NAD M23 VS NCore500 A/B test

[Garlic]

I wonder as well...

As more amp manufacturers use the same modules in their products, and all have similar scores in scientific SINAD tests, how does one make a choice ?
Like comparing NAD 298 vs Boxem 4215, same amps right ? 500€ price difference but NAD has XLR and RCA, and trigger input. Is one better built, more reliable than the other, better heat dissipation ? Am i missing something ?

Interesting times we live in, we have a myriad of smaller "independent" manufacturers (Boxem, Vera, Nord, Audiophonics, etc...) building the same products as larger mainstream brands for (in the case of the C298) almost the same price...

At some point it may come down to just a difference in box style, size and color.

 

[MaxwellsEq]

@PaulHC I don't think anybody can answer your questions. This is a site attempting to bring scientific test methods to the technology and environment that allows us to enjoy music as faithfully reproduced as possible in the home, office or studio. A scientist has to posit an idea, devise and run tests to DISPROVE that idea. If they cannot disprove it, they open it up to expert peers to also devise and run tests to disprove it and if they cannot disprove it, then it becomes canon until someone comes up with a better idea or a test that disproves it.

Is it possible that a device with an over-specified power supply under test measure differently than an identical unit with a less well specified power supply? Of course it's possible! In fact, this has been proven over decades:

• Poor mains ripple and noise rejection is easily measured and is shown on the graphs that get published here by those with E1DA setups and the more expensive AP equipment. These will impact sound, especially likely to mask quieter aspects of the sound
• Insufficient current swing or voltage swing from the power supply limits the maximum output and may be OK at middle frequencies, but struggle at frequency extremes
• Poor regulation leads to "saggy" power supplies unable to sustain amplifier delivery when called on to work hard

There have been generally accepted, engineering ways to measure all of these aspects for decades. The tools are getting more resolute and able to display much more informative graphs (e.g. multitone and FM testing).

But whether you can hear a difference between these two amplifiers is a different matter. The gold standard for whether something is audibly different is the accurately level-matched to 0.1dB (using a scope or high-frequency AC voltmeter) double-blind listening test. What this shows up time and again is that much of what gets described subjectively as audibly different cannot be reliably detected by listeners.

To answer your question - you need to have a double-blind, accurately level-matched shootout between the two devices, perhaps at 75dB, 85dB, 95dB and (for very short periods) 105dB. Since I don't think this has been done, none of us can answer your questions.

 

[PaulHC]

Many thanks for all your helpful comments above and without proper testing i agree it is hard to comment concretely !

I think things are so even now in audio we r left with the burden on the salesman of hi end gear and the nuances of audio placebo effects and perhaps it is just good quality components and longevity of use that matters.

I used my nCore recently vs a very very expensive system (circa 100k ukp) and it failed dismally with audible distortion of the vocals - turned out to be a failing capacitor which has been replaced - it has crossed my mind that as class Ds are switched on for 10 years these failures may be inaudible at first then appear as distortion later which is why my impression (dont shoot me guys) was the m23 pleased my individual auditory cortex more ........

Not trying to deny the science but are class Ds complex enough in its filtering that a gradual reduction in performance may happen with time ? Ive heard a class d is effectively a switching power supply with output in the audio range and power supplies do fail ......... do they degrade ???

 

[PaulHC]

Just getting my kevlar vest on ......

 

[MaxwellsEq]

I used my nCore recently vs a very very expensive system (circa 100k ukp) and it failed dismally with audible distortion of the vocals - turned out to be a failing capacitor which has been replaced - it has crossed my mind that as class Ds are switched on for 10 years these failures may be inaudible at first then appear as distortion later which is why my impression (dont shoot me guys) was the m23 pleased my individual auditory cortex more ........

Not trying to deny the science but are class Ds complex enough in its filtering that a gradual reduction in performance may happen with time ? Ive heard a class d is effectively a switching power supply with output in the audio range and power supplies do fail ......... do they degrade ???

