Hegel H95 Review (Streaming Amplifier)

[Phorize]

Cool, so that means they are willing to pick out their amp in a double blind test?

That’s quite a tantalising thought.

 

[Geert]

Hegel H95 DAC Measurements
This being an amplifier, I immediately jumped to measuring that part of it first. But then I had to backtrack when I noticed digital input performance was worse than analog. Given the line out, I decided to measure the DAC using coax input. For some reason, like a Hegel DAC I tested in the past, my ASIO4ALL interface would not see the USB input so I decided to test with Coax. Here is our dashboard with volume control adjusted to produce the nominal 2 volts

@amirm, for my understanding, how far did you need to open the volume control to reach 2 V? And say people would use the line output to drive a sub, is this volume control setting something people might use in practise when using a digital input or would this level drive the power output stage of the amp into clipping?

Just trying to understand if people might effectively run into the issue of overdriving the line output. If not, do you consider remeasuring the DAC at a lower output level?

 

[aliqaz]

My recent post on the facebook group:
I think we are now entering into the damage control phase by Hegel. Unlike companies D'agostino or Pass Labs that actively add coloration, Hegel have promoted them selves as an engineering company that strive to create neutral amps: No coloration, low noise and low distortion. As they own the same audio analyzers as ASR I am certain they knew exactly the performance of their products. The problem is that we as consumers lacked context. With the ASR review we can directly contextualize and compare Hegel's engineering performance claims with competitors and thus see if the premium prices they charge are worth it. If you can get something with the same level of engineered performance for pennies on the dollar, why would you buy Hegel.

And Hegel's official response:

"We are not in any damage control phase. We do stand by our claims. I should know as I have been here for 13 years. That has been 13 years of reviews and measurements that confirm our claims. Then comes one that says exactly the opposite.
As mentioned earlier we want to get to the bottom of finding out how ASR came to these results and then we can have a solid opinion about them.
There is no damage control, there is only a sensible approach to claims that you do not understand."

 

[thyristor]

Hi All,

I am the original author of the facebook post highlighting this review. I'll paste all official Hegel company employees comments here:

"You have totally missed the point. Noise and distortion measurements have some questionable values when it comes to sound quality in an amp. You can easily achieve very good measurements by introducing a lot of feedback loops and elaborate circuitry, however, that doesn't matter when the amp sounds bad and boring in the end of the day. Marantz PM90 is a typical Japanese late 80's construction, it sounds ok, but rather dull and lacks dynamics. NAD 2200 is powerfull and dynamic, but coarse and lacking in detail. Nobody with working ears would consider any of these remotely close to any current Hegel amp in sound quality.

Marantz PM-90 is a properly engineered amplifier unlike this junk.

Marantz PM-90 Review (Vintage Amplifier)

This is a review and detailed measurements of the Marantz PM-90 Vintage stereo integrated amplifier. It was purchased by a member from japan and drop shipped to me for testing. It came out in 1991 with a price of 160,000 Yen (about $1,500 in today's exchange rate). I looked online and I can...

www.audiosciencereview.com

 

[M00ndancer]

It would have been nice to see the Hegel USB port measured. That's my favorite sound wise. The USB port shows up as a sound card in the MacBook Pro sound preference. It's labeled as “TE7022 Audio w/ SPDIF” as the Output Device.

It's the same chip as in my HAD-1 from swissonic, (Thomann in-house brand)

[Teardown and Measurements] Swissonic HAD-1 AD/DA converter + headphone amplifier

Hey guys, this is actually my first post on this forum so maybe a little introduction is in place. I am an electrical engineer from Finland and I've been in the headphone audio scene for quite some time now, went through a lot of gear and decided to do some measurements of my own. My measurement...

www.audiosciencereview.com

 

[voodooless]

Just trying to understand if people might effectively run into the issue of overdriving the line output. If not, do you consider remeasuring the DAC at a lower output level?

Actually we don’t need to. We have SINAD data from just before clipping. It’s about 80.

 

[Geert]

Actually we don’t need to. We have SINAD data from just before clipping. It’s about 80.

The amplifier section won't perform any better indeed, but I wonder what it would do to the DAC's performance. As you can read in Hegel's response this line output level thing is important for the credibility of the review.

 

[Somafunk]

Tips to Hegel. If you are going to go to war, you should know what your enemy is capable of.

Before you release the dogs it’d be prudent to wait for Hegel’s response, not a fan of a smug forum pile-on as it does no one any favour.

 

[voodooless]

The amplifier section won't perform any better indeed, but I wonder what it would do to the DAC's performance. As you can read in Hegel's response this line output level thing is important for the credibility of the review.

Like I said, we have that:

 

[PuX]

H95 is an integrated amp, not a preamp, so a 2 volt signal on the variable output is outside parameters and will be heavily distorted.

I don't get this argument at all.

2V is the most common output level for a DAC on RCA outputs, isn't it?

why would it be a problem?

and why didn't they restrict it lower if 2V too high? (it isn't, their argument makes no sense).

