Bluesound Node Icon Streamer Review

[Sbrtoy]

Figured I’d chime in since I sent my unit to Amir to test. In non-scientific comparison with the Wiim I felt there was some audible differences, again take as heresy if you like as they aren’t measured but were just a couple weeks of A/B which was roughly level matched. I’d listen for about 30 mins then swap boxes and play the same material. Yes sighted bias but who knows because not everything favored the more expensive unit as that would seem to infer. A few of my thoughts:

The Wiim sounded significantly better with my turntable, frankly the analog in on the Icon subjectively sounded very poor. Perhaps Amir’s measurements support why. It seemed to lack detail while also sounding much harsher than the Wiim, this was my biggest disappointment in the unit.

The Icon and Wiim were very similar sounding streaming FLAC but I found the Icon to have more depth of soundstage consistently and have more specific placement of instruments and vocals. How you relate this to a test measurement I have no idea, admittedly not technical enough for that. The potential IMD that Amir mentions is interesting, obviously I would prefer if Qrono was switchable on the Icon, I am very curious if the poorly measuring filter led to the audible difference in soundstage, but also at what cost? Being able to swap filters would be a big plus to me.

I’m undecided if I will keep the Icon, before I sent it to Amir Dirac wasn’t enabled yet. That was a huge selling point for me and still is, despite me playing with REW and the PEQ on the Wiim for awhile the results haven’t equaled what other Dirac components (NAD) seemed to in the past for me. Those who state “get a Topping and a streamer” aren’t the market target, my days of shelves full of components and cables are done, I want simple, sleek and user friendly in one box with strong performance and was hoping the Icon was it, but maybe it isn’t quite there yet. BlueOS isn’t unbearable for me and my wife has figured it out too, though I admit the Wiim is more intuitive at first glance.

Finally my Icon was dead quiet, comparatively the Wiim was noisier with my ear to the tweeter, however some people have mentioned huge noise issues so maybe I was just lucky.

 

[JSmith]

Figured I’d chime in since I sent my unit to Amir to test.

Good on you for sending the device in for testing.


JSmith

 

[pogo]

Subjective listening test comments have been very positive. Maybe QRONO sounds better but doesn't measure quite as good. The Absolute Sound tested it and commented that the dac in the Icon sounds better than his Topping D70 Pro. Might be slight changes in filter sound. He was very impresssed

Could it be that we are not taking all the necessary measurements for an objective evaluation and that the weighting is wrong?
All sound impressions of the MQA QRONO d2a filter go in the same direction and that can't be a coincidence.

 

[amirm]

No one who has the budget to pay $1000 for a streamer, will hate it.

??? I can easily budget for $1000. But if I do, I want a properly executed system. So while I don't "hate" it, it won't be something I would be buying. So please don't make this about affordability. You don't know the financial position of members here.

 

[wwenze]

You know, this looks like irony to me

For many decades the haters of "digital music" complain about the "staircase waveforms" that digital is supposed to have (not even noting the fact that I'm not seeing any staircase on my monitor screen right now)

Yet by having a bad-to-no LPF, this is exactly what causes the staircase to appear

Or is it that only if the original signal has no staircase then it counts? In which case a slow filter doesn't matter?

 

[Art of sound]

This is a review and detailed measurements of the Bluesound Node Icon Streamer, DAC and Preamplifier (ADC). It is on kind loan from a member and costs US $999.
View attachment 433588
I can't say I am overly impressed by the design as it kind of looks plasticky. There is a nice large display but doesn't support touch which is strange in this day and age. Instead there are a set of touch buttons on top. I did not use any of this and instead of downloaded the BlueOS app. I found it super confusing, taking me almost 15 minutes to figure out how to change inputs! The option is not under "Audio." Nor the "Player." Instead it is under a "Music" button on the bottom! Half hour spent doing usability testing would have shown them this to be an issue. I understand it supports Dirac but I did not play with that. I updated it as it requested on first connection.

Back panel shows impressive set of inputs and outputs which you may have a hard time deciphering due to low contrast labels:
View attachment 433589

I could not test everything but did my best to get wide coverage. I suggest watching my tutorial on DAC measurements if you are not familiar with my testing:

And lease subscribe to the channel.

