Uniti Atom/Nova/Star - New firmware - March/April 2018 Comments

Can we please stay on topic ref SQ issues post firmware changes?

dlb27 posted:

Can we please stay on topic ref SQ issues post firmware changes?

Do you have any? 

Alll threads tend to drift, but yes, perhaps we should stick to SQ.

Just a quick link to a post from Trevor Wilson on the ND* 4.6 thread that reiterates his and Naim’s commitment to understanding the reasons for the divergent views on Uniti 2.6 firmware.

https://forums.naimaudio.com/to...42#75342530622519042

Eoink posted:

Just a quick link to a post from Trevor Wilson on the ND* 4.6 thread that reiterates his and Naim’s commitment to understanding the reasons for the divergent views on Uniti 2.6 firmware.

https://forums.naimaudio.com/to...42#75342530622519042

Good to see - I was unsure until that if they'd actually twigged that older device users were also split on improvements.

Eoink posted:

Just a quick link to a post from Trevor Wilson on the ND* 4.6 thread that reiterates his and Naim’s commitment to understanding the reasons for the divergent views on Uniti 2.6 firmware.

https://forums.naimaudio.com/to...42#75342530622519042

Hi

I read Trevor Wilson's post with interest, especially his reference to " we aren't ignoring those few customers on the 2.6 release that have raised concerns "

I have been in dialogue with Naim technical support regarding problems I have experienced with the 2.6 update in terms of my Nova/Unitiserve and the feedback so far is that, in terms of the missing artwork they are working on a fix but in relation to sound quality issues they are happy with the outcome and that very few, in relation to the 15,000 users, Atom/Star/Nova, have complained.

glasnaim posted:

Eoink posted:

Just a quick link to a post from Trevor Wilson on the ND* 4.6 thread that reiterates his and Naim’s commitment to understanding the reasons for the divergent views on Uniti 2.6 firmware.

https://forums.naimaudio.com/to...42#75342530622519042

Hi

I read Trevor Wilson's post with interest, especially his reference to " we aren't ignoring those few customers on the 2.6 release that have raised concerns "

I have been in dialogue with Naim technical support regarding problems I have experienced with the 2.6 update in terms of my Nova/Unitiserve and the feedback so far is that, in terms of the missing artwork they are working on a fix but in relation to sound quality issues they are happy with the outcome and that very few, in relation to the 15,000 users, Atom/Star/Nova, have complained.

So if there are 15,000 users we must represent well under 0.05 % who have actually posted concerns.

Could that reflect problems with our hardware?

Hi Alleycat

I haven't done the math but yes a very small subset, I don't know if it reflects a problem with our hardware, might be a possibility, I just thought it would be a useful insight to Naim's assessment of the situation.

glasnaim posted:

Hi Alleycat

I haven't done the math but yes a very small subset, I don't know if it reflects a problem with our hardware, might be a possibility, I just thought it would be a useful insight to Naim's assessment of the situation.

Yes it certainly is, and perhaps reflects why there doesn't seem to be much interest if  >99.9% are happy....

To be honest I was very surprised by the numbers but in fairness to Naim I get the feeling they are listening and trying to ensure there are no known issues. Sound quality unfortunately doesn't have any objective assessment criteria, it is at the end of the day down to Naim and the beta testers to decide, and I guess customers if they stay or go elsewhere. Naim's response was they wouldn't put out something they weren't happy with.

Eoink posted:

Just a quick link to a post from Trevor Wilson on the ND* 4.6 thread that reiterates his and Naim’s commitment to understanding the reasons for the divergent views on Uniti 2.6 firmware.

https://forums.naimaudio.com/to...42#75342530622519042

The results that you have posted on 4/28/18 8:05 PM strongly suggest a software problem. Differences in sound quality are perhaps just a symptom of this problem. The fact that only a minority of users have been affected by this symptom would be rather immaterial in this context: what would matter is that upgrading does not yield consistent outcomes. In this situation, it would be possible that 99.9% of the users that have upgraded are perfectly fine with the sound quality of their devices and, still, that many of these 99.9% are running different firmware versions! This could potentially hide problems that could become apparent in upcoming firmware upgrades. Thus, it is only natural that Naim want to fully understand what is going on.

Alley Cat posted:

nbpf posted:

nbpf posted:

Nick Lees posted:

Alley Cat posted:

Thanks nbpf that's very helpful.

Since 2.6 BubbleUPNP Server doesn't do gapless Qobuz with the new Unitis (I use Kazoo on an iPad). It's being looked at I believe. I still plays nicely with the old streamers under 4.6 thankfully (for me).

