• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Impeccable measurements. Are they the key to success in today's world of high end audio?

Mivera

Major Contributor
Joined
Mar 4, 2016
Messages
2,322
Likes
58
Location
West Kelowna
#21
I have not. But I also have the Mark Levinson 532 reference amplifier (Class AB) and the 53 outperforms it at every level.

It is just pure prejudice against the notion of "digital amplification." Nothing about the sound.
I'm not sure what it is, but the same types who hate the Hypex stock sound, change their mind when the right input buffer is used. So I'm not so sure if it's just placebo effect of knowing the amp is class D.
 

Opus111

Addicted to Fun and Learning
Joined
Mar 2, 2016
Messages
666
Likes
18
Location
Zhejiang
#22
So there must be something else going on that's beyond the scope of industry standard measurements causing this.
Precisely - what's causing different subjective impressions from Hypex buffers would be how noisy the buffer's power supplies get when driving into the low impedance input of the Hypex module itself. A pure classA buffer where the power supply draw is invariant with signal would be my suggestion to achieve more transparency.
 

Mivera

Major Contributor
Joined
Mar 4, 2016
Messages
2,322
Likes
58
Location
West Kelowna
#23
I can only suggest that you check the input impedance of the classD offerings. The raw classD circuits from Hypex I recall have fairly low (3kohm is it?) input impedance and this does present an issue for the electronics driving them.

Sounds to me like jumping to a conclusion that 'none of the flaws are being masked'.
That's what the buffer boards are for. The NC-500's are just an output stage. The input impedance is a very low 1.8 kohm. The input buffers are designed as proper input buffers for the amps. There's nothing wrong with them.

Well that's the way it is with these amps. They are so transparent, the sound passes through them with very little coloration. Most people are accustomed to coloration's.
 

Mivera

Major Contributor
Joined
Mar 4, 2016
Messages
2,322
Likes
58
Location
West Kelowna
#24
Precisely - what's causing different subjective impressions from Hypex buffers would be how noisy the buffer's power supplies get when driving into the low impedance input of the Hypex module itself. A pure classA buffer where the power supply draw is invariant with signal would be my suggestion to achieve more transparency.
That's not an issue with these buffers. The power supply can easily drive 4x times the load with ease. The discrete opamps aren't even flinching driving the low input impedance of the hypex amps. They are far more robust than the LM 4562's that are in the standard Hypex OEM buffer. They also run in class A way beyond the current requirements of the Hypex amp.
 

Opus111

Addicted to Fun and Learning
Joined
Mar 2, 2016
Messages
666
Likes
18
Location
Zhejiang
#25
OK, 1.8kohm and they require a high signal level to achieve the low noise do they not? Like something of the order of 20VRMS, balanced? (trying to remember the Hypex DS). This would translate to a peak input current of 8 or 9mA per phase.
 

Mivera

Major Contributor
Joined
Mar 4, 2016
Messages
2,322
Likes
58
Location
West Kelowna
#26
OK, 1.8kohm and they require a high signal level to achieve the low noise do they not? Like something of the order of 20VRMS, balanced? (trying to remember the Hypex DS). This would translate to a peak input current of 8 or 9mA per phase.
The peak current requirements to achieve full output from the amps is only 11mA. These discrete opamps can easily sink/source 60mA and run in pure class A up to 16mA.
 

Opus111

Addicted to Fun and Learning
Joined
Mar 2, 2016
Messages
666
Likes
18
Location
Zhejiang
#27
That's not an issue with these buffers. The power supply can easily drive 4x times the load with ease.
Seems you missed my point. Its not about the power supply's ability to drive the load, its about how much noise is generated on the power supply while driving the load. That is something you could measure.

Running in classA is no guarantee of power supply current invariance. Its necessary, but not sufficient.
 

Mivera

Major Contributor
Joined
Mar 4, 2016
Messages
2,322
Likes
58
Location
West Kelowna
#28
Seems you missed my point. Its not about the power supply's ability to drive the load, its about how much noise is generated on the power supply while driving the load. That is something you could measure.

Running in classA is no guarantee of power supply current invariance. Its necessary, but not sufficient.
I'm using the exact same power supply and voltage regulators on my input board and they are dead silent. Like the amps are off. Anyways what I'm getting at is 100% of people prefer the noisy boards, over the dead silent Hypex standard boards that measure impeccably. I bet the noisy board that everyone prefers would look much worse on paper.
 

Mivera

Major Contributor
Joined
Mar 4, 2016
Messages
2,322
Likes
58
Location
West Kelowna
#29
Seems you missed my point. Its not about the power supply's ability to drive the load, its about how much noise is generated on the power supply while driving the load. That is something you could measure.

Running in classA is no guarantee of power supply current invariance. Its necessary, but not sufficient.
Yeah but I'm not trying to pinpoint the source of the noise. I'm more interested in measuring what the noise looks like on paper. It's not my board that I'm trying to troubleshoot. I already have a board that has a -140Db noise floor I'm quite happy with.
 

Mivera

Major Contributor
Joined
Mar 4, 2016
Messages
2,322
Likes
58
Location
West Kelowna
#31
If Amir is up for it one day, it would be interesting to have a subjective listening session with a panel of audiophiles comparing the 2 input boards. After the votes are in, measure the amps with both input boards and see if the board that got the most votes as being superior is the best measuring board.
 

