Bluesound Node Icon Streamer Review

[Koeitje]

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

Why would that be incorrect? When designing a device that does digital to analog conversion selecting a proper filter is part of the proces. They seemed to have selected one at random. So either they didn't care or they didn't understand what they were doing. I don't know which is worse.

 

[kyuu]

Why would that be incorrect? When designing a device that does digital to analog conversion selecting a proper filter is part of the proces. They seemed to have selected one at random. So either they didn't care or they didn't understand what they were doing. I don't know which is worse.

No, it's actually worse than either of those. They chose a poor filter on purpose, as it's part of the MQA cruft that's apparently at the "heart" of their audio pipeline. That's also why they won't be updating the firmware to even give the option for a proper filter.

In other words, it's broken by design.

 

[Koeitje]

No, it's actually worse than either of those. They chose a poor filter on purpose, as it's part of the MQA cruft that's apparently at the "heart" of their audio pipeline. That's also why they won't be updating the firmware to even give the option for a proper filter.

In other words, it's broken by design.

 

[mr_bill]

I doubt that they selected the filter by random - that's a big company with global sales, some very high end products and many engineers on design and development.

 

[pogo]

If people would listen to an Icon, some of the comments here would certainly be different.

 

[kyuu]

If people would listen to an Icon, some of the comments here would certainly be different.

If anyone performed a controlled, double-blind listening test and was able to consistently pick out the Icon as sounding better than, say a Wiim Ultra, that would be interesting. Sighted, subjective listening however is not evidence of anything. Otherwise, we'd have to believe in expensive "audiophile" power cables since there's quite a lot of subjective listening that says they improve the sound from an audio device.

I doubt that will happen though as there's no reason to think one transparent DAC should sound different to another transparent DAC (assuming the poor filter and poor jitter performance doesn't result in any audible difference on the Icon). MQA never presented any evidence from controlled listening of its differences (and actually actively tried to prevent anyone from doing proper testing of it), and it doesn't seem likely Lenbrook will be any different with its perpetuation of MQA.

 

[darth]

I returned mine partially due to ground loop hum I posted in the older thread when connecting to genelecs. But I honestly did not hear a sound quality difference between the icon or Wiim. I got poor quality ears and tested poorly.

Recorder

recorder.google.com

Bluesound releases 2024 ICON streamer

I don't have the Buckeye amp (yet), I have the Icon connected to a SMSL AO200 MKII Amp using XLR cables then onto KEF LS50 Meta's. I am not experiencing this buzz.

audiosciencereview.com

 

[beefkabob]

I understand that ultrasonics can cause changes in audible sound. For me that's 15khz and below, alas. It makes me wonder, do ultrasonics in everyday sound/music cause audible "artifacts" as well? If that's the case, then we should all be listening to 192khz files to get the best sound recreation.

However, clearly a ~44khz file isn't going to introduce the correct "artifacts" if unfiltered or poorly filtered. Still, if we're down 10 db by 24 khz and 40db by 30khz, is it going to matter? A typical filter is down 10db by 22khz, so is this all that different audibly?

Has someone performed multitone tests with a DA/AD loopback and different filters so that we can see measurements of the actual differences?

 

[amirm]

I understand that ultrasonics can cause changes in audible sound. For me that's 15khz and below, alas. It makes me wonder, do ultrasonics in everyday sound/music cause audible "artifacts" as well?

It is very hard for music to contain out of band energy by itself. I have to induce that with using noise for my filter testing. Otherwise music has a brick wall response.

 

[beefkabob]

It is very hard for music to contain out of band energy by itself. I have to induce that with using noise for my filter testing. Otherwise music has a brick wall response.

At what frequency is that brick wall? I'm not arguing a point here. My mind isn't made up at all. I'm trying to understand.

I saw this with a quick search: https://www.semanticscholar.org/pap...oyk/e09b612f8bbd487b174e0701ec18395d5706a0b6n

It claims that there is at least some musical information with some notes on some classical instruments:

My assumption is that, since I can't hear above 15khz, none of that is going to be audible to me. But then I see that audible artifacts can be introduced by ultrasonics. But then I see all these DACs, even this one, filtering out those ultrasonics, abruptly or not.

The filter question definitely appears widely agreed upon, but I'm not clear as to why a slow filter is such a problem. Is there evidence of significant differences in filter speed affecting audible sound in significant ways?

