NAD M33 Streaming Amplifier Review

[Soniclife]

@amirm are you able to measure the phono stage? Given the discussion of it's value above is partly about how many things it does well knowing how good the phono stage is makes a difference.

 

[amirm]

No, I already mailed it back to its owner.

 

[Helicopter]

I would send in my NAD PP2, but seems like that would be a waste of time since it is out of production. I bet the phono stage is as good as the Rega Amir just reviewed.

 

[da Choge]

We could look to that for performance differences. If the modules are the same, then the issue is elsewhere.

Yes, perhaps, and maybe it is the NAD-designed buffer boards (and might they be possibly out-sourced for production)? You guys are the experts, but isn't this likely where the M33 could take a slightly unwelcomed dive from the Purifi amps as far as measurements?

P.S. BTW, I DO very much like that the M33 seems to have primo AD conversion unlike so many other components (high-end or otherwise) that we see today.

 

[amirm]

Yes, perhaps, and maybe it is the NAD-designed buffer boards (and might they be possibly out-sourced for production)? You guys are the experts, but isn't this likely where the M33 could take a slightly unwelcomed dive from the Purifi amps as far as measurements?

Yes. It is my guess that the buffer is at fault.

 

[pogo]

Hello…

1)
Have incorrect measurement data been used in the NAD M33 Streaming Amplifier Review?:

2)
I think the pictures in the NAD M33 Streaming Amplifier Review belong to the UUT!? You should still mention the expansion stage (MDC modules inserted, e.g. HDMI and/or the upcoming DSD/USB respectively none of them), the firmware version and the serial number of the UUT in the review. The first two points in particular can influence the measurement results.

3)
I have had my NAD M33 for about 3 weeks and it will definitely replace my NAD C388.

I‘m amazed which textures can be heard on voices / instruments. I didn't know these before and therefore probably didn't miss them. The lightness and impulsiveness also surprises me again and again when browsing through my music collection.

The headphone output also surprised me very positively (very detailed and dynamic playback). Hopefully a DIRAC correction for headphones will also come onto the market in the near future.

Phono MM / MC sound really good too.

And of course you can negotiate a discount from a dealer you trust. I don't know whether this is still possible. The amplifier seems to have been sold out for some time.

BR
Michael

 

[PeteL]

Yes. It is my guess that the buffer is at fault.

My guess is that the buffer is fine. First, we actually have at the DAC level, at 48kHZ, performance already a bit below Purifi specs when looking at high frequency THD+N , and it looks like In your tests the analog inputs digitized at 44.1, so we are already bottlenecked before even reaching the buffer. second, There is a LOT more stuff in the box, it is totally expected to have a raised noise floor, and in this case actually not that bad. They can't make miracles, there are just more power lines travelling arround the chassis. Third, the channel difference, it is obviously the same circuitry for both channels, so it has to come from physical layout and compromise based on real estate. The fact that both are not the same confirm my second point.
I believe that I've read somewhere in this thread that the sample rate of the ADC on the analog inputs could be increased? Just that could have make an appreciated improvment, like it did at DAC level no?

 

[pogo]

I miss the information in the review of the NAD M33 sample rate setting of the ADC used during the analog measurements!?
48kHz is the default value. You can set up to 192kHz!

 

[Frank Dernie]

What is disturbing is that the modules don't have the usual blue pcb seen on Nad stuff, in opposition to what is licensed from Hypex.
The first time I saw the green pcb, my reaction was that they were not ready at launch and bought a few samples to Purifi for start of production. A tear down would be interesting.

I could easily imagine the early units had bought in modules whilst production of in house modules was de-bugged.
Production engineering items often go through under-the-skin mods particularly early in the production run.
I can see how people expect an item they buy to be the same as all the others with the same model number but it often isn't so.

 

[AndrewDavis]

not to say this in a negative way for this engineering-dominated forum) marketing and product management innovation than technical innovation.

As a marketer I take no offence.

In my experience companies that have that a strong partnership between sales/marketing and engineering/r&d are more likely to create quality products. Too strong an emphasis on R&D creates products that are interesting to engineers but not useful to customers.

