NAD M33 Streaming Amplifier Review

[Reed]

Which speakers do you use?

Magnepan MMG, Schiit Vidar amp, REL T5i connected through SUB1 (top) on C658. I use the full range Dirac upgrade.

 

[pogo]

REW using umik-1 and Monitor Audio Silver 6 speakers. I can also share spectral delay plots that include the time domain, but again, the plots between the two firmware versions are the same. Happy to conduct more tests if you'd help me understand how to quantify and test "cloudy", "swing-in" and "swing-out". I think the key here is, if you're happier with the new firmware, then that's great!

3D Step Response Measurement - The sound quality of the reproduction made visible!?

The 3D Step Response measurement shows the behaviours of the sound reproduction in the non swung-in state. This measurement is very important to judge the reproduction of impulses. As testsignale preferably half sinus oscillations of different frequencies at the same time are used, since acoustic events are essentially based on these modes of motion.

View attachment 83902



The frequency response is totally inadequate for the judgement of sound quality. The frequency response is measured with non-changing (Static) signals such as sinus waves or calculated noise signals. Against that, a music signal consists of changing (Dynamic) signals. The study of music signals shows that the sound quality of a musical instrument is determined by impulses. Especially the first impulse, the stroke of a guitar string, the impact of a piano cord, the hit on a drum and the blow of an organ pipe or brass instrument is important for the sound. In the Step-Response, all acoustic parameter information of the sound reproduction is contained. This consists of the frequency response, phase response and swing-in behaviour. There for the display is so complex, that the sound quality judgement based on the measurement is not possible. That is why the 3D Step-Response measurement was developed. By the measurement, the Step-Response is analysed and displayed in a 3D graph. The 3D measurement shows the impulse reproduction for each single frequency with use of an additional frequency axis. The measurement makes it for the first time possible to analyse the sound quality through measurement technology.

(Source: Kirchner elektronik)

As a reminder, see also posts #331 and #287.
Conclusion: You need a different test setup to be meaningful.

 

[Reed]

Let me say that the placebo isn't in effect in my situation. I installed the BluOS update and there were NO differences to the sound. I have DiracLive on my Mac. I did my session in January then resaved the same session so I could do another session with my KEFs at a later date. I have three curves that have not changed. I opened Dirac to see if there was an update. None was present. I switched from one session to the other to make sure that session was working. My expectation is that nothing would happen since they are the same session, only renamed. That's not what happened. First I noticed a volume increase. I thought it Dirac was disabled. Nope. I go through my curves (Harman +4 preferred) and they are all working with the differences I've come to expect. However, everything has changed. The volume has increased. I know where I listen via the display on the NAD and I checked the db with my phone app. It's louder. And yes, I know that volume mismatches can make differences to perceived tonality so I adjusted the volume to match my known listening db using the phone app. There is an easily heard change in the tone. More highs for sure. But it's everything. It's easier to identify sounds and instruments across the soundstage. And yes, this is music I listen to all the time. And it's the same if I go back to the other identical session. I'm going to start from scratch with a new session in a few days. So no, this isn't placebo.

 

[Thalassophile]

Following the link that a member posted with the Lana Del Ray "For Free" song, at 1:10, there is a phrase, "Waiting for the walking greeeeeeeeeen". During that phrase, if I turn the subwoofer filter on, I do hear the distortion. With the subwoofer filter off however, there is no distortion. Furthermore, if I turn on the tone controls, which reduce the gain of the amplifier (to give some headroom), the distortion goes away.

To summarize:
Dirac On, Subwoofer Filter Off, Tone Controls Off: No distortion
Dirac On, Subwoofer Filter On, Tone Controls Off: Distortion
Dirac On, Subwoofer Filter On, Tone Controls On: No distortion

Therefore, for people who listen to music that is recorded at clipping levels and need to use a subwoofer, you can try to turn the tone controls on (leave them at 0db) and test if the distortion also disappears for you.

Well I'll be damned!!! I was just able to duplicate fcracer's exact results. I could hear the distortion when the Subwoofer Filter was on and Tone Controls were off. It's the word "walk" right before "in greeeeeen" that is clipped. Could hear it in the FLAC file and also playing a lossy version off of Spotify, so the bit depth was actually irrelevant.