Bluesound Node Icon DAC Measurements
I started my testing with streaming over Ethernet and XLR balanced output:
View attachment 433590
This is better performance than I expected, showing good level of competence in engineering:
View attachment 433591



View attachment 433592

It is however shy of some of its competitors which peg the above chart. Performance is no different if you use USB or HDMI ARC:
View attachment 433593
View attachment 433594
There are some small sidebands in all the FFT spectrum however. This will show up more in our jitter test. For now, here are the RCA results:
View attachment 433595

The unit was highly dependent on grounding. You can see this reflected in channel 2 having lower performance. This is the best I could achieve. Your mileage may vary.

Dynamic range is excellent:
View attachment 433596

Low distortion in the dashboard translates into the same in multitone and 50 Hz distortion tests:
View attachment 433597
View attachment 433598

Linearity was flat as it should be:
View attachment 433599

As noted, jitter performance is not good:
View attachment 433600

We have both random, low frequency jitter that broadens the "skirts" of our 12 kHz tone. But also a myriad of spurious/jitter tones. This shows lack of design hygiene. Next bit of bad news was the filter:
View attachment 433601
Gosh, we have one of these slow filters which we had seen in earlier Marantz AVRs. They get credit for making this optional in latest series but I could not find any such option in Node Icon menus. I strongly recommend the company reconsider this decision and provide at least one standard filter. Fortunately, the frequency response is not impacted:
View attachment 433602

But wideband THD+N certainly is:
View attachment 433603

Bluesound Node Icon Preamplifier Measurements
There is analog input which I appreciate as I have a Reel-to-Reel tape deck and know others have turntables:
View attachment 433604

While ground loop has now come back with vengeance, performance is still quite good.

Input is digitized using an ADC at 44.1 kHz:
View attachment 433605
Would have liked to see this at higher sample rate or made programmable as their competitors do.

Bluesound Node Icon Headphone Amplifier Measurements
Performance using high impedance of 300 ohm shows that the output power is just a step above "checklist" feature:
View attachment 433606
I like to see desktop products produce at least 100 milliwatts.

There is better performance to be had using low impedance of 32 ohm:
View attachment 433607

Conclusions
It is clear Bluesound has taken a major step forward in design of Node Icon. Alas, they didn't quite get the job done with the horrible choice of DAC reconstruction filter and to some extent jitter and interference in output stage. Analog input is good. Headphone output is mediocre to acceptable depending on your headphone impedance and sensitivity.

Because of the choice of DAC filter, I cannot recommend the Bluesound Node Icon. Company needs to revise this especially since it is just a firmware choice. Should they do this, it would go on my recommended list.
------------
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/

oooh that hdmi earc + sub out is tempting. hopefully it has a cutoff for analog out. sadly no dolby atmos support do decode that into a 2.1 channel properly.

 

[GXAlan]

But if there are zero audible consequences in this case, I can't understand all the lambasting.

… rather than a legitimate concern for the consumer.

Really, help me out, here. Should I be planning to sell my unit, and replace it with something else? It seems like it fills an excellent niche in my setup, and the UI doesn't bother me.

If you like the Node Icon, you should keep it.

There are only three outcomes for QRONO.

1) It’s inaudible and is pure marketing. Think of fancy cables with nice jackets and proper capacitance, inductance, and resistance. There is NO difference in performance but with sighted bias, you might experience an improvement. It’s definitely not a detriment to the sound.

Using Harman research on speakers, you can look at the blind testing and come to the conclusion that preference score predicts consistent preference in both sighted and blinded testing. The 100% equally valid scientific conclusion is that multiple listeners scored certain speakers very highly in sighted listening. It sounded better when they also saw the speakers. Therefore, spending some time making the speaker look pretty is an equally valid way to get happy customers. And this is 100% true. Just look at the popularity of soundbars today.

The experience of the Node Icon sounding better may very well just be sighted bias.

2) It’s worse.
The mathematical argument is that distortion is bad and ringing artifacts are not found in musical content. This leads to a circular argument that if ringing artifacts are not found in musical content, is the kind of higher frequency content also found in musical content?

3) It’s better.
The stance that ASR takes is that Bluesound should prove this to be the case, or fans of QRONO. The burden of proof is on them not ASR.

