Hmm I’m not enjoying mine ATM. The bass is now too much combined with a harsh top end at times and lost the prominent voices it once had.
Powered off, lots, factory reset, repositioning etc. I just used REW to analyse my system last night and sent to Thierry at HAF and he confirms.my listening from this and that my system seems to have a v shaped signature. Before update the bass was never this pronounced but the highs have always been an issue even on the last one. Will see what he can come up with to counter act this, until the next update I guess and it will no doubt change it again. If this changes significantly again then I will strongly consider finding a new amplifier system as I am not going to out up with things changing every update, it’s ridiculous.
I did a lot of frequency response tests with my Atom. I always used RCA connection. And several times I used WAV files (through Atom USB front input), but the results were the same.
The response curves were always exactly the same, irrespective of how Atom sounded when playing music. I measured it at worst and best sound quality cases – curves were the same. Also frequency response curves were the same with different updates (2.3, 2.6, 2.6.1, 3.2). I do not count very small differences. I think they are because of different measuring time, and slightly different mic position and other factors. Curves with 1/24 smoothing always looked exactly the same.
So I am confused why SQ changes are not visible in frequency response curves, although they are obviously heared when playing music. Maybe it is somehow related to DAC work. Maybe frequency response signal is very simple task for DAC and it sounds the same, unlike music which has huge transitions and more complex signal curve.
I have to admit, that I‘m extremely happy with 3.2 update. Two weeks with stable SQ, without random SQ fluctuations. Also I’ve never experienced such high SQ on my Atom.
But I forgot to mention one thing (maybe very important):
I disconnected all inputs (using Naim app). I left only HDMI and servers. I think this also could be somehow related to SQ changes as software is involved. Maybe some active threads are destroying SQ, and these threads are (possibly) dependent on inputs.
Inputs disabled made no difference at all to sq last then I tried and they really shouldn’t. If they do then as far as I am concerned the unit is flawed. I heard about this on the streamers and pres. It’s ridiculous, if inputs effect sq fix them or don’t have them at all.
That’s the main argument in the German Hifi-Forum, Naim section. There are some trolls (blind testing Vienna group) telling you all the time there is no difference between any amplifier, between cables … because is not possible to measure it. What you can not measure, there is no difference. We all even know the difference to a FW update . Surely there is a difference between amplifier.
(The German Hifi-Forum is very exhausting, because many (German) people just think that they know everything better . The people from UK are more liberal .).
Software updates shouldn’t alter sound quality also, but they do. Also there are a lot of “bugs” left. So there are obvious problems with software in Uniti range players. I wouldn’t be sure that disabling inputs does nothing :). I think Atom should be restarted after disabling inputs (the same procedure as after installing 3.2 update). And it should be given some time (several days, or week) to stabilize and to see if differences exist - this is a unique feature of Unitis :).
Still talking with the support guys regarding the failing remote on my Atom after the update - he mentioned they got a few reports on this, so I was wondering how widely spread this issue is?
My issue is that the remote only manage to register a couple of inputs, then the volume ring blinks and not registering any additional input again before the lights go out in the remote.
All things I could think about has be tried, factory reset, battery swap, changed from wifi to ethernet, paring the remote again - no change.
It’s very disheartening as this amp was a fair bit of an investment for me, and it’s hard to have such a prolonged issue with this beloved amp, and it’s put a bit of a damper on things. Some things can be done from the iOS app, but a lot of navigation ex in the settings, has to be done via the remote. It’s a bit of a bad design flaw in my book, that one can become so crippled with the remote out of play.
Hopefully this will be sorted, I just had to vent a little - what a strange update this has been…
I’ve only had my Atom for less than a month and have not noticed any fluctuations in sound quality so far. After reading this whole thread, I’m hesitant to update the software so I’ll probably hold off for awhile. I’m hoping that Naim get it ironed out eventually so that software updates are not going to always be an issue down the road.
Well i just got my test back from HAF, what a difference it makes, this is now what I remember my Atom sounding like when I got it, vocals front and centre and balanced. Naim to me have dropped the ball with the last two firmwares somewhere or my 2nd unit is also knackered like my first, The bass is too extended for it own good and highs too harsh. With HAF EQ wow the feet are tapping and not wanting to turn it off.
I take your point, but I continue to read (throughout the forum) about the many and varied problems that the Nova updates generate - it is not just this specific matter …
I sincerely hope that one day I’ll have the confidence to get mine out of its box
Home Audio Fidelity, run by a guy called Thiery. You give him a number of measurements from eq software such as REW with a calibrated mic and he analyses them and will asses what can be done to get a balanced sound. Since 2.9.1 I have had issues but last one has taken bass out of control in my room and the treble is still harsh like in 2.9.1.
He will then create convolution filers for DSP to be applied to make the relevant adjustment. You supply him a wav files of a piece of music you know well and perhaps covers the best range and he then apply the filters to it and return it for you to evaluate. If you like it then you have the choice to purchase the full service and he will give you the filters and tweak a little to your liking if need be.
Well with that concern obviously in the front of your mind, if I were you I would sell the Nova and buy something that I didn’t have to worry about, maybe from Currys.
Best
No I just got the test back today. Bass has been an issue since 3.2, treble been an issue since 2.6.1 so now I have both to contend with so its but it way off balance. There is a 10gb gain at around 35Hz which is the bass boost which has come from somewhere.
I still maintain that 2.3.8574 is the last ‘best’ musical firmware that keeps my toes tapping on the Nova, I keep reverting to it, and prefer it to the more detailed/forward/harsher newer variants - each to their own, it’s just a shame all users cannot access older firmware versions easily (earlier ones may have been even better, but all updates after 2.3.8574 don’t work as well sonically for me personally, and I think there was a small number of us on the old forum that felt this way) .
I have a very big peak at 50Hz (about + 20 dB), but I’m sure it is room mode. I checked with REW room simulation tool. It is very simple tool, but it showed exactly the same peak after simulation.
Maybe your 35 Hz problem is also related with room mode? From what I’ve experienced, I’m sure that my Atom didn’t play 35Hz at all. Just after new update I began feeling this deep and controlled bass. And I’m really happy with it.
In your situation, seems that deep bass occurred later. And (possibly) it turned out to unwanted room mode. But it’s just a guess :).
Or maybe you are experiencing some kind of temporary SQ fluctuations in Atom, again…
. Surely there is a difference between amplifier.
. The people from UK are more liberal 
.).
