Hello Everyone,
This is a review an measurements of the Marantz NA-11S1 Network Audio Player.
It was send to me by a kind member for review.
Marantz NA-11S1 - Presentation
The Marantz NA-11S1 was released in 2012. It is a massive unit in size and weight, which showcases exceptional craftsmanship. Its cost was roughly $3500, so we are indeed talking high-end.
@John Atkinson reviewed it for Stereophile and you'll find some insights from M. Ishiwata talking about the end of the physical media (poor me) and the beginning of a new streaming era.
Considering the recent announcement regarding Stereophile that @amirm shared, and fact that John announced he's going for (a well deserve) retirement, I thought I'd use this review as an opportunity to make a tribute to his past 40 years of measuring Hi-Fi devices for us. I wanted to thank him for that.
————
The back of the Marantz is busy:
No Bluetooth/wifi, but network and SPDIF inputs (I tested up to 24bits/192kHz) plus USB for connection to a computer (up to 352.8kHz). Analog outputs include RCA and XLR (hot pin 3, ie output is inverted). There are also digital SPDIF outputs.
Sadly, despite joggle button on front face, it is impossible to configure some audio options without the remote control. This is a big miss.
The Marantz offers the selection of two internally designed filters, processed via a DSP, and the option to turn on/off noise shaping. The latter had no impact on measurements, and my guess is that it acts at a too low digital level, probably 24bits, to be revealed.
The NA-11S1 uses BurrBrown DSD1792 DACs that have a max THD+N of 0.0004% at 48kHz, that means a max SINAD of 108dB. We'll check if the Marantz can reach that.
There is a very welcome headphones output on the front.
User Experience
That's an easy one, streamers became a standard (well, not for me). I mostly used it as a DAC for the measurements and listening. Nothing special to mention, except again that the RC is mandatory, and I don't like that.
My computer with Windows 11 recognized the USB Audio output without asking for a driver (up to 352.8kHz max). I tried to install the Marantz driver, but it did not work.
Marantz NA-11S1 - Measurements
All measurements performed with an E1DA Cosmos ADCiso (grade 0), and the Cosmos Scaler (100kohms from unbalanced input) for analog outputs, and a Motu UltraLite Mk5 for digital.Like I said, the
I measured 4.618Vrms from the Right balanced channel, and 4.628Vrm from the left one (with Filter 2), that is a low 0.02dB channel imbalance. The Filter 1 increased the output by 0.09dB which could be heard by some, as it plays louder.
I measured 2.303Vrms from right unbalanced channel, and 2.302Vrms from the left one (Filter 2), meaning a near perfect channel balance.
The balanced outputs inverted signal polarity, meaning that the XLR jacks had their pin 3 wired as hot (not pin 2, as is usual), as shown on the back of the Marantz.
As often, RCA are slightly less performing than XLR, but they especially suffer some low level PS leakage, that are absent from XLR. That being said, for the rest of this review, all measurements you'll see were performed from balanced outputs.
And allow me to deviate a little from my usual measurements, but I wanted to replicate some performed by John at the time, to see how this Marantz ages but also to check the performances are consistant between two totally unrelated items, an ocean and 14 years appart.
----
Let's start with the standard dashboard, at 4Vrms output (-1dBFS), so to comply with the reviews of @Amir with Filter 1:
Well, that is better than BurrBrown specs for that DAC, with a SINAD of almost 111dB. The max SINAD, at full output power increases slightly to 111.5dB. And that puts the Marantz, as a DAC, in the "Excellent" category of Amir's reviews!
It is good to see that Marantz has nailed the implementation of the BurrBrown DAC in this player.
Now, to really comply with Amir's review, I need to extend the span of analysis from 10Hz to 22.4Hz, in which case:
The SINAD reduces to 110.5dB, here right channel, and 110.2dB from the left one. This still makes the Marantz enter the "Excellent" range of Amir's review.
----
As I used to measure at -6dBFS, with all my other reviews (same span as above), let's try:
Good to see that the SINAD did not lose the full 6dB difference from -1dBFS (4Vrms).
----
This is the multitone test that you like:
I tested with both filters, and it's the same result. We get more than 20bits of distortion free-range, which is excellent (forget that spike at 150Hz, I get it sometimes from my testing setup, and I don't know why).
----
As @amirm does, let me share the THD+N vs Frequency view, with noise included up to 90kHz, and alternatively 45kHz:
I guess you understand why I overlaid the two. Indeed, the result with 90kHz Bandwidth implies high frequency constant noise (-90dB). At 45kHz Bandwidth, we get below that noise, and so we see a better -105dB value, which is nice. Note that the two filters made no difference, the noise is probably coming from the delta-sigma modulator of the DSD1792, so unavoidable.