For #1, money was spent in marketing and software encoding and IP. If it’s inaudible, that’s R&D money that could have been spent elsewhere. For #2, then it definitely requires a “not recommended” score. For #3, we need more “proof” like ABX or blind testing.

The Icon and Wiim were very similar sounding streaming FLAC but I found the Icon to have more depth of soundstage consistently and have more specific placement of instruments and vocals.

Finally my Icon was dead quiet, comparatively the Wiim was noisier with my ear to the tweeter, however some people have mentioned huge noise issues so maybe I was just lucky.

I then went from Fosi ZD3 to the WiiM Ultra and I also noticed more noise with the WiiM than the Fosi, and then when I got the Node Icon, it was clearly the quietest. Again, this is maybe one or two feet next to the speaker and not at the listening position.

@amirm ’s measurement of dynamic range is 125.785 dB for a voltage of 3.691V and 125.157 for a voltage of 3.671V.

0.0020274 mV and 0.0018962 mV residual noise by the math.

The WiiM Ultra is 0.0026465 mV and 0.002562 mV by the same approach.

Is 1.89 uV versus 2.56 uV enough to make an audible difference? If we take a 29 dB gain amplifier, then the extra 0.7 uV is just 0.02 mV which shouldn’t be audible, either.

I also found the Node Icon to be quieter when close to the tweeter. Maybe grounding affects RCA connections more, and I was getting worse real-world RCA noise.

If Amir hasn’t packed up the Icon yet, it would be an interesting subjective test for him go connect it to his Neumann speakers to see if there’s anything he can appreciate in stereo listening about noise.

Could it be that we are not taking all the necessary measurements for an objective evaluation and that the weighting is wrong?
All sound impressions of the MQA QRONO d2a filter go in the same direction and that can't be a coincidence.

This is why we need science. All sound impressions of the Node Icon have been favorable. However, we cannot assume QRONO is the reason the sound impressions are favorable. It could be the extra low noise. It could be a dual mono design. It could be avoiding intersample overs because it doesn’t run a full 4V so there’s some sort of headroom. What if it was EVEN better with a standard filter? All we know is that subjective reviews of the Node Icon have been favorable and objective measurements have also been favorable, save for the filter.

We should easily be able to record the effects of QRONO by recording the output of the Node Icon playing 44.1 at a higher sampling rate and recording the same thing through a standard DAC and then comparing.

I just haven’t had time to put a test together...

 

[pogo]

The experience of the Node Icon sounding better may very well just be sighted bias.

Bluesound made the comparisons internally and the results were in favor of the MQA QRONO d2a filter. Switching filters was probably considered unnecessary: Link

 

[amirm]

Bluesound made the comparisons internally and the results were in favor of the MQA QRONO d2a filter.

Is there a white paper/write up on this?

 

[Haskil]

Yeah. I get it. It’s a double standard. If we were dealing with Emotiva or Monoprice, we’d just shrug and say that they have no talented software engineers.

Here, WiiM is unfairly held to a higher standard because their software engineers are clearly top notch.

So their thought is: By giving you these permissions, you don’t have to enter your WiFi credentials since we will just pull it. Great. Lovely.

Now you have people saying — yeah… but no one else is asking for that kind of privilege. Can you just add the option of manually entering my credentials? Or let me use Ethernet to setup my device?

Like I said, it’s okay to say that something is the best value and best choice but also push the developer to make it even better.

Can't we really use Wiim without going through Wifi, but in ethernet...

 

[somebodyelse]

BluOS runs everything at root too, so they have bad security practices too.

Have they fixed the arbitrary command injection yet? That one's beyond merely 'bad' and it sounds like it wasn't the only one like that.

 

[TNT]

No. I already stated I'm not a technical expert. I'm merely reporting my understanding of the issue.

In a technical oriented forum there is no need for persons that know that they have a limited technical knowledge (quite some...) to report their understanding. It would make this place unbearable... and confusing ... and misleading...

The least you could have done is to phrase it like a question!?

//

 

[Kolllak]

Is there a white paper/write up on this?

QRONO white paper

https://mqalabs.com/wp-content/uploads/2024/12/MQA-Labs-QRONO-White-Paper_updated.pdf

We started with a question, “what was missing from digital music?”. Perhaps we should have asked, “what was added by digital capture and playback?”. We now know that the ADCs used in recording studios, and our DACs used for playback, added time blur from the ringing in their filters.QRONO d2a playback of a CD-level audio master has the time response of conventionally played 96kHzHi-res file.

