Hegel H95 Review (Streaming Amplifier)

[watchnerd]

He he.

Edit:
Question: Did not Duglas Self, said that he would rather take a class B amplifier with decent power than a class A with low power? To make it absolutely clear. That was a question. So no one comes and claims that I said Douglas Self said that. I need to be clear now when I mix a nestor into the game.

Douglas Self - Wikipedia

en.wikipedia.org

Class B or class AB?

 

[Septentrio]

Further commentary from Scnieder including comments re ASR and Amir

" From what we can tell, the measurements there try to assess the H95 almost as if it is a standalone DAC. But it's not a standalone DAC -- it's an integrated amplifier with a DAC. Also, it assumes that the line outputs mean it's a preamplifier -- but they're really line outputs that *could* be to use it as a preamplifier, but some people are saying it's to attach a sub or output a signal to something else. Nevertheless, what appears to have happened there is that in treating it like a standalone DAC, the measurer put the volume up to 2V, which resulted in the output from the line outputs clipping, so it was distorting. Now some could argue that they'd like to see it output 2V, which is valid in some ways. But the fact of the matter is that the measurer should've realized that it was distorting and start experimenting at lower levels and simply reporting the behaviors to give readers more complete information. Also, the measurer should've really been looking at its behavior in the context of what the product is -- an integrated amp. Why they got such poor linearity AND jitter measurements, however, we don't understand. For us, the jitter immunity was excellent and the deviation from linearity respectable given the DAC architecture."

" If you want kind of our overriding view, it's this insofar as consumers go -- use the H95 within its rated power (60WPc) and don't drive speakers less than 4 ohms with it. For us, those are the biggest takeaways from the measurements. The stuff that ASR dwelled on, when you look at it in the context of an integrated amplifier, not a standalone DAC, they missed the point."

" BTW, one thing to remember about Amir's history in publishing was that he was one of the founder's of WhatsBestForum.com, which is a forum based on promoting the most expensive equipment possible using purely subjective methods. They have promoted some of the most nonsensical ideas for years -- as well as promoted some of the most expensive gear in the industry. He departed from there some time ago, I understand, and started AudioScienceReview.com. I don't know what the goal this time is, but he seems to have swung in the opposite direction using measurements as a club and trying to tear companies down. It seems that whatever he does -- the outright lunacy of WhatsBestForum.com to the "subjectivized" objective take of AudioScienceReview.com -- it appears to be just for show. I'd take most of it with a grain of salt. Though that's only my opinion -- everyone should draw their own conclusions. I withdrew my account from AudioScienceReview.com a long time ago and asked that all my posts be removed from that forum because I didn't want to be associated with the ideas being promoted there, such as the "gotcha" attitude I outlined above."

" I would hope so. If someone wants to take a truly objective approach, do some investigation and digging in. BTW, one approach that's promoted there that has NO basis in science or the research that's been done in hi-fi is the use of measurements to listen. I can't believe that's going on there -- it's one of the reasons I wanted no association with that forum (when ASR was being created, Amir emailed me several times to contribute, because of my experience with speaker measurements). But what's being trumpeted as "science" in that regard is nonsense again. I made a video about it.

"

Further commentary from Scnieder including comments re ASR and Amir

" From what we can tell, the measurements there try to assess the H95 almost as if it is a standalone DAC. But it's not a standalone DAC -- it's an integrated amplifier with a DAC. Also, it assumes that the line outputs mean it's a preamplifier -- but they're really line outputs that *could* be to use it as a preamplifier, but some people are saying it's to attach a sub or output a signal to something else. Nevertheless, what appears to have happened there is that in treating it like a standalone DAC, the measurer put the volume up to 2V, which resulted in the output from the line outputs clipping, so it was distorting. Now some could argue that they'd like to see it output 2V, which is valid in some ways. But the fact of the matter is that the measurer should've realized that it was distorting and start experimenting at lower levels and simply reporting the behaviors to give readers more complete information. Also, the measurer should've really been looking at its behavior in the context of what the product is -- an integrated amp. Why they got such poor linearity AND jitter measurements, however, we don't understand. For us, the jitter immunity was excellent and the deviation from linearity respectable given the DAC architecture."