But let's have a closer look at all of that.
----
And this is where I'll redo the measurements of John, and that you can read here.
First, let's have a look at the behavior of the two Marantz filters, with a wider bandwidth (300kHz) than the one of John (100kHz). This is Filter 1:
It is an overlay of white noise and the 19.1kHz test tone that John used. The filter one is slow and so the alias of the 19.1kHz test tone is still very high at 24kHz. I can zoom on the frequency (80khz):
Note also the delta-sigma modulator's noise of the DAC (noise floor increasing after 35kHz).
And as you can see from the above, the filter's roll-off is significant, and reaches -3.5dB at 20kHz (with 44.1kHz bandwidth).
Let's do the same with Filter 2:
And zoom:
This is a sharper filter but not very efficient at attenuating the aliases of the conversion (barely -60dB). Same noise floor beyond 35kHz, which makes me say this comes from the high speed modulator of the DAC.
The above results are the same that John reported at the time, of course.
----
Let me add a view of the 3DC at 16bits that John liked to perform, and that I'm reusing when measuring a CD Player:
This is at 16bits and shows that the Marantz is very silent. The three primary levels of the conversion are very well represented. Only the latest well implemented DACs achieve better.
Let's run the same test with 24bits:
We get a clean nice sine, as expected.
----
And let's continue with the Jitter Test:
I again did it "à la" John, with a restricted BW (7500Hz zo 14500Hz). Both coax and optical showed the same little amount of jitter. And similar to what John reported, I got a better result from the USB input:
In case you're interested, I explained in the review of the SMSL PO100 PRO what this J-Test is made of.
----
Let me also add the CCIF intermodulation test (19kHz+20kHz), first with Filter 1:
Very weak attenuation of their aliases (at 24.1kHz and 25.1kHz), but you see that the calculated result is a very good -110dB distorsion (-101dB with noise).
Let's do the same with Filter 2:
Results are basically the same, but the aliases are better attenuated, generating also less noise in the audio band, and I prefer.
----
Other few measurements:
We can see all odd harmonics created by the interpolator clipping.
Marantz NA-11S1 - Digital in to out
Since the DAC of this Marantz is excellent but not SOTA per today's standards, I wanted to check how it behaves as a digital pass-though. In the past, I've seen some devices modifying the digital stream, so I wanted to check if the Marantz truly keeps it intact.
When I review CD Players, I use the intersample overs test at 5512.50Hz, with a phase shift of 67.5°, like I did for the TASCAM CD-200 review. This signal generates an overshoot of +0.69dB and so if the digital signal would be modified before being sent (by an ASRC for instance), it would show either a reduction of amplitude or we'd see some sort of saturation/increase noise/distorsion. So here we go, below, and note this is 16bits PCM:
We get the +0.69dBFS overshot and no noise increase or added distorsion. PCM 16bits is therefore kept intact.
Let's try at PCM 24bits, this time with a more standard 997Hz test @0dBFS, and without dither:
The above is an overlay of a capture at the output of the Marantz and one directly from the SMSL PO100 Pro output. The traces are identical, difference being when I captured them. Indeed, the 997Hz test tone generates a lot of unique PCM subcodes, so an averaged 4x FFT is unlikely to be exactly twice the same. No noise and no distortion here.
With these two tests. I'm confident the Marantz has a bypass for digital data, to feed directly the SPDIF outputs, and it is good to know.
Conclusion
The Marantz NA-11S1 was certainly best in class in its time. John concluded "superb measured performances", and I can only concur.
Besides a little low level noise which prevented it from achieving a higher dynamic range, it reached the max performances of the BurrBrown DAC.
With its SINAD of more than 110dB at 4Vrms, the NA-11S1 accesses the "Excellent" category of Amir's DAC reviews! Well done big boy!
In case of doubts, you can also use it with an external current SOTA DAC, to push results even further.
It is still, 14 years later, a very good performer and so, happy owners can continue to enjoy the sound of it.
Happy weekend!
This is a review an measurements of the Marantz NA-11S1 Network Audio Player.
It was send to me by a kind member for review.
Marantz NA-11S1 - Presentation
The Marantz NA-11S1 was released in 2012. It is a massive unit in size and weight, which showcases exceptional craftsmanship. Its cost was roughly $3500, so we are indeed talking high-end.
@John Atkinson reviewed it for Stereophile and you'll find some insights from M. Ishiwata talking about the end of the physical media (poor me) and the beginning of a new streaming era.
