• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Hegel H95 Review (Streaming Amplifier)

KSTR

Major Contributor
Joined
Sep 6, 2018
Messages
2,732
Likes
6,101
Location
Berlin, Germany
But regarding the statement the DAC cannot be tested; are you sure the line out signal is derived from the speaker outputs, or can it be from a buffer stage behind the volume control. In the latter case you test the DAC + volume control, and I expect S/N to be OK.
That could be easily found out. Monitor line-out and speaker-out (unloaded) simultanously. If they are seperate the speaker-out should clip way before the line-out.
The distortion is plain clipping and it would easily identify if Amir had looked at the residual, not only the spectrum.
 

oivavoi

Major Contributor
Forum Donor
Joined
Jan 12, 2017
Messages
1,721
Likes
1,938
Location
Oslo, Norway
There are very few amplifiers that have passed for example LTS' blind test in being audible transparent and that's in a room that is far from great. That also corresponds with good number of AB tests I've conducted of both power amps and DSPs (DACs).
That's why I said "under most listening conditions" :) The LTS tests are very demanding and often focused on tiny details. And there have been a few other blind tests as well which have told amps apart, but also lots of blind tests where people can't tell any differences.

The big question for me is more or less like this: If someone switched out my amplifier without me knowing, would I ever notice that? For most listeners, in most setups, I'm fairly sure that the answer would be no. Whereas the answer is probably yes for loudspeakers, headphones and room acoustics. But of course it's difficult to switch out those things without people noticing - this is just a thought experiment for me.
 

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,602
Likes
10,768
Location
Prague
That could be easily found out. Monitor line-out and speaker-out (unloaded) simultanously. If they are seperate the speaker-out should clip way before the line-out.
The distortion is plain clipping and it would easily identify if Amir had looked at the residual, not only the spectrum.

Klaus, you have this measurement in post #1

6082EA7C-8F21-4E31-82AF-82B863961FD5.png


Imagine you want to use this product as a DAC with coax input and analog output at usual, 2V output level. Would you be satisfied with the result and at the same moment, do you see hard clipping? Hard clipping is usually 10% and more THD. Sorry, this behaviour should have been fixed by design engineer.
 
Last edited:

KxDx

Senior Member
Joined
Nov 1, 2019
Messages
347
Likes
674
Location
Tidewater Virginia
When I jointed Microsoft, we had a key competitor. I changed my home page in the browser to be their site! That way, every time I started it I would see if they had announced anything. If they had, I would immediately download, test and benchmark what they had. As a result, I was way ahead of everyone else in having that knowledge. And made sure to keep our product group accountable to match and beat them.
AMIR KILLED NETSCAPE!!!

:oops:o_O
 

Tangband

Major Contributor
Joined
Sep 3, 2019
Messages
2,994
Likes
2,796
Location
Sweden
That's why I said "under most listening conditions" :) The LTS tests are very demanding and often focused on tiny details. And there have been a few other blind tests as well which have told amps apart, but also lots of blind tests where people can't tell any differences.

The big question for me is more or less like this: If someone switched out my amplifier without me knowing, would I ever notice that? For most listeners, in most setups, I'm fairly sure that the answer would be no. Whereas the answer is probably yes for loudspeakers, headphones and room acoustics. But of course it's difficult to switch out those things without people noticing - this is just a thought experiment for me.
A little of topic, but: the cheapest dac LTS have tested in blindtesting, that couldnt be detected in the before/after listening test, was Yamaha wxc50 used as a dac , toslink in , ”playermode” without volume regulation , and the analog sound output from the RCA outputs. :). So one can not do better than that regardless of price and it only cost 350 dollars, if all that you want is a good dac. The LTS measurements of the unit was good, but not outstanding in any way.

They have tested a lot of dacs and there are many expensive dacs with impressive measurement results , even better than the yamaha wxc50, that failed the before/after listening test.

Now back to the subject:)
 

Bjorn

Major Contributor
Audio Company
Forum Donor
Joined
Feb 22, 2017
Messages
1,291
Likes
2,576
Location
Norway
That's why I said "under most listening conditions" :) The LTS tests are very demanding and often focused on tiny details. And there have been a few other blind tests as well which have told amps apart, but also lots of blind tests where people can't tell any differences.