 

[caught gesture]

Before you release the dogs it’d be prudent to wait for Hegel’s response, not a fan of a smug forum pile-on as it does no one any favour.

Hegel’s comments on Facebook are not helping them. Making comments like, If somebody would prefer this to a Hegel H95 when it comes to sound quality, please be my guest. They are probably the same people that would prefer a Budweiser over a Budvar, comes across as elitist and derogatory.

 

[nyxnyxnyx]

I have a hard time understanding this result, fellas.
What's more likely to be the case here?:
A) They are already famed and successful so they just take whatever unfinished designs they have and sell it at a high price because they know people will just buy it anyway
OR
B) Despite having years of experience and receiving praises in this industry, they are somehow still incapable of creating objectively sound products?

I mean like, I don't even NEED or EXPECT every product to have 0,0000000001% distortion or 120dB+ SINAD with ruler flat FR or something. But don't they want to at least design some statement products just to guarantee that they CAN make good products when they aim to.

Schiit really did make a lot of people shut up and stop talking shit about their products when they released Modius & Magnius. It proved that despite having inadequate measurements on other products from them, they are more than capable to design SOTA devices to prove their worth as engineers.

 

[Geert]

Like I said, we have that:

You're right, the issue is clear and Amirm identified the saturation issue. But there's still the first dashboard with an FFT graph and SINAD numbers which refer to the saturation scenario if I'm not mistaken. It also doesn't answer the question if people might run into the issue of saturating the line output in practise.

 

[Geert]

I don't get this argument at all.

2V is the most common output level for a DAC on RCA outputs, isn't it?

why would it be a problem?

and why didn't they restrict it lower if 2V too high? (it isn't, their argument makes no sense).

Has been discussed earlier on. Hegel commented the line output is not a preamp out.

Ages ago the standard reference level used to be -10 dBV (0,316V). Not saying all is fine, but it's clear lots of people will laugh this review away because of this issue.

 

[Ein]

We're all doing it wrong. Need to put our laptops on big spikes to get the sound right

We're all doing it wrong. Need to put our laptops on big spikes to get the sound right.

Yes, that's change everything at all! Laps on spikes and you have audio-nirvana. Only for 500$ after 10% diss.

BTW. Digital path on H95 is bad joke, also headamp... for what, for V on spec only? Also unusable pre-out. So, summary of H95: only two RCA in (not 2V compatible LOL!) and average power amps here. For 2k dolla... just "wow"

 

[KMO]

You're right, the issue is clear and Amirm identified the saturation issue. But there's still the first dashboard with an FFT graph and SINAD numbers which refer to the saturation scenario if I'm not mistaken. It also doesn't answer the question if people might run into the issue of saturating the line output in practise.

The -63dB @ 2V SINAD figure is can be seen on the graph at 0dBFS. You can see it achieves its best SINAD of -82dB is as -12dBFS, which is about 0.5Vrms (1.4Vp-p), then it starts to clip.

That -82dB is better, but still bad, in the middle of the red group for DACs. The output limit is not its only problem.

Has been discussed earlier on. Hegel commented the line output is not a preamp out.

Ages ago the standard reference level used to be -10 dBV (0,316V). Not saying all is fine, but it's clear lots of people will laugh this review away because of this issue.

It still is, for normal interconnects. It's only power amplifiers that typically demand more - they expect the pre-amplifier to have actually pre-amplified a bit! Pre-outs can usually do at least 1.2Vrms, but you sometimes need a bit more. Amir's 2Vrms should be plenty for any power amp.

So yes, this amplifier only has "line outs", which are perfectly adequate if you wanted your streamer output to send off to some other integrated amp, but it's not really ideal for connecting to a power amp.

Whereas I gather a lot of modern DACs are designed to run high enough to go straight into a power amp.

I do agree that only showing the FFT for 2V in this case is unhelpful - it's probably not going to be representative of what the "normal" 80dB SINAD that doesn't hit clipping looks like.

 

[Kegemusha]

I dont get it, dont they measure their own products? This amp should have never left the HQ. How can this be possible? Should be fairly easy for them, and many others, to do some quality measurements. And cost a lot too.

 

[CedarX]

US patent 6,275,104 was granted to Bent Holter (Hegel AS) on Aug 14, 2001…. Is H95 sudden bad performance caused by an expired patent?

 

[restorer-john]

To give them a fair chance; line level for consumer products used to be -10 dBV, being 0,316 V. Although that standard got lost once CD players started using the full digital dynamic range instead of maintaining head room (the reason why on a lot of amplifiers you can only use 1/4th of the range of the volume control).

The defacto 'standard' for line level inputs on consumer HiFi for many decades, right up to and well past the introducing of Compact Disc was 150mV for full rated power.

 

[AdamG]

I dont get it, dont they measure their own products? This amp should have never left the HQ. How can this be possible? Should be fairly easy for them, and many others, to do some quality measurements. And cost a lot too.

It all comes down to this (try not to laugh): Quote; “There is no damage control, there is only a sensible approach to claims that you do not understand”. We simply just don’t understand, understand?