Hypex Ncore

Status
Not open for further replies.
I agree somewhat Barrows - I haven't heard an amp that can do electronica like the Ncores and in fact on some vocals they are great as well - but there are times that I prefer my other amps which have a slight lushness that the Ncore misses. I've discovered that the Ncores are the better amps with my vinyl source (PS Audio GCPH / Ortofon 2M Black) but that I prefer my A/B amp with digital.
Yes the Ncore is transparent and breathtakingly clear - specs don't always tell the whole story though.
 
Pass X150.5 in my case. I would not extend my listening experience to a general point of class A and class A/B amps though. Only what I have actually compared.
And: to respond to your other comment: no, what I was describing was definitely not coloration, this is specifically why I mentioned it. Coloration changes tonality of sound which already exists. What was lacking in my (admittedly too brief to come to a full conclusion, why I mentioned I am going to listen more) listening was the upper harmonic decays of bells, chimes, cymbals, and other very high frequency sounds with tonal decays. With nCore amps they just seemed quite truncated. This observation does not indicate a coloration in the Pass (although that exists as well, as with any amp) but something omitted from the output of the nCores. I have heard enough actual bells, chimes, and cymbals in real life to know.
Perhaps longer break in, and/or a slightly different build could ameliorate this lack in the nCores. I am going to do another build soon and intend to do a lot of listening and fine tuning in order to find out.

It does not help to find the real perfromance of the nCores, if everytime someone has a listening observation which could be considered negative, then someon chimes in with: "you want coloration", or "you like the sound of distortion". As if the nCore amps were "perfect". Sorry, but nothing is perfect. I find the nCores very, very impressive and promising, and am very interested in a longer term audition of them. Hence, I will be doing a high quality stero build with them. If you must have measurements which indicate that the nCores are not "perfect", I would point you to their bandwidth, though I do not know that the limited bandwidth is a problem or not. What I do know, is that many, very well respected amplifier designers, with quite a bit more experience than Mr. Putzeys, do consider extended bandwidth important for amplifiers.
Do you mind posting some measurements of your speakers in the listening position?
And what other electronics are you using?
 
Not Relevant...

"Do you mind posting some measurements of your speakers in the listening position?
And what other electronics are you using?'

1. Frequency response measurements would not be relevant to the specific observation I made. Additionally, the room, and its response, is the same in both of my observations: the difference(s) heard are only the differnces between the two amplifiers under test. The observations of the differences between amps which I reported were not of tonal differences, I have tried to point this out previously, and just cannot say it any more clearly than that.

2. A custom server, running Voyage/mpd, with a audiophile quality USB output card (SOtM), powered by a custom dual power supply, with a separate linear supply for the USB output card-a highly optimized Buffalo DAC build, with Salas regulators, and an onboard asynchronous USB receiver (SOtM, XMOS based), synchronously clocked with very low phase noise clocks, dedicated clock power supply (separate transformer) with two stages of regulation, each oscillator module being fed from its own shunt reg-DAC is direct connected to the amplifier via Nordost Frey cables-Nordost Frey speaker cables to Focus Audio FS-888 Signature loudspeakers.
 
I find it strange that someone makes obejective statements about an amp based on experiences primarily in their own setup. There are so many variables that come into place that it's almost impossible to know for sure what the amp's signature is and what is something else. One is hearing a complete system where speakers and acoustics contributes the most.

It's better in my opinion to share subjective experiences subjectively and not try to make objective conclusions.
 
Curious?

I find it strange that someone makes obejective statements about an amp based on experiences primarily in their own setup. There are so many variables that come into place that it's almost impossible to know for sure what the amp's signature is and what is something else. One is hearing a complete system where speakers and acoustics contributes the most.

It's better in my opinion to share subjective experiences subjectively and not try to make objective conclusions.

my subjective observations are exactly that and nothing more. Neither did I present them as anything more than my observations. Everyone else is free, and encouraged, to make their own observations. I presented them in a transparent way, and did not draw any overeaching conclusions, in fact, I expressed my desire to spend more time with an nCore amplifier to learn more about how it sounds.
But: I did make direct comparisons between the nCore and my Pass X150.5, in the exact same system. Within this context, which of course is what matters as it is what I listen to, the comparison is totally valid.

Interestingly, I have heard from quite a few people, both in this thread, and privately, who have experienced similar observations of the sound of the nCores in comparison to other well regarded class A/B amps-I am not drawing conclusions for others from these reports though, you are free to listen and come to your own conclusion.
 
