Review and Measurements of Benchmark AHB2 Amp

[GXAlan]

thx @pjug.
So in normal language a well designed and engineered amplifier working not in clipping or distortion have inaudible differences? Can we say this?

I agree, but why don’t we make the statement

Well designed and engineered amplifiers working below clipping and audible levels of distortion are indistinguishable under controlled listening. Audible differences between can be heard with non-ideal “designs and engineering” such as a very “difficult” speaker load in combination with a highly load-variant amplifier design such as a chip amp.

all facts. Less dogmatic, more scientifically accurate.

 

[RichB]

Which decision if I may ask?

There is a clue in my signature

- Rich

 

[Pjetrof]

@RichB
nice System!
Personal question if you replace the benchmark with 5xhypex nc400
at your place everything else stay the same would you hear a diffrence?

 

[Pjetrof]

@GXAlan

yes I agree with you.
But the clip from audioholics in my opinion a difficult load like they say themselves and you drive them with a 15 watt amp and then with a 150 watt amp. I m not an EE but could it be that the diffrence in watt is the biggest factor there was an audible diffrence?

 

[RichB]

@RichB
nice System!
Personal question if you replace the benchmark with 5xhypex nc400
at your place everything else stay the same would you hear a diffrence?

I would say so, but others would find it offensive to do so on this site

- Rich

 

[cjm2077]

@GXAlan

yes I agree with you.
But the clip from audioholics in my opinion a difficult load like they say themselves and you drive them with a 15 watt amp and then with a 150 watt amp. I m not an EE but could it be that the diffrence in watt is the biggest factor there was an audible diffrence?

No. The lack of feedback from output to the chip amp control system is the main difference. They weren't hitting a current limit, or seeing distortion.
The majority of amps with the majority of loads sound identical until you hit clipping, it's true. But there are edge cases where the amp may have a design issue or the load may be particularly difficult in some way that can lead to two different amps behaving differently. And some amps are just intentionally designed to be inaccurate in some manner, or incompetently designed. Those amps are also spread across the price spectrum from the cheapest to most expensive, so paying more for an amp garauntees you nothing. Places like this forum where products are measured and differences and designs discussed provide a valuable resource exactly because of that.

 

[etc6849]

I honestly noticed a difference when I swapped all my amps out. In my case, I went from hearing hiss to hearing no hiss, so 100% audible for me with my sensitive speakers. I also had decent amps before too. My setup is pretty complex, but I can tell you that I absolutely still get goose bumps when listening because it sounds that real, even three years later. I have never experienced that from any other setup I've owned.

It is also much easier to hear background chorus and small details with the benchmark too. I would bet that the IMD (https://en.wikipedia.org/wiki/Intermodulation) is vastly lower than the amps I had before too.

Benchmark has also done double blind testing:
https://benchmarkmedia.com/blogs/application_notes/power-amplifiers-the-importance-of-the-first-watt

I appreciate the skepticism (don't blame you as there is a lot of snake oil and BS on other forums), but there are measurable and audible differences between these amps and traditional amps as pointed out by others. If what you say is true, we'd all be getting the same sound with AVRs, which of course we don't.

My system is not normal by any means (each speaker is rated to over 120dB and I'm actively tri-amping them), but I just love the dynamic range and clarity I'm able to get. It absolutely sounds like I am there whether the volume is low or high.

where the previous amps broken? There should be no audible diffrence!

 

[GXAlan]

IMD in music is definitely under appreciated and poorly quantified with the 19/20khz measurement. I keep saying this, but The Echo Game from the House of Flying Daggers Soundtrack is an incredible IMD test. Compared to the movie mix, which sounds great on any decent setup, the CD sound track sounds like a bad mix UNTIL you play it back on a low IMD/low distortion system. Another Day of Sun from the La La Land sound track is also good.

All that said, I have gone all-in to assess another “edge” case. I picked up a Kenwood L-08M with a 15,000 damping factor thanks to Sigma Drive. Native damping factor is supposed to be as high as 20,000 at the amp terminals.

http://www.cieri.net/Documenti/Cata...rive Amplifier System (07-1981) (English).pdf

 

[witwald]

So in normal language a well designed and engineered amplifier working not in clipping or distortion have inaudible differences? Can we say this?

It potentially depends a little on how one defines "well designed" when considering an amplifier. An amplifier that has a flat frequency response in its pass-band, but is -3dB at 20 Hz and 20 kHz, could be regarded as being well designed for the task at hand, all other things being equal. However, there would likely be quite audible differences in sound quality if this amplifier were compared to another amplifier that was -3dB at 5 Hz and 80 kHz, but those differences would be quite dependent on the frequency content of the programme material. Of course, the measurements would immediately bring the differences to light.

We also need to decide what damping factor we would take as being the minimum acceptable value for a well designed amplifier. A damping factor of 20 rather than 200 will result in an amplifier that is much more sensitive to the varying impedance presented by a typical loudspeaker system. That sensitivity will show up as peaks and dips in the amplifier's frequency response curve while driving any given loudspeaker system. That is, the frequency response of the amplifier will depend on the loudspeaker that it is connected to it (it is load dependent).

