• 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!

Review and Measurements of Benchmark AHB2 Amp

OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,388
Location
Seattle Area
So how do you explain a number of audiophiles who usually prefer tube designs sound?
Same way a doctor would explain why stuff you read online is not real medical science. :)

None of these audiophiles have done a proper controlled test. I have. And I can tell you they will flunk all such tests. That "tube sound" is much more folklore than real. That pre-conditions people, together with the warm glow of tubes, that there is something good there.

I have lost track of how many tube amps I have tested with such prior reviews, only to find nothing in measurements, nor listening tests indicating there is truth to it.

There should be a scientistic research blind test proof on 'listeners prefer listening to gear which objectively measures better' to accept this as a fact.
No, there needs to be one in reverse. Put those people in front of a curtain, and then switch amps behind the scenes between tube and solid state with levels matched. Once you show reliable data there, then we can investigate. Until then, audiophiles are notorious for manufacturing things in their brain that has nothing to do with the sound entering their ears. Test after test has shown this.

So please, don't bring us anecdotes. We have heard them, and there are countless proof points to say that they are just false information. Start your journey right from here on: only believe in controlled tests where the only variable is what you are interested in (not level, brand, looks, price, etc.). Then you can make easy sense of all of this.
 

zalive

Active Member
Joined
Sep 4, 2018
Messages
263
Likes
38
Answer is simple: nothing is happening in time domain.

If you do understand science then you understand you've written a non-scientific claim. Scientific experiments cannot prove non-existence. It either prove existence or don't show such things exist. But mere fact it didn't show doesn't prove non-existence. It may as well mean that methodology used was not well adjusted for the purpose.

If you think there is, then develop the test and demonstrate it.

Gladly if you send me equipment of my choice to do the testing. It costs money, right?
It's not my job to prove; my job is to ask questions and to shake some assumptions which can falsely be considered as facts.

No, there needs to be one in reverse.

Any claim needs a scientific proof to be considered a fact, this applies to your claims as well.
 
Last edited:

LTig

Master Contributor
Forum Donor
Joined
Feb 27, 2019
Messages
5,760
Likes
9,442
Location
Europe
So how do you explain a number of audiophiles who usually prefer tube designs sound?
There should be a scientistic research blind test proof on 'listeners prefer listening to gear which objectively measures better' to accept this as a fact.
One serious technical reason is the higher output impedance of tube amps, especially triode amps without negative feedback. Using a typical loudspeaker you end up with a frequency response which is no longer flat but has more bass and more treble. Of course this results in descriptions like powerful bass and more air. You can get the same result (a) with a cheap power resistor (1 - 4 Ohm) in series with the speaker, or (b) using tone controls or loudness correction.
 

March Audio

Master Contributor
Audio Company
Joined
Mar 1, 2016
Messages
6,378
Likes
9,317
Location
Albany Western Australia
Those issues may or may not be related to the power amp (other than transformer buzz). That's something you'll have to track down. You may have some issues with your sources, no way for me to tell remotely.
Some amplifiers do suffer a mechanical transformer buzz caused by mains conditions. Naim are infamous for this and I have experienced it with their amps. If an amp caused signal derived buzz / mains hum to come out the speaker I would throw it in the bin :)
 

Sir Sanders Zingmore

Addicted to Fun and Learning
Forum Donor
Joined
May 20, 2018
Messages
947
Likes
1,898
Location
Melbourne, Australia
So how do you explain a number of audiophiles who usually prefer tube designs sound?
There should be a scientistic research blind test proof on 'listeners prefer listening to gear which objectively measures better' to accept this as a fact.
I did a single blind test comparing my Line Magnetic tube amp against my devialet 120.
Sighted, the differences were enormous. Blind, I couldn’t tell them apart.
 

March Audio

Master Contributor
Audio Company
Joined
Mar 1, 2016
Messages
6,378
Likes
9,317
Location
Albany Western Australia
I really hope some light will be shed within a reasonable period over cables, in a scientific way.
Cable manufacturers claim it's not the metal that needs the burn-in, but dielectric.
Their general explanation on how dielectric (outside of burn-in time issue) affects the signal is charge-discharge cycle: dielectric charges from the signal from the conductor, then discharges back to the conductor, but with some time delay. Some describe it as a time smear. I imagine it as an echo from the signal.

So to determine whether this is true or not, I'd say we need a specifically designed measurement. Discharge from the dielectric won't supposedly generate harmonic or non-harmonic distortion because it follows the original frequency. Instead, distortion supposedly happens in the time domain only - time delayed artifacts instead of changed artifacts.

So I guess cable measurement designed to prove or disprove whether something like this happens should use something like a signal impulse of a very short duration and then screen whether it generates additional waveform (of a same frequency) in time domain, trailing the original signal impulse.

I point out, these are only my own thoughts/speculations on how the truth about this should be experimentally researched, based on what sounds logical to me.

Cable performance is very much a known quantity. There is no unknown magic going on. It has been thoroughly researched and implemented in the real world. With respect this is simply a case of you not personally knowing the technicalities and science behind it.

Think about and read around cables designed for RF use, transmission lines and use of time domain reflectometers etc.

Audio is actually a very simple application and audible effects would be eminently measurable.