A 192kHz hi-res file played with QRONO d2a can exceed the time performance of the best analogue systems. The audible improvement, while subtle, is immediate. Removing the time smear reveals textures previously obscured. Micro dynamics are improved, and instruments are clearly delineated, enhancing thestereo soundstage and image. Music flows naturally, adding realism and reducing listener fatigue. Listen for yourself.

 

[anphex]

Why does changing devices to 192 Khz suddenly clean up the high frequency distortion due to the slow filter? I always thought DAC filters are always just working at multiples of F(s).

And when you change to 192 Khz and play a 44,1 Khz file, aren't the bits just filled, causing the file to not get the necessary oversampling filter and getting aliasing in the high frequency?

I am super confused about this right now.

 

[Matt_Holland]

Bluesound doesn't give a shit about this product, so why should we?

Is this a serious comment?
If so I think it’s unfair and almost certainly incorrect.

 

[AdrianusG]

Subjective listening test comments have been very positive. Maybe QRONO sounds better but doesn't measure quite as good. The Absolute Sound tested it and commented that the dac in the Icon sounds better than his Topping D70 Pro. Might be slight changes in filter sound. He was very impresssed

" The Absolute Sound tested it and commented that the dac in the Icon sounds better than his Topping D70 Pro. Might be slight changes in filter sound. He was very impresssed"

I bet he was, after all they were very impressed with MQA as well back then, all subjective opinion without any real proof, so no surprises here.

 

[Mark411]

Thanks for the review! I recently purchased a Node Icon and can offer my perspective on what made it a no-brainer for my particular use case and why its price didn’t seem out of line for me (a recently retired married guy in a smallish house). I use the Icon along with an Apollon Purifi 1ET6525SA ST amp, KEF LS-50 Meta speakers, and a Bluesound Pulse Sub+. The Icon and amp live out-of-sight in a cabinet.

1. It runs BlueOS. After having an older Node and an NAD C399 with BlueOS module, my wife and I have gotten used to and actually like the app despite its “quirks” and are loathe to change. The app has been rock-solid stable, runs well on iPhone, and is frequently updated and well-supported.
2. It integrates wirelessly with the Pulse Sub+. This is a great sub for me. I have in-home auditioned at least six other subs and the combination of form, size, and sound of the Pulse hit the sweet spot. Because of its location, running a sub cable is out of the question. I’ve tried three wireless adapters and found them either unreliable or to hum/buzz. In contrast, the integrated wi-fi connection to Bluesound devices is reliable and silent.
3. It has integrated Dirac support. I find it really easy to calibrate for various listening positions and love the ability to switch instantly between up to five filters in BlueOS. (I don’t think I need to convince ASR members of the benefits of equilization!)
4. It has an IR in jack. I know, “a what??” My hi-fi doubles as an AV receiver and I love the ability to control the TV, Roku box, and Icon with a single Logitech Harmony remote when streaming movies or watching OTA TV.

Of course, now I’m bummed to discover from this review that my new Icon has a sub-optimal reconstruction filter, which keeps it out of the “recommended” category. Can someone help me understand the impact of this filter choice? Is it more like “for $1K you should expect a better filter even though the sound difference is inaudible” or more like “this unit would sound better if the filter didn’t suck” or even as one commenter implied, “this crummy filter will fry your tweeters”?

Thanks! --Mark

 

[morillon]

the end result ,finallity , is subjective...
if 7 out of ten people prefer this approach... so much the better, and the other 3 will just go for something else...
that's all...
;-)

 

[Mikig]

Devices honest, as demonstrated by the tests with lights and shadows in the overall performance.

I think if I had to look for a streamer, Dac, preamp, at that price, I would go for Eversolo.

Thanks Amirm as always for the test!!

 

[somebodyelse]

It has an IR in jack. I know, “a what??” My hi-fi doubles as an AV receiver and I love the ability to control the TV, Roku box, and Icon with a single Logitech Harmony remote when streaming movies or watching OTA TV.

Isn't this more about:

The Icon and amp live out-of-sight in a cabinet.

Hence you need an external IR receiver because a built in one won't be able to see the IR from the remote.