If you turn the Tone Controls from "Off to On" during playback, it's obvious that the volume does drop a few decibels, even when the treble and bass are at "0." That will tick some purists off. But I am satisfied with the solution and feel a sense of some relief.

 

[Kaboo]

Let me be clear, all NAD processors (C658, T778, M33) have the distortion issue when any processing such as Dirac or bass management is activated. This is a hardware limitation due to lack of processing headroom. Tone control and using replaygain on can limit the distortion but will also limit the output level and scarify dynamic range. NAD cannot and WILL NOT fix this. This is the final replay I got after escalating the issue to their management:

-----------------------------
Hi Eli

Thanks for the sample. We have been investigating this matter for some time now. The inter-sample clipping (Rhye distortion) you are hearing is created at the time of recording on some "hot" tracks and cannot be resolved in any other way than reducing the gain on the track.

Please make adjustments accordingly. Sorry.

--------------------------

There is no FW fix! This is what happens when digital processing is carried out using the original bit depth of the recording, you can easily run out of headroom. I've had it with NAD. I've ordered the MiniDSP SHD, which will be connected to dual Hypex NC400.

See other similar discussions about NAD distortion here:

https://www.audiosciencereview.com/...st-this-for-me-female-vocal-distortion.22151/

I’m a bit lost here. If the distortion is due to the recording clipping (“too hot” in the words of the NAD tech), how can you expect firmware to fix a recording issue?

Firmware will not fix this issue .We are screwed

 

[Kaboo]

Following the link that a member posted with the Lana Del Ray "For Free" song, at 1:10, there is a phrase, "Waiting for the walking greeeeeeeeeen". During that phrase, if I turn the subwoofer filter on, I do hear the distortion. With the subwoofer filter off however, there is no distortion. Furthermore, if I turn on the tone controls, which reduce the gain of the amplifier (to give some headroom), the distortion goes away.

To summarize:
Dirac On, Subwoofer Filter Off, Tone Controls Off: No distortion
Dirac On, Subwoofer Filter On, Tone Controls Off: Distortion
Dirac On, Subwoofer Filter On, Tone Controls On: No distortion

Therefore, for people who listen to music that is recorded at clipping levels and need to use a subwoofer, you can try to turn the tone controls on (leave them at 0db) and test if the distortion also disappear

I’m a bit lost here. If the distortion is due to the recording clipping (“too hot” in the words of the NAD tech), how can you expect firmware to fix a recording issue?

It can't be fixed it is not a fault of the recording .it is bad engineering.we have been screwed and if you own one so have you.

 

[Kaboo]

Well I'll be damned!!! I was just able to duplicate fcracer's exact results. I could hear the distortion when the Subwoofer Filter was on and Tone Controls were off. It's the word "walk" right before "in greeeeeen" that is clipped. Could hear it in the FLAC file and also playing a lossy version off of Spotify, so the bit depth was actually irrelevant.

If you turn the Tone Controls from "Off to On" during playback, it's obvious that the volume does drop a few decibels, even when the treble and bass are at "0." That will tick some purists off. But I am satisfied with the solution and feel a sense of some relief.

It will limit the distortion but will also limit the output level by about 6 db and effect dynamic range. This so called Masters amp is now hobbled no matter how you look at it .If you can justify their incompetence to make yourself feel better that's good for you .In reality what have you spent your hard earned money on ?

 

[Kaboo]

It will limit the distortion but will also limit the output level by about 6 db and effect dynamic range. This so called Masters amp is now hobbled no matter how you look at it .If you can justify their incompetence to make yourself feel better that's good for you .In reality what have you spent your hard earned money on ?

The percentage of music this cluster fuck effects is large. Keep listening

 

[Thalassophile]

It will limit the distortion but will also limit the output level by about 6 db and effect dynamic range. This so called Masters amp is now hobbled no matter how you look at it .If you can justify their incompetence to make yourself feel better that's good for you .In reality what have you spent your hard earned money on ?