That is possible - many devices degrade with age and/or use: bearings and motors, tubes/valves, electrolytic capacitors can as well. But this decay is common across designs. It will impact a Class A/B as well as a Class D. Given 10,000 years I suspect every component will fail (e.g. due to electron migration, quantum effects). Class D amps are more "complex" than relatively simple Class A or Class B amps, so they may not last as long or degrade quicker. Although Sony built amps of this type several decades ago, until we've had the current crop around for 30 odd years, it won't be possible to know (MilSpec destructive testing may give results). In their favour - they don't normally run very hot so heat-based failures (e.g. to electrolytics) may be less.

BUT such degradation is measurable compared to the original measurements. PSUs demonstrate more ripple or pass through more noise. THD and IMD increase. Separation reduces. Noise increases. Inter-channel imbalances emerge. These can ALL be measured.

I once commissioned 11 24-channel analogue sound mixing desks. I ran a gamut of tests across each channel in each desk. Almost all of them measured identically, but one had an unexpected roll-off above 5kHz and two had noise 10dB worse than the others. I listened to all three "broken" channels and was not sure if I could hear a difference, but I swapped them anyway.

 

[PaulHC]

Many thanks again for your reply

 

[kemmler3D]

how does one make a choice ?

Lots of the things you mentioned - performance is a given with the modules in question, so what else matters to you? Build quality, heat dissipation, warranty, customer service quality in your country, the I/O you want, amenity features beyond that, sheer looks - all things that may be worth paying for.

Not trying to deny the science but are class Ds complex enough in its filtering that a gradual reduction in performance may happen with time ? Ive heard a class d is effectively a switching power supply with output in the audio range and power supplies do fail ......... do they degrade ???

Yes, definitely, ask @restorer-john , he's always ready to beat the drum about how Class D is less reliable and tends to be hard to repair if it fails. What people will often tell you, especially on ASR, is that nobody has ever noticed (read: correctly identified in a blind test) an audible difference between two well-built amps that were not driven into distortion. What is implied but not spoken about much in that context is the amp needs to be working properly! And as you said, sometimes they break. Class D has a lot going for it, but reliability isn't necessarily it's strongest point as a group. And naturally the sound can degrade in all sorts of interesting and sometimes subtle ways if the amp itself is broken.

 

[pogo]

NAD build the purifi module under license so i presume it isnt modified but will check with NAD
I wonder if it was a better power supply in the NAD that at high volumes i perceived a difference - or am i, after over 45 years of hifi finally floating and hallucinating in a tub of snake oil ????

No, you are not
This has all been discussed here before: Link
NAD M23/M33 seem to use a different firmware of the amplifier modules, otherwise no output power >800W at 2 ohms would be possible.
Also, the Purifi modules built in license seem to be more tightly tolerated, so that they can also be bridged.

 

[Tim Link]

It's interesting when you have a perception of a difference during sighted listening and it wasn't expected and you can't shake it, especially when it's a startling difference on some music. I think it's safe to guess that something is going on but then you have to do a real controlled test if you really want to verify it. I understand now why some subjectivists aren't interested in doing it. If you can't shake the perception during sighted listening, and you like the perception you are getting, why test it unless you're trying ot save money or are just curious? I'm pretty sure I've percieved distinct differences between three different receivers I've hooked up to my system recently. I've also percieved differences with a change in dacs and also the source feeding those dacs. I subscribe to Apple Music, and I play it from my Mac Mini and from an AppleTV. I hear glitches and dullness in sound from the Mac Mini. It sounds better playing the same tracks from the Apple TV, even though it's signal is being processed through the Mac Mini. I haven't blind tested or done any comparitive measurements yet but I'm going to try soon because I'm curious. I wouldn't bother if I didn't think I was hearing something. So that's the starting point.
BTW, someobody did a comparison of that M23 to some other solid state amp as a Youtube video. The differences I percieved just listening to that video of those amps being switched out matches pretty close what you described as hearing. The M23 sounded tight, clean and neutral. The other amp, which most people claimed to prefer, was a little "juicy" sounding. I could see why people might like it better. It's not good science but it shouldn't be ignored either when you think you hear a difference and you can describe that difference, and there seems to be some consensus about that description. It makes me want to understand it better - get that perception verified and then correlated with some measurements - hopefully.