Considering the recent announcement regarding Stereophile that @amirm shared, and fact that John announced he's going for (a well deserve) retirement, I thought I'd use this review as an opportunity to make a tribute to his past 40 years of measuring Hi-Fi devices for us. I wanted to thank him for that.
————
The back of the Marantz is busy:
No Bluetooth/wifi, but network and SPDIF inputs (I tested up to 24bits/192kHz) plus USB for connection to a computer (up to 352.8kHz). Analog outputs include RCA and XLR (hot pin 3, ie output is inverted). There are also digital SPDIF outputs.
Sadly, despite joggle button on front face, it is impossible to configure some audio options without the remote control. This is a big miss.
The Marantz offers the selection of two internally designed filters, processed via a DSP, and the option to turn on/off noise shaping. The latter had no impact on measurements, and my guess is that it acts at a too low digital level, probably 24bits, to be revealed.
The NA-11S1 uses BurrBrown DSD1792 DACs that have a max THD+N of 0.0004% at 48kHz, that means a max SINAD of 108dB. We'll check if the Marantz can reach that.
There is a very welcome headphones output on the front.
User Experience
That's an easy one, streamers became a standard (well, not for me). I mostly used it as a DAC for the measurements and listening. Nothing special to mention, except again that the RC is mandatory, and I don't like that.
My computer with Windows 11 recognized the USB Audio output without asking for a driver (up to 352.8kHz max). I tried to install the Marantz driver, but it did not work.
Marantz NA-11S1 - Measurements
All measurements performed with an E1DA Cosmos ADCiso (grade 0), and the Cosmos Scaler (100kohms from unbalanced input) for analog outputs, and a Motu UltraLite Mk5 for digital.Like I said, the
I measured 4.618Vrms from the Right balanced channel, and 4.628Vrm from the left one (with Filter 2), that is a low 0.02dB channel imbalance. The Filter 1 increased the output by 0.09dB which could be heard by some, as it plays louder.
I measured 2.303Vrms from right unbalanced channel, and 2.302Vrms from the left one (Filter 2), meaning a near perfect channel balance.
The balanced outputs inverted signal polarity, meaning that the XLR jacks had their pin 3 wired as hot (not pin 2, as is usual), as shown on the back of the Marantz.
As often, RCA are slightly less performing than XLR, but they especially suffer some low level PS leakage, that are absent from XLR. That being said, for the rest of this review, all measurements you'll see were performed from balanced outputs.
And allow me to deviate a little from my usual measurements, but I wanted to replicate some performed by John at the time, to see how this Marantz ages but also to check the performances are consistant between two totally unrelated items, an ocean and 14 years appart.
----
Let's start with the standard dashboard, at 4Vrms output (-1dBFS), so to comply with the reviews of @Amir with Filter 1:
Well, that is better than BurrBrown specs for that DAC, with a SINAD of almost 111dB. The max SINAD, at full output power increases slightly to 111.5dB. And that puts the Marantz, as a DAC, in the "Excellent" category of Amir's reviews!
It is good to see that Marantz has nailed the implementation of the BurrBrown DAC in this player.
Now, to really comply with Amir's review, I need to extend the span of analysis from 10Hz to 22.4Hz, in which case:
The SINAD reduces to 110.5dB, here right channel, and 110.2dB from the left one. This still makes the Marantz enter the "Excellent" range of Amir's review.
----
As I used to measure at -6dBFS, with all my other reviews (same span as above), let's try:
Good to see that the SINAD did not lose the full 6dB difference from -1dBFS (4Vrms).
----
This is the multitone test that you like:
I tested with both filters, and it's the same result. We get more than 20bits of distortion free-range, which is excellent (forget that spike at 150Hz, I get it sometimes from my testing setup, and I don't know why).
----
As @amirm does, let me share the THD+N vs Frequency view, with noise included up to 90kHz, and alternatively 45kHz:
I guess you understand why I overlaid the two. Indeed, the result with 90kHz Bandwidth implies high frequency constant noise (-90dB). At 45kHz Bandwidth, we get below that noise, and so we see a better -105dB value, which is nice. Note that the two filters made no difference, the noise is probably coming from the delta-sigma modulator of the DSD1792, so unavoidable.
But let's have a closer look at all of that.
----
And this is where I'll redo the measurements of John, and that you can read here.
First, let's have a look at the behavior of the two Marantz filters, with a wider bandwidth (300kHz) than the one of John (100kHz). This is Filter 1:
It is an overlay of white noise and the 19.1kHz test tone that John used. The filter one is slow and so the alias of the 19.1kHz test tone is still very high at 24kHz. I can zoom on the frequency (80khz):
Note also the delta-sigma modulator's noise of the DAC (noise floor increasing after 35kHz).