What was lacking in my (admittedly too brief to come to a full conclusion, why I mentioned I am going to listen more) listening was the upper harmonic decays of bells, chimes, cymbals, and other very high frequency sounds with tonal decays. With nCore amps they just seemed quite truncated. This observation does not indicate a coloration in the Pass (although that exists as well, as with any amp) but something omitted from the output of the nCores. I have heard enough actual bells, chimes, and cymbals in real life to know.
distortion". As if the nCore amps were "perfect". Sorry, but nothing is perfect. I find the nCores very, very impressive and promising, and am very interested in a longer term audition of them. Hence, I will be doing a high quality stero build with them. If you must have measurements which indicate that the nCores are not "perfect", I would point you to their bandwidth, though I do not know that the limited bandwidth is a problem or not. What I do know, is that many, very well respected amplifier designers, with quite a bit more experience than Mr. Putzeys, do consider extended bandwidth important for amplifiers.

As a user of Both UcD and NCore.
On looking at building Hypex some years ago I had my reservations about the bandwidth. My understanding (thanks to Fourier when at college)was that the odd harmonics were essential to the faithfull reproduction of a square wave. While we do not have theoretical square waves in our audio signal one can argue that we have a close similarity with rising wavefronts.
Looking at a square wave at 15 Khz and we want to faithfully reproduce the 3rd 5th and 7th harmonic we are looking at a bandwidth in excess of 100Khz with the correct phase response.
Perhaps a close inspection of the waveforms generated by the various instruments in the orchestra would give a much more accurate picture of the harmonic content that an amplifier must be able to handle and I would be very surprised if Bruno has not already been down that route.
I agree some designers do consider extended bandwith up to 150 Khz essential but there is a lot more to amplifier design than bandwidth looked at in isolation.
My own experience, which is naturally constrained by the bandwith of my ears, is that the Ncore is the best all round amp Ive owned.
Perhaps you are much younger than me and blessed with young ears?
 
Barrows, thanks for sharing your observations. Especially for reporting them in such a reflective manner, which sadly is too seldom seen.

Your personal observations are of course particular to your taste, ears, and system, though nevertheless valid for that particular event. When you support your perspective with that of other highly acclaimed amplifier constructors, you do nevertheless provide your claim with some kind of "general" -quality.

But for that maneuver to be valid these amplifier constructors has to have pointed out exactly the point you make here about truncated upper harmonics. Otherwise you "just" agree on the ncore not being entirely up to the best you have heard. -Which on the other hand isn´t so bad for a plug n´play DIY amp kit sold for much less.

Personally I can relate to your observation on the sound of the ncore to have some kind of quality to it. That is that the ncore does do something to the signal besides adding gain.
I haven´t had enough time with ncore to add anything to your statement other than I have noticed this in various degrees on all class D amps. That is not to say that class D is inferior to other classes but it´s a quality that seems to stick with them.

As a reason for this "sound" of class D hereunder the ncore´s that you speak of, the bandwidth limitation might have some validity when looking how it affects the structure of upper harmonics. I found this read interesting on that very subject:

The Sound of the Machine

On the other hand I have a lot of experience with a class A/B construction also with a fair share of nfb inherent to its operation but with a bandwith of > 600kHz and some 3,5 kHz for small signals.

That amp has some of the truncatedness too, but it seems to depend on it´s bias setting. With this amp I can either have truncation of highs or blooming lows. I prefer the best class D to that, as I find this amp unbearable to listen to for either reason...

I am not knowledgable enough to make any statement on what the reasons are for various flavors that these different amps introduce, but my guess would be that the operation of the nfb circuitry is having as much to say on these amps as their ultimate bandwidth.
How does the nfb operation condition the upper harmonics?

If I understand the ncore right, it is the nfb operation that is its major attribute compared to the ucd.
My humble guess is that the algorithm integrated to the nfb on the ncore is both responsible for its major leap forward in SQ to UCD recreating a much bigger and finely rendered image, but also responsible for its "sound".
To what extend can the ncore operation hereunder its algorithm simulate and control upper harmonics?
Could this be a more to the point answer that includes role of the control loop frequency extension?

best,
 
The new 'ncd' NFB loop seems to have some cunning tricks to increase the open-loop gain, together with real-time fast-acting circuitry to modify the O/L transfer function under the conditions that could normally cause the amplifier self-oscillation to go unstable as a result of the higher O/L gain. (Such as driving the amplifier to clipping with an open-circuit load, which can make the normal ucd circuit switch to a self-osc frequency around 40kHz, as a result of the sharp phase shift transition of the output filter into a High-Z load). I have not used an example of an NCD amp, but worries about upper harmonics seem to be misplaced as the published intermodulation distortion of a a 19kHz/20kHz signal pair look very impressive for a class-D amp.
 