 

[boXem]

I wonder as well...

As more amp manufacturers use the same modules in their products, and all have similar scores in scientific SINAD tests, how does one make a choice ?
Like comparing NAD 298 vs Boxem 4215, same amps right ? 500€ price difference but NAD has XLR and RCA, and trigger input. Is one better built, more reliable than the other, better heat dissipation ? Am i missing something ?

Interesting times we live in, we have a myriad of smaller "independent" manufacturers (Boxem, Vera, Nord, Audiophonics, etc...) building the same products as larger mainstream brands for (in the case of the C298) almost the same price...

At some point it may come down to just a difference in box style, size and color.

A few thoughts,
- 500 € is a 33% increase.
- Manufacturing location can influence people decision, or not.
- If you look at the build quality, the "big" competitor is more the NAD M23
- and the most important, these modules need a proper integration. We already have 2 two reviews on ASR of amplifiers using 1ET400A showing sub-par performance. So just looking at the module is like just looking at a DAC chip: no guarantee of quality.

No, you are not
This has all been discussed here before: Link
NAD M23/M33 seem to use a different firmware of the amplifier modules, otherwise no output power >800W at 2 ohms would be possible.
Also, the Purifi modules built in license seem to be more tightly tolerated, so that they can also be bridged.

Could you please link the review in which the M23/M33 were measured at 800W? I can't find it anymore.
The Vera and the boXem can also be bridged. It is not a question of manufacturing tolerances, it is that a supervisor needs to be put in place in order to shutdown both modules when one experiences some failure mode.

 

[Papillon]

Indeed
The real interesting test would be to compare the boxem 4215 and the nad m23…

 

[pogo]

Could you please link the review in which the M23/M33 were measured at 800W?

M23 = 730W / M33 = 815W: Link

 

[pogo]

The Vera and the boXem can also be bridged. It is not a question of manufacturing tolerances, it is that a supervisor needs to be put in place in order to shutdown both modules when one experiences some failure mode.

NAD Response about Bridging the M23 Amplifier

With regards to the trick of lowering the distortion by re-wiring of the feedback signal for bridge mode; we’re well aware of this possibility but decided against this. We decided to optimise the M23 for normal stereo use. With distortion levels as low as they are, just adding switching for re-routing the feedback signal will affect distortion in normal operation mode.

We channel-match the Left and Right modules to be as close as possible; this helps reduce distortion in bridge mode also. Indeed, if bridge operation would be permanent and thus not switchable, it’s logical to apply re-routing the feedback loop.

As performance in bridge mode still is as good as some of the best amps out there, we decided this was an acceptable compromise.

Source: audioholics.com

 

[boXem]

M23 = 730W / M33 = 815W: Link

Thank you.
That is dynamic power, measured in a certain way. It cannot be compared to the continuous power that we quote and that @Amir measures.

NAD Response about Bridging the M23 Amplifier

With regards to the trick of lowering the distortion by re-wiring of the feedback signal for bridge mode; we’re well aware of this possibility but decided against this. We decided to optimise the M23 for normal stereo use. With distortion levels as low as they are, just adding switching for re-routing the feedback signal will affect distortion in normal operation mode.

We channel-match the Left and Right modules to be as close as possible; this helps reduce distortion in bridge mode also. Indeed, if bridge operation would be permanent and thus not switchable, it’s logical to apply re-routing the feedback loop.

As performance in bridge mode still is as good as some of the best amps out there, we decided this was an acceptable compromise.

Source: audioholics.com

Good to learn that NAD is matching their modules. But this is not the reason why they can bridge them. No supervisor, no bridging.

 

[pogo]

That is dynamic power, measured in a certain way. It cannot be compared to the continuous power that we quote and that @Amir measures.

Too bad that this was not measured by AMIRM. Continuous power should be around 600W @2ohm.