You are right Kaboo. The more I listen, especially with headphones (with Dirac off, obviously) I realize it was not just a reduction in decibels with Tone Controls on, it flattens the soundstage a bit. I mean, at least I think so, it's hard to tell if it is just the -6db affecting my perception or not. It's not dramatic, but the loss of dynamics seems to be there. Damn.

That's the thing, I keep the tone controls off but do not use a subwoofer, so I did not notice the distortion before or the effect Tone Controls have on the sound stage. Only after I read fcracer's post I figured out how to replicate it the problem you were all having.

 

[Kaboo]

Well I'll be damned!!! I was just able to duplicate fcracer's exact results. I could hear the distortion when the Subwoofer Filter was on and Tone Controls were off. It's the word "walk" right before "in greeeeeen" that is clipped. Could hear it in the FLAC file and also playing a lossy version off of Spotify, so the bit depth was actually irrelevant.

If you turn the Tone Controls from "Off to On" during playback, it's obvious that the volume does drop a few decibels, even when the treble and bass are at "0." That will tick some purists off. But I am satisfied with the solution and feel a sense of some relief.

limit the distortion but will also limit the output level and scarify dynami

You are right Kaboo. The more I listen, especially with headphones (with Dirac off, obviously) I realize it was not just a reduction in decibels with Tone Controls on, it flattens the soundstage a bit. I mean, at least I think so, it's hard to tell if it is just the -6db affecting my perception or not. It's not dramatic, but the loss of dynamics seems to be there. Damn.

That's the thing, I keep the tone controls off but do not use a subwoofer, so I did not notice the distortion before or the effect Tone Controls have on the sound stage. Only after I read fcracer's post I figured out how to replicate it the problem you were all having.

Yes we are all part of this design flaw .There must be something that can be done to compensate us for this farce we are all a part of now?

 

[EB1000]

Best sol is to turn subwoofers off, have then connected using the front speakers pre outs via minidsp 2x4 HD. After first REW calibration, run Dirac live. The results are actually better and no distortion. The only problem is driving your small speakers with full range bandwidth. Usually not a problem. Either way, as I said, I'm done with NAD for good. I'll give minidsp SHD a try...

 

[Reed]

Before everyone sets their hair on fire, try to replicate this. I just applied today’s BluOS update for my C658. Sounds like before or flat, afraid to say. I then opened the DiracLive software on my Mac laptop and loaded the project while music is playing. Only do this with your volume low because it immediately gets loud(er). Confirmed with the volume display on the C658. Maybe jumped by 10db. Quit DiracLive and everything remains. And to me it changes a number of things. IOW, it ain’t flat, at least not on my system. Confirmed that all curves are working. I think there might be an issue with the handshake between DiracLive and BluOS. And opening DiracLive while playing music through BluOS somehow clears the cobwebs. Someone try this please.

 

[Deleted member 65]

Looking at this from outside it looks to me as an Intersample peaks/overs problem. There’s an excellent article written by Archimago here: https://archimago.blogspot.com/2018/09/musings-measurements-look-at-dacs.html

If any of you users are using Roon you can enable Headroom Management with -3/-6 dB and verify if the Intersample peaks clipping occur.

 

[pogo]

Let me say that the placebo isn't in effect in my situation.
First I noticed a volume increase.
It's easier to identify sounds and instruments across the soundstage.

I fully agree with you. After the update with activated DL, the volume probably increased by 3dB for me. My distortion issue (see post #494) now occurs at a volume of -21dB (previously -18dB) and now also has a different signature. NAD has definitely improved the DL issue 'cloudy sound'. This issue, which I and other users posted last year, was probably at the top of the priority list and before the issue 'distortion'. I can understand that, because this effect was clearly audible with very good speakers and there is a short-term remedy for the issue 'distortion' with tone control on.

 

[fcracer]

Looking at this from outside it looks to me as an Intersample peaks/overs problem. There’s an excellent article written by Archimago here: https://archimago.blogspot.com/2018/09/musings-measurements-look-at-dacs.html

If any of you users are using Roon you can enable Headroom Management with -3/-6 dB and verify if the Intersample peaks clipping occur.