Just to avoid confusion: I do not use Bubble UPnP Server ...

I can confirm that Qobuz gapless replay works flawlessly under upmpdcli. I am listening to ...

Many thanks - much appreciated - sorry to ask another detail, can you confirm this is streaming Qobuz titlesfrom the internet not downloaded files on your LAN.  

Yes, this is Qobuz streaming through a 55,5 Mbit/s (download) connection. Gapless replay works just fine via upmpdcli, as one would expect. Best, nbpf 

Very helpful - I must give it a try.

Alley Cat posted:

Very helpful - I must give it a try.

You can easily try upmpdcli on a Raspberry Pi. With an Allo DigiOne hat, you have a very competent (server +) S/PDIF renderer solution for about 150 EUR, see also https://forums.naimaudio.com/topic/allo-digione-player. I understand that the DigiOne it is not at the NDS niveau but it is still quite good and certainly much more flexible. If you plan to do a lot of internet streaming via wireless, I suggest that you get a Raspberry Pi 3B+. The B+ model has a very decent support for the 5GHz band and the CPU can be run at up to 1.4GHz. I have set it to constant 1GHz and when transcoding 24/192 .flac to .wav, the CPU loads never exceed 20%. You can use the DigiOne with Volumio and dietpi but I very much prefer a minimal headless Raspbian distribution with no web interface. Back to the topic now and let's hope that Naim have meanwhile sorted out the problems with the firmware upgrades! 

nbpf posted:

Alley Cat posted:

Very helpful - I must give it a try.

You can easily try upmpdcli on a Raspberry Pi. With an Allo DigiOne hat, you have a very competent (server +) S/PDIF renderer solution for about 150 EUR, see also https://forums.naimaudio.com/topic/allo-digione-player. I understand that the DigiOne it is not at the NDS niveau but it is still quite good and certainly much more flexible. If you plan to do a lot of internet streaming via wireless, I suggest that you get a Raspberry Pi 3B+. The B+ model has a very decent support for the 5GHz band and the CPU can be run at up to 1.4GHz. I have set it to constant 1GHz and when transcoding 24/192 .flac to .wav, the CPU loads never exceed 20%. You can use the DigiOne with Volumio and dietpi but I very much prefer a minimal headless Raspbian distribution with no web interface. Back to the topic now and let's hope that Naim have meanwhile sorted out the problems with the firmware upgrades! 

Thanks - we have a couple of Pis I need to get to grips with.

A very quick look at upmpdcli did not suggest an easy option to install all software components on a Mac so the Raspberry Pi could get a good alternative.

Received the firmware rewind email yesterday from Naim but haven’t bothered as I’m pretty happy with 2.6 on my Atom now I think and don’t want to wreck my head now that I’ve adjusted to it. Just been listening to Dave Brubeck Take 5 with a fair bit of welly in to the A4’s and the drum solo is stunning, best I’ve heard the system I’d reckon, could be helped by the recent addition of the Cisco 2960 and a boggo Ethernet cable but very little not to like. Having gone through the mill on this and been hot and cold re 2.6 I’ve still no idea what was going on, either a genuine technical issue or just a matter of adjustment, I may never know but somehow suspect the latter! 

Dear forum members,

here's another Atom owner who is unhappy with the sound of the 2.6 update.

But first of all I wish to thank everyone who has posted here so far. I've been reading quite a bit around here and it was very helpful and educational for me!

After updating, I too felt that something had gone. The music felt less involving than it did before. There was this overwhelming presentation that I experienced when auditioning the Atom at my dealer, and I could happily reproduce that at home. Until the update... I just felt I lost something. Or was it just imagination?? Then I did some research on the web, stumbled upon this forum and learned that I was not the only one. Following the information given in this thread, I e-mailed Naim support for help to revert to 2.3. This brought that involving sound back that made me buy the Atom. I did try to go back to 2.6, and the change in SQ was immediately obvious to me. Some people mentioned an increase in detail, and a loss in volume with 2.6. However I could not say that 2.3 was less detailed then 2.6, neither is there a notable change in volume. Maybe that's because I use active speakers on the pre-out, and not the Atom's internal power amp? Anyway, with 2.6 the music seems to be more distant, less involving, cooler, with less impact. The warmth is gone. I feel less dynamics. (It's a bit hard to describe that in English as it is not my 1st language).