A needlessly large emphasis on sales & marketing creates very shiny turds.

 

[PeteL]

I could easily imagine the early units had bought in modules whilst production of in house modules was de-bugged.
Production engineering items often go through under-the-skin mods particularly early in the production run.
I can see how people expect an item they buy to be the same as all the others with the same model number but it often isn't so.

All this is true, but in this particular case all evidences point the other way. I am not sure why people want so much to believe NAD are modyfying the modules. It's the most accomplished class D tech around. They couldn't improve on this. Now we've got a tear down of the m28 from NAD, that appears to be using the stock Purifi module, we got pictures of he m33 wher the modules are obviously green and look just the same.... NAD can twist that marketing wise how they want, but when it looks like a duck, swim like a duck and quacks like a duck... It’s probably a duck...

 

[pogo]

I looked at the measurement curves again and came to the conclusion that the measurements involving the analog input section were carried out with the lowest quality level possible at the NAD M33 (ADC sample rate 48kHz). For that they seem to be good to very good.

There are no measurement data of the full potential of the NAD M33, if the analog input stage is used at 192kHz ADC sample rate. Please measure again, if possible!

To make matters worse, the sample rate settings only work correctly since the following FW-version and have previously reset themselves to the default value of 48kHz:

BluOS 3.12.9. Released November 4, 2020:

• Update for inputs and settings on NAD M33. This release applies to the NAD M33 only.

 

[boXem]

All this is true, but in this particular case all evidences point the other way. I am not sure why people want so much to believe NAD are modyfying the modules. It's the most accomplished class D tech around. They couldn't improve on this. Now we've got a tear down of the m28 from NAD, that appears to be using the stock Purifi module, we got pictures of he m33 wher the modules are obviously green and look just the same.... NAD can twist that marketing wise how they want, but when it looks like a duck, swim like a duck and quacks like a duck... It’s probably a duck...

Where have you seen people believing that NAD would modify the modules?
NAD tell that the modules are built by themselves under Purifi license. We wonder why they don't have the NAD blue PCB (which is not a modification). One hypothesis is that manufacturing is not ready. Another one is that they have decided to use green PCB for whatever reason.
What would be the point for NAD to tell that they manufacture the modules while they don't intent to do so?

 

[PeteL]

I looked at the measurement curves again and came to the conclusion that the measurements involving the analog input section were carried out with the lowest quality level possible at the NAD M33 (ADC sample rate 48kHz). For that they seem to be good to very good.

There are no measurement data of the full potential of the NAD M33, if the analog input stage is used at 192kHz ADC sample rate. Please measure again, if possible!

To make matters worse, the sample rate settings only work correctly since the following FW-version and have previously reset themselves to the default value of 48kHz:

BluOS 3.12.9. Released November 4, 2020:

• Update for inputs and settings on NAD M33. This release applies to the NAD M33 only.

@amirm already said somewhere that the integrated amp have already been sent back to the owner. With all respect to our host, it is part of the dynamic of this forum. We get granted with new gear measured almost daily, anybody can sent out any type of gear to be measured, but the compromise is that they are not always fully torough, you won't find all the options/inputs measured.

 

[PeteL]

Where have you seen people believing that NAD would modify the modules?
NAD tell that the modules are built by themselves under Purifi license. We wonder why they don't have the NAD blue PCB (which is not a modification). One hypothesis is that manufacturing is not ready. Another one is that they have decided to use green PCB for whatever reason.
What would be the point for NAD to tell that they manufacture the modules while they don't intent to do so?

They said the "amplifiers" are by built by NAD. They didn't say the modules, but why would it matter then if they are the same? "Manufacturing" is a vague term... I am about 99% sure that NAD don't "manufacture" their own PCBs, that would not do any sense... literally EVERYBODY outsource unpopulated boards, and even the pick and place/assembly of smd centric boards, that would be very, very very, very... well you get the idea... very rare. Even Purifi themselves, I'd be very surprised if they have the facility to populate their board, and EXTREMELY surprised if they made their own PCBs. So bottom line, you have somewhere a PCB manufacturer that push tens of tousands, if not hundreds of thousands of these green PCBs, and instead of just grabbing them, NAD would insist on going trough the tooling process with an other manufacturer, for maybe one thousand pcbs, just for the sake of having em Blue? Come on...