The big question for me is more or less like this: If someone switched out my amplifier without me knowing, would I ever notice that? For most listeners, in most setups, I'm fairly sure that the answer would be no. Whereas the answer is probably yes for loudspeakers, headphones and room acoustics. But of course it's difficult to switch out those things without people noticing - this is just a thought experiment for me.
Let's not forget that many blind tests ar done with no pre-listening, getting to know both the music material and test objects. Plus often the listeners are sitting at diferent posisions in the room. That will give a null result almost every time unless the difference is huge.

It's very difficult to generalize. What some would consider as smaller details might be very audible to some. Smaller details also become much more audible with great speaker designs and well acoustic treated rooms.

One could blame the LTS test for not being conducted under great conditions (i.e. sitting close to rear wall with only som Skyline "diffusers" behind them), mostly use older listeners, etc. and therefore not being revealing enough. It will also depend on several factors.

How much AB testing with level matching between components have you done yourself?
 

Dogcoop

Active Member
Joined
Feb 12, 2021
Messages
136
Likes
269
Did anybody actually notice that it's clear the DAC itself is not distorting at all? Amir tested the "line out" in good faith but the line out is the dimmed down speaker out and to reach 2Vrms the amp is already on the edge of clipping. The DAC itself was never tested and cannot be tested.
The DAC itself performs just fine im 100% confident, as does the amp. Not SOTA performance but completely OK. The price/performance ratio looks to suffers a bit which is also quite clear why.... Norway is an expensive country, Hegel is a small-scale outlet.

Ok, this "line-out" issue was not to be expected by Amir but now I come down to the main point:
ASR tests now have gained so much impact (typically including a shitstorm from all the backseat drivers and hobby-dissers) that it is absolutely required that those test be as thorough as they can. The internet doesn't forget anything.

There is a huge responsibility and sorry to say it so directly to you, @amirm, you seem to lack that engineering thoroughness that is required by now. Don't post review/measurements when you are not 100% sure you can really nail down what's going on. Check back with the manufacturer in case you see unexpected behavior, or at least check back with the local forum experts, you know who those are.
And most importantly: take your time. IME (and I'm doing this for along time now), a proper device test takes at least one work day, double- and triple-checking everything. In case of odd behavior this can easily extend to several days, even up to a week, including some "settling time".
Thanks for your input, but I’m not familiar with “settling time.” Pardon my ignorance, but what exactly is it? Thanks.
 

KSTR

Major Contributor
Joined
Sep 6, 2018
Messages
2,732
Likes
6,101
Location
Berlin, Germany
Thanks for your input, but I’m not familiar with “settling time.” Pardon my ignorance, but what exactly is it? Thanks.
"Settling Time" (note the quotes) : Don't rush, take a break, rethink, etc.
 

oivavoi

Major Contributor
Forum Donor
Joined
Jan 12, 2017
Messages
1,721
Likes
1,938
Location
Oslo, Norway
How much AB testing with level matching between components have you done yourself?
Why is this relevant for the discussion and why are you asking, Bjørn?

EDIT: I have actually participated in several blind tests, but I'm not sure I'm comfortable with you bringing such things into the conversation
 

KSTR

Major Contributor
Joined
Sep 6, 2018
Messages
2,732
Likes
6,101
Location
Berlin, Germany
Imagine you want to use this product as a DAC with coax input and analog output at usual, 2V output level.
That's not a valid use-case for this product and of course Hegel have failed to communicate this clearly.

EDIT: And as for the IMD+N plot, this is a very condensed 1-dimensional view. We always, always have to look at the spectra (and residual, when possible) to find out what the noise is, notably when those plots (also THD+N) are noise-dominated for the larger part (which they always are and hence are of little use -- practically no insight... unless at least the standard noise slope is also shown).
 
Last edited:

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,602
Likes
10,768
Location
Prague
That's not a valid use-case for this product and of course Hegel have failed to communicate this clearly.
I have some problems with the "valid use-case" style of reasoning, to me it sounds a bit like a late excuse that must be done for marketing purposes. Nothing personal, Klaus, just different approach.
 

Bjorn

Major Contributor
Audio Company
Forum Donor
Joined
Feb 22, 2017
Messages
1,291
Likes
2,576
Location
Norway
Why is this relevant for the discussion and why are you asking, Bjørn?
Because of the claims you brought up. I'm wondering what they are based on. Actual testings you have done or something else.

I have done numerous tests of these mysels and I've find the opposite of what you're saying. I can agree though if you room is very poor (typical very small with no treatment) and/or speakers of low quality, smaller differences between well designed electronics become challenging to hear and wouldn't matter much.
 