" If you want kind of our overriding view, it's this insofar as consumers go -- use the H95 within its rated power (60WPc) and don't drive speakers less than 4 ohms with it. For us, those are the biggest takeaways from the measurements. The stuff that ASR dwelled on, when you look at it in the context of an integrated amplifier, not a standalone DAC, they missed the point."

" BTW, one thing to remember about Amir's history in publishing was that he was one of the founder's of WhatsBestForum.com, which is a forum based on promoting the most expensive equipment possible using purely subjective methods. They have promoted some of the most nonsensical ideas for years -- as well as promoted some of the most expensive gear in the industry. He departed from there some time ago, I understand, and started AudioScienceReview.com. I don't know what the goal this time is, but he seems to have swung in the opposite direction using measurements as a club and trying to tear companies down. It seems that whatever he does -- the outright lunacy of WhatsBestForum.com to the "subjectivized" objective take of AudioScienceReview.com -- it appears to be just for show. I'd take most of it with a grain of salt. Though that's only my opinion -- everyone should draw their own conclusions. I withdrew my account from AudioScienceReview.com a long time ago and asked that all my posts be removed from that forum because I didn't want to be associated with the ideas being promoted there, such as the "gotcha" attitude I outlined above."

" I would hope so. If someone wants to take a truly objective approach, do some investigation and digging in. BTW, one approach that's promoted there that has NO basis in science or the research that's been done in hi-fi is the use of measurements to listen. I can't believe that's going on there -- it's one of the reasons I wanted no association with that forum (when ASR was being created, Amir emailed me several times to contribute, because of my experience with speaker measurements). But what's being trumpeted as "science" in that regard is nonsense again. I made a video about it.

"

Abhorrent ad hominem attacks.

 

[DanielT]

Class B or class AB?

That said, that was a question.

Attached is what Duglas Self wrote about class A in all cases. But we let it be now, I think. Partly OT partly because it does not have dignity compared to Amir now in the thread must answer various accusations against him. We give Amir free rein. So in another thread what Douglas Self may have said.

 

[amirm]

BTW, one approach that's promoted there that has NO basis in science or the research that's been done in hi-fi is the use of measurements to listen. I can't believe that's going on there -- it's one of the reasons I wanted no association with that forum (when ASR was being created, Amir emailed me several times to contribute, because of my experience with speaker measurements). But what's being trumpeted as "science" in that regard is nonsense again. I made a video about it.

I never asked Doug to come here let alone write anything for us (or anyone else for that matter). I extended a warm welcome to him once he did show up on his own based on misconceptions I had about him caring about measurements and such. When I realized those measurements were made by NRC and not him, and he actually not caring about them, I put the screws on him in a thread here resulting in him bolting out.

 

[pozz]

It should be clear now how much stupidity and pettiness you'll find in mainstream hifi.

By the way, the SS and ASR linearity graphs are largely the same except that they use different filters and scaling. Not at my computer but someone else could easily extract the data and plot it for comparison.

 

[sarumbear]

Yeah. Nothing.

This:View attachment 168308
is not brighter than this:View attachment 168310

Trust your ears!

If you are younger than 50 years old you may consider having your ears cleaned.

 

[steve59]

I just read the link to soundstage measurements page and am struggling to understand what's wrong with the H95?

 

[amirm]

First of all thank you for taking the time to measure the H95.

Thank you for your participation as well. I have a question about your diagram and feedback here:

The "DAC" output is prior to amplification block. It is capable of going up to 2.4 volt. Why would that stage care that the amplification block *after* it clips? If I cascade an external DAC to your amplifier and your amp clips, nothing happens to the performance of the external DAC. Why is it that by pulling that functionality in, you wind up disturbing the performance of the DAC in the same scenario?

The answer in AV receiver world is that when their amplifier clips, it impacts the voltage rails feeding the DACs and hence the increased distortion. Is that the same here? If so, it is a shame that in a high-end device like this, you don't have independent power supply for the digital/DAC section. Or simple ability to disable the amplifier.

 

[amirm]

This measurement is, in our view, wrong. This is how it should look (our own test on AP 2722 today)

This measurement shows a drop of 4 dB in performance as the unit warms up -- worse than what I showed! Many audiophiles think that by leaving their equipment on, they get better performance. Here it is backward by not an insignificant amount.

 

[MZKM]

This measurement shows a drop of 4 dB in performance as the unit warms up -- worse than what I showed! Many audiophiles think that by leaving their equipment on, they get better performance. Here it is backward by not an insignificant amount.