The comments about 2.6 becoming "better" in SQ added confusion to me. Until now I could not reproduce that when upgrading to 2.6 again. However, Simonpeterarnold mentioned a certain sequence of downgrading, not resetting, then upgrading and then resetting... Seems a bit weird to me that this could make a difference, but who knows? So thank you Simonpeterarnold for posting this detail, I will try it myself and see what happens!

Hope it helps you Robert. No idea myself why it worked all a mystery. 

Alley Cat posted:

nbpf posted:

Alley Cat posted:

Very helpful - I must give it a try.

You can easily try upmpdcli on a Raspberry Pi. With an Allo DigiOne hat, you have ...

Thanks - we have a couple of Pis I need to get to grips with.

A very quick look at upmpdcli did not suggest an easy option to install all software components on a Mac so the Raspberry Pi could get a good alternative.

Just a word of warning: I have done some more Qobuz listening via upmpdcli today and while gapless replay works flawlessly, I ran into the problem documented at Computeraudiophile. It does not happen very frequently but it is very annoying. Best, nbpf

Which problem is that nbpf?  Is it an issue described in  forum there - did a quick search but difficult to pin it down.

Alley Cat posted:

Which problem is that nbpf?  Is it an issue described in  forum there - did a quick search but difficult to pin it down.

Sorry Alley Cat, my original post contained a link to the pertinent CA thread but has been moderated. The problem that I am referring to has originally shown up on Sonore devices: some micro- and ultraRendu users have experienced that, sometimes, replay from internet streams (both Qobuz and Tidal) skips to the next tack well before the current track is actually over. It does not happen frequently but it is obviously very annoying. Yesterday I did extensive Qobuz listening through the upmpdcli Qobuz interface running on my Raspberry Pi (and I can confirm once more that gapless replay works flawlessly with upmpdcli) and had three skips in about 5-6 hours. I looked up the problem, discovered the CA thread and thought I would report back here. You should be able to find the CA thread by searching for "Help Trouble Shooting Skip-to-next-track problem in my Streaming setup" in the "Networking, Networked Audio, and Streaming" Forum on CA. I suspect that the problem could lie in the upmpdcli interface to Qubuz (and Tidal) but I am running additional tests to substantiate or confute this conjecture. If the problem turns out to lie in upmpdcli, I will open a ticket. The upmpdcli developer is very responsive and I am confident that he will be able to readily fix the problem. It is also possible that the problem does not depend on upmpdcli, however. For the time being, I am listening to Rachmaninov's "Bells Symphonic Dances" by Mariss Janssons using the Qobuz interface of the Bubble UPnP control point (instead of upmpdcli's own interface) and streaming to the upmpdcli instance running on the Pi and so far I have experienced no skips. I am only at track 4, however. I will do more extensive tests later and report on the above mentioned CA thread. Best, nbpf

Sorry to sound like a moaning whinny  but this thread is meant to be about SQ issues post firmware changes NOT the intricacies of the various mechanisms you keep on going on about.   Can I please suggest that you start another thread as I am bored of reading this increasingly off topic seemingly private conversation here.  Please respect the topic the thread was raised for originally. 

dlb27 posted:

Sorry to sound like a moaning whinny  but this thread is meant to be about SQ issues post firmware changes NOT the intricacies of the various mechanisms you keep on going on about.   Can I please suggest that you start another thread as I am bored of reading this increasingly off topic seemingly private conversation here.  Please respect the topic the thread was raised for originally. 

Apologies for going off topic. As I mentioned in my post above, I am going to report on the issue Alley Cat is interested in on the CA forum. Thanks for keeping the thread on focus! Best, nbpf

dlb27 posted:

Sorry to sound like a moaning whinny  but this thread is meant to be about SQ issues post firmware changes NOT the intricacies of the various mechanisms you keep on going on about.   Can I please suggest that you start another thread as I am bored of reading this increasingly off topic seemingly private conversation here.  Please respect the topic the thread was raised for originally. 

Sadly I think the original thread discussion has died a death based on the dearth of on-topic posts.....

I'm still unclear quite how Naim are looking into these perceived negative effects for some users - has anyone been asked to provide specific feedback?

Unfortunately I think you are right Alleycat, I don't think Naim are actively pursuing feedback regarding sound quality issues arising out of the 2.6 update, I suspect the opposite. So the choices are stick with 2.6, go back to 2.3 unsupported or go elsewhere.

In relation to the artwork that went missing, if you are a Unitiserve user, Naim seemingly are still working on a fix, no timescale and no update.

I doubt anyone has been asked, roll on the same issues into the NDS555.