The effect seem to be on the transport side, since I am using an external DAC with my NDX2+XPS-DR and the change from 3.11 to 3.12 is quite something (within the bounds of reason, of course). The change is also as evident whether I use Qobuz or locally stored .WAV files.
4 Likes
My NDX2 / XPS DR is still running 3.7.0 and I have so far resisted the update urge. I can live without internet radio. Right now I accept SQ could always be better, but a previous update to 3.11 on a (now sold) ND5XS2 resulted in SQ I really didn’t enjoy, so despite all the glowing reports thus far I will sit this one out for now. One day I realise I may have to respond to the firmware update prompt, and maybe that day has come? My system with SN3 / HICAP DR sounds great, and I’m reluctant to risk upsetting the status quo.
2 Likes
So I moved on, and decided to do the upgrade. As expected, the Web Client is no longer working, but this was part of the deal.
I will now spend a bit of time to hear if I can spot a difference, but probably not my best skills.
2 Likes
If you don’t need any of the new functionality, I say don’t do it. I am a Roon user on 3.8.x and do not miss anything. And I get consistent Naim organic sound every time.
4 Likes
I wonder why you are so interested in the new firmware for Naim gear when you don’t own any?
2 Likes
Firmware and hardware obviously will interact with each other and the effect they make on sound quality changes over time.
1 Like
Different folks have different tastes. But here is why I like 3.12 so much more than any previous version (including a very brief stint with 3.8): low noise which means reverberation, sostenuto, una corda, … are clearer and easier to hear (same with different bow techniques on the violin). So, below are two examples of where it matters musically, not just sonically. The first is Lewis’ last (?) recorded version of his 1950s Django composition in his great “Evolution I” album. He plays the tune very slowly and you can hear the sustained chords all the way to the next chord and harmonically combining with it. Clever idea and beautiful rendition of the tune as a result. Now, the slow tempo makes sense. The second is my favorite rendition of “Les feuilles mortes”. Montand uses slow glissando to convey the sadness of his memories (the film where the song was first sung is a post WWII movie that’s pretty bleak). Again, the lowering of noise makes this so much clearer than previously. It gives the impression of time passing. Not trying to convince anybody, but I like it when an upgrade/update makes me understand musical intent better.
11 Likes
Yes, but you will never know whether 3.12 would sound better than what you have now, unless you try it.
Opinion without experience is emptiness.
5 Likes
You know I’d never actually run the NDX 2 without the PSU together with the Elektra.
So in the interest of understanding whether theory meets practice, thought I’d run it today …
The answer is that without PSU the soundstage narrows and is pushed forward into more of a wall of sound. There’s less space between performers/sounds and it feels constrained. Timbres are a little less defined, dynamics, body and the subtleties diminish.
It still sounds good, but very much akin to the difference adding the PSU makes to base NDX 2 if you’ve tried that?
3 Likes
I do still own a naim streamer just not the latest one, plus i have plenty of experience to know how firmware works and naim is no different to anyone else.
1 Like
I wish it were as easy as that. It’s a one-way commitment, as I have previously discovered, to my cost.
So what is that sound like? Everyone talks about the 3.8 as if it were the Holy Grail. What do you mean by the ‘consistent Naim organic sound’? Can you describe it?
1 Like
That reminds me of the late great Christopher Hitchins on a different subject but just as relevant:
“Without evidence, it’s just talk.”
3 Likes
Is it not about time that someone did before and after audio spectrogram measurements. It may help inform the debate. GOBSAT term from medical world comes to mind ‘Good all boys sat at table’ giving their views rather than evidence.
I can’t speak for gk but my own experience lies in what drew me to Naim in the first place in the early 1980s. As a musician and BBC trained sound engineer I had a reasonable idea of what both the live instrument and its reproduction in the studio sounded like. What was, and still is, IMO unique to Naim is the ability to preserve all the pitch and rhythm information such that the system plays in tune and on time. Just like in the studio. Whenever I hear non Naim systems (which these days is infrequently) I am usually taken aback at how poorly the pitch relationships within the music and its timing are portrayed. In that sense Naim sounds natural (organic?) to me, and I love that. I can and do listen for hours.
