Adding an xps2 to ndx2 or ndac?

Yes I knew but usually I dont use it.
What I can do, even because I live temporary in egypt in which there is no quobuz or other services, is to use a directly connected iphone with apple music or a pc with a usb spdif converter for my library. I was thinking to use a dedicated music server like core but i m not sure about an increased quality because of the same spdif connection

Naim’s preferred digital connection tends to be electrical SPDIF, in preference to optical or USB, so it’s probably your most useful option for getting the best from the NDAC. The Core could be a good option here, despite its limited verstility in many respects. I would also consider an Innuos server, as this is in effect a streamer too, so it gives you access to internet radio, streaming services (if you can access them!), Roon etc.
If you don’t need a dedicated ripper you might also consider an ND2XS2 which will give you the versatile access to music files on a directly attached USB drive that the NDAC lacks, plus a fully fledged streamer.

I didnt consider nd5xs, it could be a good option but the only streaming service here is apple music which is not supported by any streamer as far as i know.
Innuos could be a good option but only the zen mini mk3 with external lps which has an spdif connection and I could you it only as a ripper player
probably better core in this configuration.
to summarize I m evaluation the best spdif ripper/player and my doubt is if core is a top notch spdif player or if there is something better (like innuos melco or whatever)

The ND5XS2 would get you Apple Music via Airplay. Possibly not optimal sound quality, but it works. Maybe an iOS device connected to the NDAC USB inpit would be sufficient?
You are right that only the Zen Mini has SPDIF. How this would compare to the Core I wouldn’t know.

Better, with Spdif, are Lumin or Grimm MU1, the most expensive here. But they don’t rip. So personally I would choose the Core.

1 Like

EDIT
I made an interpretation error!!! Apologies
Thanks to @soulrider4ever for catching that I have misinerpreted these results.
These are INPUT jitter measurements not output.
I will leave my original post otherwise to own my mistake though.

Thanks again for catching and bringing my mistake to my attention

I do try to be accurate, even on internet forums, so my apologies!

-Nathan


The following is from their Lab Report section in each article.
I removed the price as its no longer accurate, reformatted the NDX2 table because it pasted oddly, and added a header.

To me that difference in jitter is significant.
Who knows if that is ‘normal’ or addressed by a unit to unit variation, but I would hope not.

ND5XS2

Measure Measurement
Maximum output level / Impedance 2.21Vrms / 23-190ohm
A-wtd S/N ratio (S/PDIF / USB-A) 105.6dB / 105.5dB
Distortion (1kHz, 0dBFs/–30dBFs) 0.00065% / 0.00028%
Distortion & Noise (20kHz, 0dBFs/–30dBFs) 0.0095% / 0.0009%
Freq. resp. (20Hz-20kHz/30kHz) +0.0 to –0.28dB/–8.1dB
Digital jitter (48kHz / 96kHz) <10psec / 15psec
Resolution @ –110dB (S/PDIF / USB) ±0.6dB / ±0.6dB
Power consumption 10W (1W standby)
Dimensions (WHD) / Weight 432x70x301mm / 6.6kg

NDX2

Measure Measurement
Maximum output level / Impedance 2.22Vrms / 17-191ohm
A-wtd S/N ratio (S/PDIF / USB) 109.1dB / 109.0dB
Distortion (1kHz, 0dBFs/–30dBFs) 0.00028% / 0.0053%
Distortion & Noise (20kHz, 0dBFs/–30dBFs) 0.0034% / 0.0012%
Freq. resp. (20Hz-20kHz/30kHz) +0.0 to –0.65dB/–15.9dB
Digital jitter (48kHz / 96kHz) 65psec / 105psec
Resolution @ –110dB (S/PDIF / USB) ±0.6dB / ±0.6dB
Power consumption 16W (1W standby)
Dimensions (WHD) / Weight 432x87x314mm / 10kg

Hope this adheres to the rules re: quotes of freely accessible online content.

Thanks very much, I obviously missed that on the phone. Interesting and quite odd.

Maybe worth noting their judgement:

For the NDX2:

Jitter suppression is excellent if not absolutely state-of-the-art [see Graph 2] with evidence of very low-rate jitter (note broadened peak) and correlated sidebands at ±384Hz and ±768Hz (65psec with 48kHz/24-bit data) and extending to ±1152Hz, ±1536Hz, etc (105psec with 96kHz/24-bit data).