 

[Pjetrof]

So after reading the last few posts, @witwald @GXAlan @etc6849 @RichB and others who have same experiences
nobody should start being offensive. Your general consensus is that amplifiers sound different.
no distortion no clipping same room same speakers same loudness same song.
question At the manufacturers who are on this forum.
the consensus is there is no audible diffrence, how do you think about that!
@March Audio
@nord audio
@Apollon Audio
@benchmark audio
@orchardaudio
@ All who make amplifiers
maybe a poll for manufacturers just yes or no answer

does amplifier sound the same
no distortion no clipping same room same speakers same loudness

 

[GXAlan]

Your general consensus is that amplifiers sound different.
no distortion no clipping same room same speakers same loudness same song.

You list manufacturers, except for @John_Siau who has explained why one owner’s vintage Accuphase amp may sound different from the AHB2.

That’s not what I am saying. I am saying that
1) A chip amp with a Polk L200 at 2.83V is audibly and measurably different with a FR sweep than an Emotiva amp.

I will add
2) An Accuphase P-20 configured for a “Damping Factor” of 1, will ABX differently from an AHB2 with a complex speaker load.

3) A JBL 708P with audible hiss in a quiet room at idle, will ABX differently than a 708i with an AHB2. This hiss will he heard in the silent periods of classical music.

 

[John Atkinson]

I am highly suspicious of that plot.

blueone said:
>> I've never seen a 1 ohm dip either, only some 2 ohm dips. Here's one I saw mentioned on another forum that dips below 2 ohms right in the heart of the range you might really need some power:

https://www.stereophile.com/floorloudspeakers/104legacy/index.html

View attachment 60885 < <

My apologies for the tardy response, as I have only just seen your post. Why are you suspicious of the Legacy loudspeaker's impedance plot?

John Atkinson
Technical Editor, Stereophile

 

[SIY]

Why are you suspicious of the Legacy loudspeaker's impedance plot?

For a ported or PR system, I'd expect to see two magnitude peaks in the bass range, with a valley between at the box frequency and the height ratio depending on the alignment. The magnitude looks like it's upside down.

The rise of magnitude toward lower frequencies is also unusual. I've seen something vaguely like that with things like a Clarke-Geddes arrangement, but there's no mention of that in the review, and it's rather rare.

 

[etc6849]

@Pjetrof

Yes clipping would give really bad distortion... But, just because there is no clipping, doesn't mean there is no distortion.

You may listen with a dishwasher going in the background and mask some of the distortion of an average amplifier or you may not be willing to change out upstream components that are limiting the performance of an amplifier like the AHB2. Imagine the dynamic range you'd lose by hooking the AHB2 to an AVR for example, versus a traditional DAC.

The AHB2 happens to have the lowest distortion of any amp that I've seen measurements for, and is the worlds best from a measurement perspective. It is so well designed, that if a lower distortion amp does come along, you really wouldn't be able to hear a difference and you will be limited by other components in the signal chain, room noise and acoustics, etc...

There are certain amplifiers like the AHB2 that are so well designed that they indeed sound more accurate; even non-musicians like myself can hear the difference. Most amplifiers have a SINAD in the 76dB range, so it shouldn't surprise anyone that if you have a quiet well treated room, somewhat sensitive speakers and a nice DAC, you can hear almost a 40dB improvement (and note that Amir's measurements are at 5 watts so it is a very fair comparison between the amps, and performance at lower wattages do make the most practical difference in real world performance).

Hopefully you aren't offended by anyone here trying to help. The simplistic view found on sites such as avsforum that all amps sound the same because only amplitude and frequency range are important and that only things like clipping matter is not the entire story. This site happens to have a large number of technical experts and a very bright owner and electrical engineer who has leading edge test equipment.

 

[John Atkinson]

For a ported or PR system, I'd expect to see two magnitude peaks in the bass range, with a valley between at the box frequency and the height ratio depending on the alignment. The magnitude looks like it's upside down.

Thanks for getting back to me. Yes, the Legacy's impedance plot - fig.1 at https://www.stereophile.com/content/legacy-audio-focus-2020-loudspeaker-measurements - does looks strange. The magnitude isn't upside down as there is the usual increasingly inductive rise in magnitude visible above 10kHz.

The rise of magnitude toward lower frequencies is also unusual. I've seen something vaguely like that with things like a Clarke-Geddes arrangement, but there's no mention of that in the review, and it's rather rare.

I wondered at the time whether there was a large series capacitor in the woofer feed, which would result in both the rising magnitude below the apparent port tuning frequency and the increasingly capacitive phase angle shown in the impedance graph. But if you look at fig.3 and 4 in the review, this is a loudspeaker with an extremely and possibly unnecessarily complex low-frequency drive-unit array.