Also, somewhat on topic, 44.1 khz is more an "artifact" of old technology than a scientifically chosen number for optimal results. According to http://www1.cs.columbia.edu/~hgs/audio/44.1.html:

In the early days of digital audio research, the necessary bandwidth of about 1 Mbps per audio channel was difficult to store. Disk drives had the bandwidth but not the capacity for long recording time, so attention turned to video recorders. These were adapted to store audio samples by creating a pseudo-video waveform which would convey binary as black and white levels. The sampling rate of such a system is constrained to relate simply to the field rate and field structure of the television standard used, so that an integer number of samples can be stored on each usable TV line in the field. Such a recording can be made on a monochrome recorder, and these recording are made in two standards, 525 lines at 60 Hz and 625 lines at 50 Hz. Thus it is possible to find a frequency which is a common multiple of the two and is also suitable for use as a sampling rate.

The allowable sampling rates in a pseudo-video system can be deduced by multiplying the field rate by the number of active lines in a field (blanking lines cannot be used) and again by the number of samples in a line. By careful choice of parameters it is possible to use either 525/60 or 625/50 video with a sampling rate of 44.1KHz.

In 60 Hz video, there are 35 blanked lines, leaving 490 lines per frame or 245 lines per field, so the sampling rate is given by :

60 X 245 X 3 = 44.1 KHz

In 50 Hz video, there are 37 lines of blanking, leaving 588 active lines per frame, or 294 per field, so the same sampling rate is given by

50 X 294 X3 = 44.1 Khz.

The sampling rate of 44.1 KHz came to be that of the Compact Disc. Even though CD has no video circuitry, the equipment used to make CD masters is video based and determines the sampling rate.

 

[JIW]

At what frequency is that brick wall? I'm not arguing a point here. My mind isn't made up at all. I'm trying to understand.

I saw this with a quick search: https://www.semanticscholar.org/pap...oyk/e09b612f8bbd487b174e0701ec18395d5706a0b6n

It claims that there is at least some musical information with some notes on some classical instruments:

View attachment 434347

My assumption is that, since I can't hear above 15khz, none of that is going to be audible to me. But then I see that audible artifacts can be introduced by ultrasonics. But then I see all these DACs, even this one, filtering out those ultrasonics, abruptly or not.

The filter question definitely appears widely agreed upon, but I'm not clear as to why a slow filter is such a problem. Is there evidence of significant differences in filter speed affecting audible sound in significant ways?

Also, somewhat on topic, 44.1 khz is more an "artifact" of old technology than a scientifically chosen number for optimal results. According to http://www1.cs.columbia.edu/~hgs/audio/44.1.html:

The ultrasonic content of physical sound sourced are removed by the anti-aliasing filter during analog to digital conversion or downsampling. The ultrasonic content produced by DACs immediately above the passband without filtering are images that are not harmonically related but instead the difference between the sample rate and the pass band content. Depending on the level, this can produce intermodulation distortion in the audible band and damage downstream components.

Instead of just 44.1 kHz, any multiple of 14.7 kHz could have been chosen. This leaves some room for optimisation as 88.2 kHz could have been chosen as well.

 

[beefkabob]

The ultrasonic content of physical sound sourced are removed by the anti-aliasing filter during analog to digital conversion or downsampling. The ultrasonic content produced by DACs immediately above the passband without filtering are images that are not harmonically related but instead the difference between the sample rate and the pass band content. Depending on the level, this can produce intermodulation distortion in the audible band and damage downstream components.

Instead of just 44.1 kHz, any multiple of 14.7 kHz could have been chosen. This leaves some room for optimisation as 88.2 kHz could have been chosen as well.

Yes. But does a slow filter allow intermodulation distortion of significance/audibility versus a fast filter?

Then there is the separate question: If some instruments do make ultrasonic harmonics, and those harmonics then cause intermodulation distortion in the audible band, are those harmonics part of the sound we would want in a recording? Would those audible IMDs be recorded anyway, or would they need to play out in a room to appear?

 

[JIW]

Yes. But does a slow filter allow intermodulation distortion of significance/audibility versus a fast filter?

Then there is the separate question: If some instruments do make ultrasonic harmonics, and those harmonics then cause intermodulation distortion in the audible band, are those harmonics part of the sound we would want in a recording? Would those audible IMDs be recorded anyway, or would they need to play out in a room to appear?

I don't know anything about that in particular. If I recall correctly some research on the audibility of ultrasonics with music found that intermodulation distortion of the tweeter used made it perceptible. This is different from reconstruction filters but indicates possible audibility of low image rejection.

intermodulation distortion of ultrasonics is produced by the recording and playback equipment. Hence, it is not desired.

 

[antcollinet]

Yes. But does a slow filter allow intermodulation distortion of significance/audibility versus a fast filter?