Lars, thanks for posting, this appears to be a problem with many DACs. It seems that the only fix from a hardware level is if the designer attenuates the DAC by at least 3db so that there’s headroom above 0dbfs. As you noted, there are some software fixes available such as headroom management. For those that want to read up on this, there are several posts on ASR about this topic as well by searching for “intersample”.

 

[pogo]

That was my guess before and that can be done via this GUI for the M33:

Here an example of the DAC GUI taken from the datasheet of the DAC ES9028:
View attachment 119667

 

[fcracer]

It can't be fixed it is not a fault of the recording .it is bad engineering.we have been screwed and if you own one so have you.

We have plenty of dynamic range to spare. Turning down the digital signal by 3dB is not going to make much difference to your audio pleasure. This intersample peak issue impacts many brands. I wonder if NAD could resolve this by providing a menu option to attenuate the source by a specific dB?

If you’re that unhappy with your device, talk to your retailer to get a refund, trade-in, or re-sell the amplifier. If your retailer doesn’t take care of you, contact senior people at NAD and ask them for a resolution. As you noted, you’ve spent your hard earned money on it and are unhappy. A good company will find a resolution.

 

[Mnyb]

Lars, thanks for posting, this appears to be a problem with many DACs. It seems that the only fix from a hardware level is if the designer attenuates the DAC by at least 3db so that there’s headroom above 0dbfs. As you noted, there are some software fixes available such as headroom management. For those that want to read up on this, there are several posts on ASR about this topic as well by searching for “intersample”.

yes it does not matter how you turn your back the solution is attenuation to handle inters ample peaks that will arise with any processing sub xover tone control EQ dirac oversampling and whatnot .

If NAD where to implement a solutions there solution would obviously be digital attenuation somewhere, not because they are lazy but because it's the only solution you cant go above 0 i digital audio . This is the cost for processing in any such product not only NAD .

As little as -0.2dB can do the trick -6dB migth be excessive try values between -dB and -3dB with your software player .

If the M33 is audible nosier because of this , NAD should thought of that when doing the DAC design . There is a reason for very high SINAD you suddenly need this headrooom when doing any processing in any product.

Benchmark has -3dB in their later products and still produce respectable performance , so it can be done .

Wonder how high the analog gain after the DAC section in M33 is this also affects noise performance ? if done sub optimally it also affects noise performance ideally you should be using almost the whole of the volume control ? btw is volume also digital ?

 

[EB1000]

yes it does not matter how you turn your back the solution is attenuation to handle inters ample peaks that will arise with any processing sub xover tone control EQ dirac oversampling and whatnot .

If NAD where to implement a solutions there solution would obviously be digital attenuation somewhere, not because they are lazy but because it's the only solution you cant go above 0 i digital audio . This is the cost for processing in any such product not only NAD .

As little as -0.2dB can do the trick -6dB migth be excessive try values between -dB and -3dB with your software player .

If the M33 is audible nosier because of this , NAD should thought of that when doing the DAC design . There is a reason for very high SINAD you suddenly need this headrooom when doing any processing in any product.

Benchmark has -3dB in their later products and still produce respectable performance , so it can be done .

Wonder how high the analog gain after the DAC section in M33 is this also affects noise performance ? if done sub optimally it also affects noise performance ideally you should be using almost the whole of the volume control ? btw is volume also digital ?

How come this issue doesn't affect Yamaha or Denon AVR that do lots of processing? I can assume that the digital data is first converted from 16 or 24 bits into a 32bit format, which adds plenty of headroom w/o scarifying dynamics. Roon converts to 64bits for processing.

 

[Deleted member 65]

How come this issue doesn't affect Yamaha or Denon AVR that do lots of processing? I can assume that the digital data is first converted from 16 or 24 bits into a 32bit format, which adds plenty of headroom w/o scarifying dynamics. Roon converts to 64bits for processing.

It’s all about the DAC design/engineering choices, RME offers +3.0dB headroom whereas Benchmark settled for +3.5dB. Note it’s all arbitrary chosen, no absolute targets exists since worst cases can reach +10dBFS.