KSTR

Major Contributor
Joined
Sep 6, 2018
Messages
2,732
Likes
6,101
Location
Berlin, Germany
I have some problems with the "valid use-case" style of reasoning, to me it sounds a bit like a late excuse that must be done for marketing purposes. Nothing personal, Klaus, just different approach.
I tend to agree... the main problem Hegel is facing here is marketing vs. reality.

Like so many manufacturers they didn't absorb that customers aren't blind mice anymore, you have to treat them at eye-level today. The days of buzzword blurb used for marketing are over because it will blow up in any indepedent technical review -- which is the main purpose of this site (and a few others).
 

voodooless

Grand Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
10,371
Likes
18,283
Location
Netherlands
Did anybody actually notice that it's clear the DAC itself is not distorting at all?
I want that DAC :cool:
The DAC itself performs just fine im 100% confident, as does the amp. Not SOTA performance but completely OK. The price/performance ratio looks to suffers a bit which is also quite clear why.... Norway is an expensive country, Hegel is a small-scale outlet.
It's a global market. It's a bit strange to complain that you produce in/from an expensive country...
Ok, this "line-out" issue was not to be expected by Amir but now I come down to the main point:
ASR tests now have gained so much impact (typically including a shitstorm from all the backseat drivers and hobby-dissers) that it is absolutely required that those test be as thorough as they can. The internet doesn't forget anything.
I don't see a problem with the test itself. It shows a specific flaw (or call it a feature if you want?) with the design. Now what might be lacking is an explanation of what's going on and possibly some further investigation. So I tend to agree with the thoroughness aspect of it. But also know that Amir probably relies on the forum to do some further digging. The issue with that is mainly that this information gets lost quite easily and doesn't make it back to the first post. That means people need to read a lot of stuff to actually figure out what's going on. Would be great to find a solution for this!
There is a huge responsibility and sorry to say it so directly to you, @amirm, you seem to lack that engineering thoroughness that is required by now. Don't post review/measurements when you are not 100% sure you can really nail down what's going on. Check back with the manufacturer in case you see unexpected behavior, or at least check back with the local forum experts, you know who those are.
And most importantly: take your time. IME (and I'm doing this for along time now), a proper device test takes at least one work day, double- and triple-checking everything. In case of odd behavior this can easily extend to several days, even up to a week, including some "settling time".
I mostly agree here, a bit more digging into the matter might be beneficial.
 

Koeitje

Major Contributor
Joined
Oct 10, 2019
Messages
2,306
Likes
3,962
I work in automotive industry and I can make a parallel to this.

I look at Hegel as a BMW. BMW by default is a premium brand with premium price on it. But, not every BMW is a M5. BMW also has striped down BMW 116i with mediocre performance then even some 30 year old cars can beat. But, 116i is still BMWs best selling and most profitable model. And its overpriced for sure. You can buy better car in every way for the price of 116i.

Same goes for Hegel H95. Not every Hegel is state of the art performance. It is actually well designed from meeting a certain criteria point of view and for who it is intended. It was a business decision to make H95.

Example: my best friend is the type of guy who would likely be the target buyer for H95. He has the funds, he likes premium brands, he is not an audiophile, but likes decent quality music and uses Spotify. He thinks I am insane because I have my own SOTA DIY DAC, custom D Class amp, spend time on forums, use Audirvana etc... He just want premium brand, nice sound and convenience.

Yes, for us, enthusiast this is a bad product, but for my friend, it is all he wants and needs.
Sorry, but this makes no sense. Nobody expects this to break SOTA territory. We'd expect at least a quality product. In your example this amplifier would be comparable to buying a brand new116i and getting one with 1990 levels of performance. The amplifier part is mediocre and overpriced, but at least its not broken. The digital part however isn't even good enough for Redbook audio. It is worse than $600 surround receivers.

Edit: reading up on some other posts. Even if its not the DAC on its own then at least the line-out from this device should be removed. No point adding something to a product with performance levels that were bad even 4 decades ago.
 

Geert

Major Contributor
Joined
Mar 20, 2020
Messages
1,944
Likes
3,548
Imagine you want to use this product as a DAC with coax input and analog output at usual, 2V output level. Would you be satisfied with the result
As discussed yesterday, there is no "usual 2V output level" for a line out. Especially not if the reference level isn't specified. All you can do is assume or wish for. Look at the recent NAD M10 and M33 amps. Also people complaining the line out level is insufficient to drive their subwoofers, and NAD making a v2 to fix it.
 
Top Bottom