Then there is the separate question: If some instruments do make ultrasonic harmonics, and those harmonics then cause intermodulation distortion in the audible band, are those harmonics part of the sound we would want in a recording? Would those audible IMDs be recorded anyway, or would they need to play out in a room to appear?

Ultrasonics cannot themselves be heard (by definition).

They can cause audible artefacts through IMD - this would be created in the reproduction gear, and then primarily speakers (distortion in speakers is an order of magnitude or two bigger than that of electronics).

So if there are ultrasonics - either in the recording, or as a result of badly filtered reconstruction images - they are never a good thing. It is one of the reasons why high-res music - rather than being a benefit - can actually damage the sound.

I suspect these things are rarely audible though.

 

[morillon]

you speak filter of dac ....but present on the recording in 44.1?

 

[GXAlan]

I have made some recordings of a regular DAC versus the Node Icon. Still running DeltaWave on it, but I will share them for blind testing. I tried two tracks quickly. One seemed to have no ultrasonic imaging artifacts since it was just a guitar and vocal.

I found a percussive sound track with detectable (in Audacity) ultrasonic imaging artifacts seem to be -80 dB or so, so at least with the test track with a lot of percussion.

I’ll create a separate thread with a poll.

 

[GXAlan]

In defense of QRONO d2a?

To be clear, this is just a scientific approach. I'm not sure it's even audible, but seeing these results was worth sharing.

I'm having trouble with doing a blind test. When I volume match in Audacity and save a FLAC, it increases in size tremendously even though I have dithering off, which prevents me from attaching the files to the post.

Test CD: IASCA 2023 Reference CD

IASCA 2023 Sound Quality Reference CD | IASCA Worldwide Inc.

2 Disc set with 'Evaluation' and 'Entertainment' CDs and liner notes. This CD includes music and test tracks for the 2018 to 2023 competition season.For current year's competition tracks, go to https://iasca.com/sqc-tracks/.

iasca.square.site

I chose this since it's a bunch of high quality 16/44.1 kHz content. I chose Track 7 on the IASCA disc, which is Anvil of Crom from Telarc's Great Fantasy Album, a very percussive track. I have that disc somewhere as well, so I could check the liner notes, but this is DDD and done by Telarc, which was one of the best audiophile labels in the 90's. I chose this because I saw the imaging artifacts.

I then recorded:
1) Node Icon from network share --> E1DA Cosmos ADC at 24/96
2) Fosi ZD3 from USB --> E1DA Cosmos ADC at 24/96

Fosi ZD3

Node Icon

There is a huge shelf above 16 kHz which is sort of weird, since I see that on the original FLAC too. Notice the spurious tones at 30 kHz? That's the problem of the QRONO d2a.

So then, it's worth looking at the original source file:

Source File

That's my fault. There isn't a lot of content above 16 kHz to begin with.

So, you'd ASSUME that I chose a poor test track. And you'd assume that all of the content above 16 kHz is buried to -70 dB or lower, and be completely masked.

I then brought things into DeltaWave. Pretty obvious that white is the Fosi ZD3 and blue is the Node Icon.

Nothing surprising yet.

But QRONO talks about phase, so I compared the Fosi ZD3 and Node Icon. So now we are seeing a difference.
Phase: Fosi ZD3 vs Node Icon

This is interesting because the difference occurs in the audible range of 6kHz and up.

So now, the test is to compare the digital file at 16/44 versus the recordings...
Phase: Recording of Fosi ZD3 vs. Digital File

Phase: Recording of Node Icon vs. Digital File

Just before you come to your conclusion.... take a look at the Y-axis. Let's set them to be equal:

Bluesound Node Icon Recording vs. Digital File

Fosi ZD3 Recording vs. Digital File

So, with a test track that is capable of generating imaging artifacts in the ultrasonic range, if you were looking how close the recording of the DAC matches the digital file from the phase standpoint, the Bluesound Node Icon with QRONO d2a actually beats the Fosi ZD3.

 

[pogo]

the Bluesound Node Icon with QRONO d2a actually beats the Fosi ZD3.

You actually need MQA FOQUS to better display temporal blurring. But the tendency is already clearly visible. TNX

 

[amirm]

At what frequency is that brick wall?

As noted, it is at half the sampling rate. Digital processing can create out of band energy but this is the exception, rather than the rule.

 

[Endibol]

What a great selling feature it would be if you could see it on and off. It’s like having Dirac on and off. A camera with image stabilization on or off.

Or a “non-selling” feature if it would turn out no difference could be heard..
I reckon that’s why the option isn’t there.