What 3.11 update did on my previous streamer was disturb that natural portrayal, along with a harsher and brighter timbre. My reaction to that update was: OMG what have Naim done? To sum up: more hifi, less music.
But that’s only my perspective.
11 Likes
Okay, I’m ready to step back into the fray 
Firstly I will say that in my mind there is no doubt this firmware release is a bug fix - to be more precise, solely directed to addressing sound quality issue and particularly with regard to the ND555 - Naim’s reference streamer. There is no doubt that since 3.8 there have been two releases (3.10 and 3,11) that significantly reduced the audio performance of the ND555 and that this was not a sustainable situation for Naim to be in. Personally, I was very frustrated with the situation because I knew from 3.8 that the ND555 was reference quality and was no longer performing at level for those of us that had updated the firmware. I had begun researching alternative products.
Secondly, for context, most of my listening has been from physical media using my CX UHD as a transport for Bluray, SACD and CD. I need to do more listening as a streamer and want to use the Core direct to the ND555using the BNC connections for a comparison.
So, my thoughts on 3.12? The ND555 returns to a reference level streamer and I couldn’t be happier. The harshness of 3.11 is gone. The high level of detail is there, but it is much more defined, distinct and spacious, resulting in a much more realistic and live presentation. This is across the spectrum, very clear top end (but not fatiguing), immersive mid-range and lovely tight and textured bass. My reference material I have listened too is astounding. Is it better than 3.8? I can’t really recall that now to make a honest comparison, but the realism, clarity, texture and drive of 3.12 is really top notch.
So, well done to the software team for finally getting this resolved. I think a lot have learned that streamers carry some risk to the vagrancies of software updates. I hope we won’t go there again, and I’ll be in the wait and see category next time. I don’t expect that we will have any formal communications of how we got here, I just hope that the lessons have been learnt and we have returned to enjoying the magic that the ND555 and other Naim streamers can deliver.
29 Likes
I’ve not had my ND5 XS2 long enough to be able to comment on the ‘sound’ of the firmware.
Imagine if when you bought a new phone that you were told it would take a few weeks to bed in, and that during that time, any payments you made with your phone were subject to +/- 10%… and if you bought one of our specially designed noise cancelling cases it would get more accurate. Repeat that for every upgrade… It’s a wonder sometimes that what you type actually appears at the destination with all that noise between where you type and where it ends up.
In my head, unless the firmware updates the DAC software, I can’t see how the bits can be changed in such meaningful ways. With streamers there’s even less need for data error correction than a CD player as there’s no optical layer that can be damaged in the way of reading said data. So I really struggle to see how a player used only as a transport shows big differences.
Within a streamer that is performing the D to A conversion, if the digital section can impart such an impact to the DAC and analog boards, why are they simply not better isolated? Computers years ago used to have optical connections between key circuit boards, so surely isolating the digital from the analog is not impossible or massively expensive, and for Naim you’d have thought well within ‘budget’.
If 3.12 has indeed got deliberate fixes for SQ that impacts the DAC section then why not just say so? Naim don’t have to give away any secrets. Likewise, if firmware changes to digital code like library updates for security patches can have such an impact on non-digital components, some actual explanation would be very helpful.
So far the “without evidence it’s just talk” line applies to both sides of the debate.
3 Likes
I think it’s simply a case that they won’t officially acknowledge the issue, they are not the most open of hi-fi manufacturers, they are actually very secretive about product road maps and firmware developments.
5 Likes
When 3.10 was introduced, things went side-ways due to extensive changes in third-party software (StreamUnlimited). Things improved from 3.10 to 3.11 and now from 3.11 to 3.12.
For as far as I know, Naim always tunes a firmware release for SQ, once all functional- and technical changes are complete and tested.
It seems that Naim went through a learning curve, finding the best way to tune the new firmware based on that new StreamUnlimited software. Apparently, they now mastered the tuning process with this 3.12 release.
I do n’t know if this is really true, but it seems that way. Anyway, I am happy that all the hard work by Naim worked out!
4 Likes