My view on this, and that was my point, NAD is an amp manufacturer, and they are very consious of their wording to not mess with the IDEA in their customers mind, that they are buying an NAD amp, not a rebranded Purifi amp, that's all there is to it, and it's not inacurate. The module is not an amp, the final product is designed and manufactured by them, and their team has to stick to this mantra. At the end of the day, a part of their design is made by someone else, and there is nothing wrong with that, who on earth would refuse to buy a product just because one of the pcb is green instead of blue? What if nad would comme to a second revision and say in a very brute honest way. Well, we tought blue is nicer, so we'll increase the price to distribute the fixed cost it involved to our customers, just because... Well... NAD is Blue...

 

[ironhorse128]

I feel like the M33 is a terrific device and quite competitive if you get it for 4k. I run a miniDSP shd studio, two RME Adi Dacs (one for the mains and one for the subs) and two hypex NP250 modules to do basically the same. All that together is also very expensive.

I also feel that the a miniDSP shd studio with the RME adi dacs provide much more freedom and features. You get four completely independent channels with independent trim level, independent time delays, independent parametric EQ. All these features are very useful to tune your setup. Dirac might not always solve all issues automatically.

Thus, the M33 might not be as power user friendly compared to the miniDSP shd. But it sure gives you good features in a great performing unit. The question is: Do you want to fiddle with all these details and options?

 

[boXem]

They said the "amplifiers" are by built by NAD. They didn't say the modules, but why would it matter then if they are the same? "Manufacturing" is a vague term... I am about 99% sure that NAD don't "manufacture" their own PCBs, that would not do any sense... literally EVERYBODY outsource unpopulated boards, and even the pick and place/assembly of smd centric boards, that would be very, very very, very... well you get the idea... very rare. Even Purifi themselves, I'd be very surprised if they have the facility to populate their board, and EXTREMELY surprised if they made their own PCBs. So bottom line, you have somewhere a PCB manufacturer that push tens of tousands, if not hundreds of thousands of these green PCBs, and instead of just grabbing them, NAD would insist on going trough the tooling process with an other manufacturer, for maybe one thousand pcbs, just for the sake of having em Blue? Come on...

My view on this, and that was my point, NAD is an amp manufacturer, and they are very consious of their wording to not mess with the IDEA in their customers mind, that they are buying an NAD amp, not a rebranded Purifi amp, that's all there is to it, and it's not inacurate. The module is not an amp, the final product is designed and manufactured by them, and their team has to stick to this mantra. At the end of the day, a part of their design is made by someone else, and there is nothing wrong with that, who on earth would refuse to buy a product just because one of the pcb is green instead of blue? What if nad would comme to a second revision and say in a very brute honest way. Well, we tought blue is nicer, so we'll increase the price to distribute the fixed cost it involved to our customers, just because... Well... NAD is Blue...

The Eigentakt modules are manufactured by NAD under license from Purifi, allowing NAD to optimise these specifically in combination with the custom designed power supply and input stages of the C 298.

NAD Intros C 298 Power Amplifier With Purifi Eigentakt Amplification

 

[PeteL]

NAD Intros C 298 Power Amplifier With Purifi Eigentakt Amplification

Fair enough, but inside pictures of the c298 shows green PCB's too. You realise that this sentence could literally mean changing the value of one resistor or part right? Nobody wants to reinvent the wheel. Manufacturing the modules, OK, I believe that, but it by no way means that they can't buy the unpopulated pcb board from them, or even Populated ones if the modifications are not on the SMD parts it's still a manufacturing process. Plus, it's a lower cost product, so "optimize" can mean in this case, pull back on actual available power.

 

[boXem]

https://yourbias.is/the-dunning-kruger-effect

 

[PeteL]

https://yourbias.is/the-dunning-kruger-effect

This relate to this discussion how? The fact that we don't know what "manufactured" and "optimize" mean for NAD where people with less expertise on the subject would think they do?