IIRC Bruno said he spent some considerable time comparing (listening to) the amplifier input signal to the output signal

It would be an interesting exercise for everyone to also try that with their amplifier of choice. Hopefully that should help understand which amplifiers add to or take away from the original signal and in what way
 
19kHz/20kHz signal pair look very impressive for a class-D amp.

Exactly. As you say, for a class D.

Another question is how this speaks for the composition of higher order harmonics.

A different factor that could have some technical validity in terms of explaining the top end performance of ncore is its high D/F // Low O/P impedance. This represents a major leap forward in terms amplifier-speaker control and may endure such different damping properties to the top end than what "we" are used to experience from electronics. If this has some credentials then the characteristics of the control loop (again) responsible for achieving this high DF will probably show even more compared to previous loops offering lower DF.
 
I would be interested in hearing where/how people get source material that has any contents at frequencies above 100 kHz...

The fundamental frequencies of source material is oft described as 20-20Khz.
There will always be a extension to this bandwidth due to the presence of harmonics, these in turn being dependant on the precise shape of the fundamental frequency.
A 10 Khz square wave will consist of a fundamental sine wave at 10Khz and sinusiodal harmonics at 20, 30 40 50 60 70 Khz + etc. These harmonics will decrease in amplitude as they rise in frequency.
In simple terms, this describes the constituent part of a 10Khz square wave and to reconstitute it faithfully the fundamental and all the harmonics must be reproduced in the correct amplitude and phase.
 
A 10 Khz square wave will consist of a fundamental sine wave at 10Khz and sinusiodal harmonics at 20, 30 40 50 60 70 Khz + etc. These harmonics will decrease in amplitude as they rise in frequency.
In simple terms, this describes the constituent part of a 10Khz square wave and to reconstitute it faithfully the fundamental and all the harmonics must be reproduced in the correct amplitude and phase.

I am familiar with the Fourier series. My point is that there is no need to be able to reproduce a 10 kHz square wave, as your source material (music) won't contain any 10 kHz square waves - the material is already, by the time it gets to the amp, low-pass filtered.

If your source is a CD, there is absolutely no original material over 22 kHz. If your source is vinyl, you might get up to 30 kHz the first couple of times you play the record. Even "hi-res" digital recordings won't contain material above 100 kHz...

"If it's not there, it is not there" is true for harmonics, too...
 
I am familiar with the Fourier series. My point is that there is no need to be able to reproduce a 10 kHz square wave, as your source material (music) won't contain any 10 kHz square waves - the material is already, by the time it gets to the amp, low-pass filtered.

If your source is a CD, there is absolutely no original material over 22 kHz. If your source is vinyl, you might get up to 30 kHz the first couple of times you play the record. Even "hi-res" digital recordings won't contain material above 100 kHz...

"If it's not there, it is not there" is true for harmonics, too...

The very reason for the 18+19kHz intermodulation plot is that higher order harmonics are understood to modulate the audible signal. One thing is whether or not something is there, another thing is whether or not it´s modulational artifacts can be reproduced faithfully...
 
Without wishing to cause a riot, I have to observe that talk of very high frequency (Ultrasonic) components of a sound somehow influencing a waveform after filtering is simply not thought through. All audio we hear (digital, analogue or live) is low-pass filtered, during the recording and mastering process or simply by the human ear.

That process not does selectively leave high frequency artifacts alone to modulate the carrier. There are no higher order harmonics in cd sound, above the 22k limit. None.
 
Last edited:
There are no higher order harmonics in cd sound, above the 22k limit. None.

Exactly.

One thing is whether or not something is there, another thing is whether or not it´s modulational artifacts can be reproduced faithfully...

Well, if something is not there, it's modulational artifacts (sic) probably aren't there either, so whether they can be reproduced faithfully or not is a bit of a moot point...
 
My understanding (thanks to Fourier when at college)was that the odd harmonics were essential to the faithfull reproduction of a square wave. While we do not have theoretical square waves in our audio signal one can argue that we have a close similarity with rising wavefronts.

Sound is created by physical objects' motion.

It has been correctly observed that no signals contain the leading edge of a perfect square wave.

Even if they did, it would not matter.

Sound id produced by objects moving air, and no object can move with infinite acceleration/velocity, so there are no acoustic square waves to be reproduced.
 
Last edited:
Status
Not open for further replies.