March Audio

Major Contributor
Manufacturer
Patreon Donor
Joined
Mar 1, 2016
Messages
4,084
Likes
4,069
Location
Perth Western Australia
#32
A lot of people call them harsh sounding, fatiguing and not musical. Same folks who say the same about the Mola Mola Kaluga's. They call it "The class D sound" But I think what it really is, is the sound of the source that they don't like because none of the flaws are being masked. Because when a different input buffer is used, that "class D sound" seems to disappear.
My experience of NC400s is that they just sound clean and neutral. I call that "class D lack of sound" I can understand why some dont like this, they are looking for euphony. If you partner this with the "wrong" speakers or room it could turn you off.

Wasnt the whole point of the 500s that manufacturers could tune to their own house sound with their own input buffer designs?
 
Last edited:

Mivera

Major Contributor
Joined
Mar 4, 2016
Messages
2,322
Likes
58
Location
West Kelowna
#33
My experience of NC400s is that they just sound clean and neutral. I can understand why some dont like this, they are looking for euphony. If you partner this with the "wrong" speakers or room it could turn you off.
Yes I owned the NC 400's for 4 years. Bruno designed the input buffer for absolute transparency. They must be paired with very good source gear or else they can sound cold and dry. However I have been able to achieve IMO much better sound out of the NC500's, yet still achieve measured results superior to the NC400's. So I don't think his buffer is the holy grail.

Anyways, back to the NC-500. The LM4562 based OEM buffer for the NC-500's is closer to the discrete buffer that is used on the NC400's than anything subjectively and measurably. However It's not as good as the NC400 buffer. Yet the noisier Nord buffer is subjectively preferred by all so far.
 
Last edited:

Opus111

Addicted to Fun and Learning
Joined
Mar 2, 2016
Messages
666
Likes
18
Location
Zhejiang
#34
Bruno designed the input buffer for absolute transparency.
As far as I can tell, 'absolute transparency' to Bruno means lowest measurable distortion. I got that from reading about his Mola Mola DAC where he went the extra mile driving the THD+N down below any other competitors' efforts. So no surprise then if it didn't sound transparent.

<afterthought> Perhaps the Nord buffer gets its sonic advantages by virtue of stochastic resonance?
 

Mivera

Major Contributor
Joined
Mar 4, 2016
Messages
2,322
Likes
58
Location
West Kelowna
#35
As far as I can tell, 'absolute transparency' to Bruno means lowest measurable distortion. I got that from reading about his Mola Mola DAC anyway where he went the extra mile driving the THD+N down below any other competitors' efforts. So no surprise then if it didn't sound transparent.
Him and the Hypex team are real measurement buffs that's for sure. They were very happy with the feedback Nord was getting on the NC-500 based input buffer with swappable opamp socket. Until as promised I shared my feedback on them on the Diyaudio forum saying they they were noisier than the Hypex OEM buffer. Shortly after the manufacturer of the Nord amps got his hand slapped by Hypex and warned not to discuss his amps on the Ncore thread. They also ran out of power supply and amp modules for almost 3 months. Then they also told him he has to raise his pricing because they are so low that they are cutting into the NC400 sales, as well as generating complaints from the competition. So I feel a bit bad, but the board wouldn't exist if it wasn't for me, and I recommended it to many already, so I felt the obligation to share my results. Although the noise isn't much. I'm estimating an S/N ratio of around -100dB. Even the Mark Levinson No 53 has a much higher noise floor of -85dB, so in reality it's not a big deal. However it is still noise that isn't there with both the Hypex board and my board that inspired it. If someone had a pair of 100dB+ efficient speakers it would be annoying.
 
Last edited:

Opus111

Addicted to Fun and Learning
Joined
Mar 2, 2016
Messages
666
Likes
18
Location
Zhejiang
#36
Sounds like bully-boy tactics - why not just raise the price of the OEM NC500? The cutting off of his supply isn't going to endear Hypex to anyone who learns about that is it? Nord seems to be the customer's best friend in that he's doing it on very slim margins.
 

Mivera

Major Contributor
Joined
Mar 4, 2016
Messages
2,322
Likes
58
Location
West Kelowna
#37
Sounds like bully-boy tactics - why not just raise the price of the OEM NC500? The cutting off of his supply isn't going to endear Hypex to anyone who learns about that is it? Nord seems to be the customer's best friend in that he's doing it on very slim margins.
Well they didn't exactly say they are cutting him off. But the supply ran out at the same time as all of this happened. So now he's stuck unable to fill his orders. He had to let his employee go and everything. He's just a small operation building the amps in his garage. So his overhead is very low. And he's not greedy either. He does a very nice job as well.
 

Opus111

Addicted to Fun and Learning
Joined
Mar 2, 2016
Messages
666
Likes
18
Location
Zhejiang
#38
He's saying he's now got to wait until May for the next shipment from what I recall. I hope he doesn't get discouraged as I agree he's doing a nice job there.
 

Mivera

Major Contributor
Joined
Mar 4, 2016
Messages
2,322
Likes
58
Location
West Kelowna
#39
He's saying he's now got to wait until May for the next shipment from what I recall. I hope he doesn't get discouraged as I agree he's doing a nice job there.
Yeah I'm not sure if everyone has to wait or not. I haven't tried ordering any after this went down. The timing was quite coincidental though. This is one drawback of using 3rd party parts. You are at their mercy.
 
Top Bottom