John Atkinson
Technical Editor, Stereophile

 

[SIY]

Thanks for getting back to me. Yes, the Legacy's impedance plot - fig.1 at https://www.stereophile.com/content/legacy-audio-focus-2020-loudspeaker-measurements - does looks strange. The magnitude isn't upside down as there is the usual increasingly inductive rise in magnitude visible above 10kHz.

I wondered at the time whether there was a large series capacitor in the woofer feed, which would result in both the rising magnitude below the apparent port tuning frequency and the increasingly capacitive phase angle shown in the impedance graph. But if you look at fig.3 and 4 in the review, this is a loudspeaker with an extremely and possibly unnecessarily complex low-frequency drive-unit array.

Appreciate the response. Still, I'd love to get my hands on one to see if I get the same results. There's just too much weird going on there.

 

[Pjetrof]

@etc6849
no no not offended.
its from this site I took the stand amplifiers diffrences is inaudible!
and find it very strange none of the other forum members who in my thread “amplifiers can you hear a diffrence”
do not participate in this thread. I m not going to re quote plenty of forum members.

can someone help me with putting up a poll simple question like my thread amplifiers can you hear a diffrence
yes or no
is this possible?

 

[etc6849]

These two application notes are worth reading:
https://benchmarkmedia.com/blogs/application_notes/ahb2-driving-pmc-mb2s-studio-monitors
(short discussion on audible noise level)

https://benchmarkmedia.com/blogs/ap...preting-thd-measurements-think-db-not-percent
(has a great example calculation on noise level SPL)

I really don't see a point in participating in a thread like the one you mentioned, aside from making one or two posts. Whether you'll hear the noise from an amp really depends on too many factors like playback chain, what you are playing back, speaker sensitivity, listening environment and even how sensitive your hearing is.

In general, Benchmark shows in those two application notes how their amps can offer no audible noise for a very wide range of speaker sensitivities in any listening environment. If you read and study those, you'll be able to do general claculations showing which amps will have audible noise with your speakers.

As for IMD, I think lowering it too matters greatly, and it's one of the reasons I originally actively tri-amped my speakers before I bought the benchmark amps. There is no going back to a traditional amp/speaker setup for me!

@etc6849
no no not offended.
its from this site I took the stand amplifiers diffrences is inaudible!
and find it very strange none of the other forum members who in my thread “amplifiers can you hear a diffrence”
do not participate in this thread. I m not going to re quote plenty of forum members.

can someone help me with putting up a poll simple question like my thread amplifiers can you hear a diffrence
yes or no
is this possible?

 

[GXAlan]

@etc6849
its from this site I took the stand amplifiers diffrences is inaudible!

It’s not black and white. Different speakers account for the biggest difference in sound quality so you should always spend the most amount of money you can on speakers.

With level matching and running in the “most linear” part of the amplifier, the differences will be small, and generally inaudible.

But that is not dogma. All of the quotes talk about “well designed” and that simply isn’t the case for everything.

Most amps have a reasonable enough SNR that you won’t have hiss at the listening position. But with a high efficiency speaker, that’s a different story.

Some speakers dip down to 1 or 2 ohms which may not be driveable by all amps reliably at high volumes.

Damping factor is often said to be good enough past a certain point but the actual measured damping factor of some AVRs according to audio.pl is only about 50-60. That is below the threshold of what might matter, depending on your speaker cable length.

Then music matters. Not all content will be sensitive to these minute differences. Hiss at baseline only matters if your music has periods of silence or near silence (classical). IMD differences are a another example because highly processed multitrack recordings are where you see the biggest differences.

edit: some classical recordings have a baseline hiss/noise that exceeds amplifier noise as well.

 

[Coach_Kaarlo]

Interesting discussion.

Presumably one can measure everything one hears therefore the measurements are an accurate tool to determine what works/ sounds best.


In my mechanical engineering experience the process of metrology is largely about trying to avoid contaminating what you are trying to measure - getting true repeatable and accurate measurements is very difficult - particularly as the speed of the variable being measured increases. Holding a steel bar in your hand while measuring microns will result in a measurement of the thermodynamic properties of the metal - not it's true size (the bar expands as it warms up). Measuring gravity on the Earth's surface results in a range of results; from 9.7639 m/s2 on the Nevado Huascarán mountain in Peru to 9.8337 m/s2 at the surface of the Arctic Ocean. The built environment is engineered using an approximate value of 9.81 m/s2. An audio engineering example of what I mean is the way the nominal impedance of a speaker is used singularly by manufacturers and consumers when the measured impedance is actually variable and moves within a range of values - and arguably the minimum value is the most important.


So when we are discussing the measuring of amps do we mean steady state conditions or do we mean dynamic live loads?

If yes to dynamic live loads - then what are the common methods to measure dynamic changes in speaker demand on an amp? Some audio is clearly more difficult to drive accurately, how are amplifiers tested for this?

What are the common methods to find the most difficult load case the amp must drive dynamically? And to test this load case?


For example a large sudden current demand to extend a woofer cone is governed by a number of factors;

• voltage rise time
• current flow
• damping factor
• phase angle
• power available

How is this measured?