And for the ND5XS2:

Jitter suppression, meanwhile, is not only superior to that observed with the NDX 2 but defines the current state-of-the-art at <10psec with 48kHz/24-bit inputs and ~15psec at higher sample rates.

I doubt that this has anything to do with the transport, but I have no idea about what the cause is, considering it uses the same code as per my earlier quote from the reviews (and I believe the same SHARC chip)

It’s nearly an order of magnitude difference as well. I suppose they double-checked, if I got such a difference in measurements without a good explanation my first impulse would be to check the measurement gear. I wish they had expanded on this / asked Naim about it.

1 Like

@Nathan_Klassen I figured we should compare their ND555 results as well:

Although both units [NDX2 and ND555] benefit from Naim’s ‘40-bit’ 16x upsampling/jitter-suppressing DSP, the key measurable advantage is realised in the ND 555 as a near-total elimination of correlated jitter

Digital jitter (48kHz / 96kHz) <5psec / <10psec

I have a hard time believing that the ND5XS2 is nearly up there with the ND555 while the NDX2 is nearly an order of magnitude worse, although all 3 have apparently very similar hardware and code in this area. It just doesn’t add up unless it’s quite an engineering failure in the NDX2. Unfortunately I could not find any other review online with these measurements. I’ll have to wait until I am back from vacation to look at my PDFs of German mags, they usually are quite fond of measurements

Is this just another demonstration that the flat-earth-ists were right? What ears hear is not what lab tests measure?

2 Likes

I don’t think so. Naim would not go this far to eliminate jitter if they didn’t care (and they are quite proud of it, see the ND555 white paper). At most, it shows that jitter at this level (they still judge the NDX2 as excellent) is not that big of an issue because ears are not that great after all.

However, there is no good reason for the NDX2 to measure (that much) worse. Something is amiss here.

2 Likes

I think measurements are useful, but they are not the be all and end all.

There are extremely well measuring products that
 I really don’t enjoy.

There was an interesting article a couple years back out of Stereophile where Atkinson compared a few different speakers that measure extremely similarly that sound quite a bit different.

For something functioning as a pure digital transport - I personally seem to be sensitive to higher jitter transports so it is something I consider.

If those measurements are representative, I probably wouldn’t consider a NDX2 as a pure transport, but I’d at least listen to it.

** Edited to indicate I’d at least listen to it **

1 Like

I’ve had both players. Regardless of any measurements the NDX 2 is a much more grown up player, as it should be, considering its higher price and where it sits within the Naim hierarchy.

The ND5 XS 2 is the balanced player for the ‘XS’ range of components. Whereas the NDX 2 is the balanced player for the Classic range. ND555 for 500 Series.

Naim are not silly, they obviously know where their range of products sit within their performance categories and pricing.

2 Likes

The ndx2 is good, but the streaming board is identical to the nd5xs2? and that the screen on the ndx2 must be adding noise somewhere. should naim just release a streaming board in a stageline box and power it externally. The less noise more signal was the consensus??

Run ND5XS2/nDAC/PS555 through 552/500/DBLs - on par with my CD555 in terms of musical enjoyment :sunglasses: (can only hear up to 14KHz which may be one explanation)

1 Like

Same with ND555 - I keep my screen turned off. However when I have my screen turned on, I hear absolutely no change/decline in sound quality :thinking:

Yes that would be the case , but to compensate naim shielded the streaming board in the nd555 and they’ve put it as far away from the screen as they can. Which means that the screen must be noisy.

1 Like

Naim made the display screen implementation as quiet as they could on the latest streamers according to the presentation they gave at the factory back in 2018.

The faraday cage screening on the ND555 streaming section (which is a source of noise) is more for the benefit of the rest of the player (particularly the sensitive analogue sections) rather than to shield it from the display screen.

2 Likes

It’s questionable , but there’s still a compromise going on somewhere. Nothings perfect .

Would be interested to know how the Nds measure vs Ndx2. Will try to find.

1 Like

The screen isn’t even getting updated (it’s a static picture that is sent once), and whether any remaining screen noise causes jitter is at least to be determined. We also don’t know if Hifi News tested with and without screen enabled, and if it made a difference.

Generally we know way too little and should not jump to conclusions based on guesses. All we know for sure is that at least one measurement by Hifi News had this result. You would not base an opinion or decision on one data point in any other realm, in particular without a factual explanation, so why would you in hifi?

3 Likes