I think they were mainly addressing the erratic nature shown in your graph, as they mentioned only via wiggling the cables could they achieve it. Is it possible to smooth this measurement?

 

[Human Bass]

77ohm on a headphone output is borderline criminal.

 

[sarumbear]

I think they were mainly addressing the erratic nature shown in your graph, as they mentioned only via wiggling the cables could they achieve it. Is it possible to smooth this measurement?

Why would smoothing matters? It should be as flat as possible.

 

[Human Bass]

The philosopher is historically important but unarguably very weird and arguably (and in fact) complete crap.

Hegel was extremely totalitarian. All his obnoxious talking about geists, and he comes to the conclusion that the state is the will of God on earth.

 

[PeteL]

v

I want to make sure we don't lose the plot here. So much attention is put on the DAC output but there is a real, undisputed problem with the amplifier. Here is how it performs using analog input:

Here is how it performs when you switch the analog input to digital, with nothing else changed:

We lose 8 dB of performance in the form of raised noise floor due to using the internal DAC.

In other words, if the amplifier used precisely as they intend, either with analog input or digital, the digital input takes a major toll on performance in noise department which is far more audible than distortion. This of course impacts streaming to the amplifier just as well.

Company has confirmed the amplifier measurements so what we have is the final word.

I showed the reason for above in my IMD measurements, showing the high noise floor:

So the DAC measurements were definitely investigative and proved the high noise floor of the internal DAC.

For an integrated amplifier and streamer, this is unacceptable. I, and I am sure many others consider digital input to be "cleaner" due to analog path being very short internally, and gains optimized. Yet we have the opposite here.

The job was not hard. All the DAC had to do was be 10 dB better than the amp. Then it would have 0.5 dB impact on its SINAD.

Conclusions
Measuring Hegel H95 as intended, i.e. with analog or digital input producing just 5 watts, produces far more noise with digital input than analog. This is a bad thing in even mass market products. It is really, really bad in the case of a premium product.

I hope Hegel looks at sources of this performance loss with digital input and remedies it. Had it not shown this problem, it would have gotten an acceptable review from me.

But the Dac is measured against a -24 dBSPL reference Tone (which is your zero on your scale), of course the noise floor will show as high. Measure any of the DAC at -24 dB SPL, what do you get as a SINAD?

 

[MZKM]

v

But the Dac is measured against a -24 dBSPL reference Tone (which is your zero on your scale), of course the noise floor will show as high. Measure any of the DAC at -24 dB SPL, what do you get as a SINAD?

Does the Hegel offer input level offsets? If not, then if the digital input is closer to 0dBFS, it’ll get much louder when you switch from an analog input to a digital one, as the volume needs to be at ~95 on the Hegel to achieve the THX 29dB amp gain.

 

[PeteL]

Does the Hegel offer input level offsets? If not, then if the digital input is closer to 0dBFS, it’ll get much louder when you switch from an analog input to a digital one, as the volume needs to be at ~95 on the Hegel to achieve the THX 29dB amp gain.

??? Analog inputs can come at any volume. Offset against a analog source at which voltage do you want? In this case an analog input of 0.157 V correspond to a signal at -24 dBFS We could do the math for an input at 0 dBFS.

 

[amirm]

v

But the Dac is measured against a -24 dBSPL reference Tone (which is your zero on your scale), of course the noise floor will show as high. Measure any of the DAC at -24 dB SPL, what do you get as a SINAD?

So was the analog voltage level. The amplifier volume control was not touched so the internal DAC was producing the same voltage as I fed it analog input.

 

[amirm]

Why would smoothing matters? It should be as flat as possible.

It can filter out the ups and downs that my measurement was showing.

 

[PeteL]

So was the analog voltage level. The amplifier volume control was not touched so the internal DAC was producing the same voltage as I fed it analog input.

Of course, sure, but think about it Amir, for exemple, look at this curve:

And look at the delta between 0 dBFS and -24 dBFS.

Now look at this curve, and look at the Delta btween 0.157V and 2V

You know better than draw conclusions based on comparison that can't compare.

 

[Bogda]

This measurement shows a drop of 4 dB in performance as the unit warms up -- worse than what I showed! Many audiophiles think that by leaving their equipment on, they get better performance. Here it is backward by not an insignificant amount.

Their measurement does not show two big spikes that are present on your diagram.