And as you can see from the above, the filter's roll-off is significant, and reaches -3.5dB at 20kHz (with 44.1kHz bandwidth).
Let's do the same with Filter 2:
And zoom:
This is a sharper filter but not very efficient at attenuating the aliases of the conversion (barely -60dB). Same noise floor beyond 35kHz, which makes me say this comes from the high speed modulator of the DAC.
The above results are the same that John reported at the time, of course.
----
Let me add a view of the 3DC at 16bits that John liked to perform, and that I'm reusing when measuring a CD Player:
This is at 16bits and shows that the Marantz is very silent. The three primary levels of the conversion are very well represented. Only the latest well implemented DACs achieve better.
Let's run the same test with 24bits:
We get a clean nice sine, as expected.
----
And let's continue with the Jitter Test:
I again did it "à la" John, with a restricted BW (7500Hz zo 14500Hz). Both coax and optical showed the same little amount of jitter. And similar to what John reported, I got a better result from the USB input:
In case you're interested, I explained in the review of the SMSL PO100 PRO what this J-Test is made of.
----
Let me also add the CCIF intermodulation test (19kHz+20kHz), first with Filter 1:
Very weak attenuation of their aliases (at 24.1kHz and 25.1kHz), but you see that the calculated result is a very good -110dB distorsion (-101dB with noise).
Let's do the same with Filter 2:
Results are basically the same, but the aliases are better attenuated, generating also less noise in the audio band, and I prefer.
----
Other few measurements:
- Crosstalk is -120dB at 1kHz, rising to -100dB at 20kHz, so that's good.
- Linearity is good down to -115dBFS (+0.2db) and deviates by only +0.7dB at -120dBFS which is very good.
- The Dynamic range (CCIR-2K) is 112dB (116dB A weighted) and that is limited by the noise floor of the Marantz. On that the BurrBrown DSD1792 can theoretically do better.
- When tested from a CD Player, with my usual "max resolution DAC" test file, the Marantz achieved 18.5bits, very close to the max theoretical 18.7bits (see SMSL PL 200 review to know more about the test).
- Pitch error (GPS DO corrected) is a small -8ppm (19'996.84Hz for 19'997.00Hz requested)
- Both Filters showed no resistance to intersample overs with an equal THD+N of -30dB (90kHz BW) from a test tone of 5512.50Hz @0dBFS with 67.5° phase shift, which generates a +0.69dB over "only", see below:
We can see all odd harmonics created by the interpolator clipping.
Marantz NA-11S1 - Digital in to out
Since the DAC of this Marantz is excellent but not SOTA per today's standards, I wanted to check how it behaves as a digital pass-though. In the past, I've seen some devices modifying the digital stream, so I wanted to check if the Marantz truly keeps it intact.
When I review CD Players, I use the intersample overs test at 5512.50Hz, with a phase shift of 67.5°, like I did for the TASCAM CD-200 review. This signal generates an overshoot of +0.69dB and so if the digital signal would be modified before being sent (by an ASRC for instance), it would show either a reduction of amplitude or we'd see some sort of saturation/increase noise/distorsion. So here we go, below, and note this is 16bits PCM:
We get the +0.69dBFS overshot and no noise increase or added distorsion. PCM 16bits is therefore kept intact.
Let's try at PCM 24bits, this time with a more standard 997Hz test @0dBFS, and without dither:
The above is an overlay of a capture at the output of the Marantz and one directly from the SMSL PO100 Pro output. The traces are identical, difference being when I captured them. Indeed, the 997Hz test tone generates a lot of unique PCM subcodes, so an averaged 4x FFT is unlikely to be exactly twice the same. No noise and no distortion here.
With these two tests. I'm confident the Marantz has a bypass for digital data, to feed directly the SPDIF outputs, and it is good to know.
Conclusion
The Marantz NA-11S1 was certainly best in class in its time. John concluded "superb measured performances", and I can only concur.
Besides a little low level noise which prevented it from achieving a higher dynamic range, it reached the max performances of the BurrBrown DAC.
With its SINAD of more than 110dB at 4Vrms, the NA-11S1 accesses the "Excellent" category of Amir's DAC reviews! Well done big boy!
In case of doubts, you can also use it with an external current SOTA DAC, to push results even further.
It is still, 14 years later, a very good performer and so, happy owners can continue to enjoy the sound of it.
Happy weekend!
Last edited:
