transport in an optical signal to a DAC (accepting CD standard only as I believe there are HD limitations with Toslink, though I might be wrong) does the electrical isolation afforded by the optical link render the transport less important? 

I don’t really understand the intricacies of digital to analogue conversions so I am happy to be corrected but if the digital ones and zeros are replicated accurately (I.e. information is properly transferred)  and the timing is correct and there is no electrical interference from the transport why does the transport matter?

Please understand I am NOT saying the transport doesn’t matter, mearly trying to understand why it does, particularly in the configuration mentioned above.

The Optical connection provides galvanic isolation (no conductors) so removes the transmission of RFI to the receiving endpoint which could affect analog circuits. That's a bonus.

But the Toslink Audio standard is data rate limited and not suitable for high rate HD multi channel audio. There is also the additional steps involved. Optical is generally provided as the only digital interface on very low cost products simply because it is cheaper. But electrical-optical/optical-electrical transducers introduce their own potential for inserting errors, causing noise, or introducing jitter.

That said, the downsides are somewhat overstated. Modern high end products use very good off the shelf Toslink Audio ports. It is really for you to decide which you prefer. I use optical for everything that is not my main source simpliy because cheap long optical cables have significantly less negative impact on sound quality than cheap coax cables (both make a difference but definitely the impact is less). It makes sense since my optical sources all need >5m cables. Personally, if the source was important to me, I'd sit it much closer to the DAC/streamer and use coax with a very decent cable - but that's just me. As it stands, for the computer and TV and so forth I use cables in the $5/meter range.

Sloop John B posted:

ChrisDH posted:

Sloop John B posted:

ChrisDH posted:

 

Hi, I demo’d the NDX 2 & ND5XS2 as digital transports into my Hugo TT and personally found the difference in performance to be very small. My original plan was to upgrade my NDX to the NDX 2 but after hearing both the new streamers side by side through the Hugo TT I decided to go with the ND5XS2. I have to say I was surprised as I was fully bought into buying the NDX 2 prior to the demo but in the end for me the difference in performance didn’t justify the extra cost of the NDX 2, also from a purely aesthetic view, I prefer not to have a display, so an added bonus.

Interesting, when you say the difference was very small, did you actually hear a difference at all?

i'm just asking as sometimes we are careful not to make statements that will create controversy and so call no difference a small difference. Of course one man's small difference is another man's "night and day", "jaw dropped" etc. moment.

.sjb

Hi, I don’t think I said anything controversial, just in my personal opinion when being used as a transport the difference to me wasn’t big enough to justify going with the NDX 2. Yes there was a difference but just not enough for me.

Perhaps I phrased myself clearly. I’m wondering the size of the difference perceived. Would you clearly recognise it in a blinded comparison do you think?

 

thanks

John. 

Hi, as mentioned, yes I could tell them apart when I went back and forth but as I said, to my ears the difference was pretty small. I really liked the sound of the NDX5XS2 and personally thought it to be better than my NDX so decided to go with it as opposed to my original plan of the NDX2. All personal preference. No idea what it sounds like using the internal dac or how it compares as have no plan ever to use it like this.

Just  commenting on a point by Feeling Zen.. Galvanic isolation is nothing to do with RFI or any other high frequency... galvanic isolation will not necessarily impede alternating signals, unless very low frequency. The whole point of galvanic isolation is that it provides a DC break, this means the grounds are disconnected, there is no DC path through. However AC signals may pass through the galvanic isolation coupling largely un impeded. A common use of galvanic isolators is to break ground path so as to avoid a large induction loop.. which otherwise would typically induce a low frequency mains voltage/hum into the loop.

So to the point from Mattnbarns, whether the transport uses SPDIF or I2S either electrically or electrically, it makes no difference to the function and importance of the transport.. the clocked transport frames are the same in both. Although technically there is a slight distortion of the transport framing phase with many Toslink connections and so a slight data dependent modulation of the transport clock can be introduced  with Toslink. This modulation will be coupled into the receiver’s front end switching circuitry. This is one reason you may prefer coax or Toslink on the same quality digital transport source. Internally the higher end Naim streamers use quality  optical couplers to disconnect the ground planes between the DSP and the DAC so as to keep any ground plane modulation decoupled... nut these are very different from Toslink connections.

Simon-in-Suffolk posted:

Just  commenting on a point by Feeling Zen.. Galvanic isolation is nothing to do with RFI or any other high frequency...

I never said it did. Avoiding RFI is simply a beneficial by product of the fact optical provides galvanic isolation by virtue of simply being a non conducting link.

On a related note, has anyone tried wifi on the new streamers? 

feeling_zen posted:

Simon-in-Suffolk posted:

Just  commenting on a point by Feeling Zen.. Galvanic isolation is nothing to do with RFI or any other high frequency...

I never said it did. Avoiding RFI is simply a beneficial by product of the fact optical provides galvanic isolation by virtue of simply being a non conducting link.

Thanks for clarifying, some do get RF filtering and galvanic isolators mixed up thinking they are the same thing.. however some if most galvanic isolators won’t stop RF.. it’s worth remembering than AC couples via electromagnetic fields, it doesn’t need a physical conductor to pass it.. after all that is how transformers work. But depending on the inductance and therefore the reactance of the galvanic link, HF/RF will be proportionally attenuated with frequency where the signal is represented by electrical current.

Like Engjoo, I would be very much interested in hearing somone's experience with WiFi and NDX2 / ND555.

I posted on this topic here :
https://forums.naimaudio.com/t...28#78019131158064928

engjoo posted:

On a related note, has anyone tried wifi on the new streamers? 

 

I used it on my Atom for a while before wiring it, and it seemed to work very well, as one might hope given that they use 802.11ac, rather than n, or even g in the old streamers. The WiFi on the new ND streamers is presumably different, as they appear to have dual antennae unlike the Unitis. 

I can’t comment on any sound quality effects of WiFi vs wired, as I don’t tend to listen that critically to the Atom in a second system. 

ChrisDH posted:

Sloop John B posted:

ChrisDH posted:

Sloop John B posted:

ChrisDH posted:

 

Hi, I demo’d the NDX 2 & ND5XS2 as digital transports into my Hugo TT and personally found the difference in performance to be very small. My original plan was to upgrade my NDX to the NDX 2 but after hearing both the new streamers side by side through the Hugo TT I decided to go with the ND5XS2. I have to say I was surprised as I was fully bought into buying the NDX 2 prior to the demo but in the end for me the difference in performance didn’t justify the extra cost of the NDX 2, also from a purely aesthetic view, I prefer not to have a display, so an added bonus.

Interesting, when you say the difference was very small, did you actually hear a difference at all?

i'm just asking as sometimes we are careful not to make statements that will create controversy and so call no difference a small difference. Of course one man's small difference is another man's "night and day", "jaw dropped" etc. moment.

.sjb

Hi, I don’t think I said anything controversial, just in my personal opinion when being used as a transport the difference to me wasn’t big enough to justify going with the NDX 2. Yes there was a difference but just not enough for me.

Perhaps I phrased myself clearly. I’m wondering the size of the difference perceived. Would you clearly recognise it in a blinded comparison do you think?

 

thanks

John. 

Hi, as mentioned, yes I could tell them apart when I went back and forth but as I said, to my ears the difference was pretty small. I really liked the sound of the NDX5XS2 and personally thought it to be better than my NDX so decided to go with it as opposed to my original plan of the NDX2. All personal preference. No idea what it sounds like using the internal dac or how it compares as have no plan ever to use it like this.

Forgot to mention is was the dealer who suggested I should listed to the ND5XS2, as he was impressed with it.

I'd booked the demo for the NDX2 only.

It's definitely worth a listen.

For reasons I wont go into here, I was obliged to run my NDX2 using Wifi, for a couple of weeks, over the festive period. It was perfectly stable and no significant difference, to my ears, in sound quality.