TDRs do what you allude to. I have used them along with VNAs. You can see cable imperfections, where the connectors are and other impedance imperfections etc from the reflected signal. However at low audio frequencies with cables that are not km in length this is not relevant and there is no appreciable effect.
 
Last edited:

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,523
Likes
37,056
It's supposed to be of the same frequency as the signal. If you don't screen the time domain you can't see such in a graph, signal wave overshadows it.

Let us think about what you are saying. The burned in vs non-burned in cable has effects in the time domain at the same exact frequency as the signal. So you don't see it in the frequency domain. Okay, not really a big deal. Remember frequency and time domain are opposite sides of the same coin.

Now if a change is audible by definition the waveform must have changed. You are claiming a change in time domain with no change in frequency no added distortion or noise. Just going with some touted cable company explanations they'd say the dielectric when new absorbs some of the energy and then releases it after a short period of time. If well designed and once burned in there is less of this store and release behavior. Now what does that mean? It would have to mean the capacitance of the constructed cable has changed. What happens when you change the capacitance of a cable? One of the effects is it changes the response at high frequencies. So you could use however high a frequency response needed to begin to get droop in response. Using the - 3 db of response would be fine. You would expect your brand new audio cable to have a different frequency where droop was -3 db than it does after it is burned in. You would expect if the explanation is true that the frequency causing a - 3 db droop would be higher after burn in.

Now if the claim is true, would it be enough for you to hear it? We know enough about how cable works to estimate that and the change in capacitance from burn in would have to be relatively huge. And if so would be relatively easy to measure. You could feed a cable with a signal with a 100,000 ohm resistor on the source end and measure at the other end. This would accentuate the loss vs frequency. You measure when new and when burned in. Is there a change and how large would it be. And however large it is, the change would be about 1000 times less in normal use of a cable for interconnect, and even much much less for speaker cables.

For a more detailed explanation read this:

https://www.audioholics.com/audio-video-cables/dielectric-absorption-in-cables-debunked
 

zalive

Active Member
Joined
Sep 4, 2018
Messages
263
Likes
38
Let us think about what you are saying. The burned in vs non-burned in cable has effects in the time domain at the same exact frequency as the signal. So you don't see it in the frequency domain. Okay, not really a big deal. Remember frequency and time domain are opposite sides of the same coin.

Now if a change is audible by definition the waveform must have changed. You are claiming a change in time domain with no change in frequency no added distortion or noise. Just going with some touted cable company explanations they'd say the dielectric when new absorbs some of the energy and then releases it after a short period of time. If well designed and once burned in there is less of this store and release behavior. Now what does that mean? It would have to mean the capacitance of the constructed cable has changed. What happens when you change the capacitance of a cable? One of the effects is it changes the response at high frequencies. So you could use however high a frequency response needed to begin to get droop in response. Using the - 3 db of response would be fine. You would expect your brand new audio cable to have a different frequency where droop was -3 db than it does after it is burned in. You would expect if the explanation is true that the frequency causing a - 3 db droop would be higher after burn in.

Now if the claim is true, would it be enough for you to hear it? We know enough about how cable works to estimate that and the change in capacitance from burn in would have to be relatively huge. And if so would be relatively easy to measure. You could feed a cable with a signal with a 100,000 ohm resistor on the source end and measure at the other end. This would accentuate the loss vs frequency. You measure when new and when burned in. Is there a change and how large would it be. And however large it is, the change would be about 1000 times less in normal use of a cable for interconnect, and even much much less for speaker cables.

For a more detailed explanation read this:

https://www.audioholics.com/audio-video-cables/dielectric-absorption-in-cables-debunked

Those claims are audio cable industry claims, not mine. I'd like to see experiments which either prove them or disprove them.
Also, charge-discharge of a dielectric is physics phenomena which will happen - and there will be a time delay between charge and discharge. It's physics as well. What's unknown, without a properly designed experiment, is whether it happens in a measurable quantity and if yes, whether it's in quantity which allows to qualify it as audible.

If dielectric discharges into a signal with the time delay, it will certainly be a distortion - because this signal belongs to a signal/frequency which has already passed through, and doesn't belong to a signal which is happening at the moment of transfering a discharge into a signal. So, relative to the signal at the moment of a discharge, it's not just out of phase signal. It's a parasite signal from the perspective of a moment in which it's generated; it's an echo (of something that happened before) that doesn't belong to that musical moment. So it represents a distortion.

Only, if you use a test which won't show it, it just won't show. If a continuos tone is being the music, discharge will create the same content as one existing, so it won't represent a distortion. However if content has changed in between charge and discharge then discharge will add something to current content that simply doesn't belong in the moment. And this is distortion by definition.

No continuous tone testing may therefore disclose it. Test has to include a content change. A short duration tone impulse as I suggested represent this, a signal content change. So there will be the impulse and after it there will be a silence. Echo(es) caught in this silence represent distortion regardless of having the same frequency.

All this has nothing specifically related to burn-in. The idea of burn-in of dielectric materials should be about dielectrics changing their properties over its initial time of use (possibly even related to dynamics of its use). If this is the case I guess somewhere should be a scientific evidence on this by now (I don't mean from the audio cable industry - even if they do scientific research for their own needs, they've business reasons not to disclose results).

If we assume that burn in changes the sound (according to my subjective experience, it does and it does change it a lot at some moment in between, at least if PTFE is dielectric) it might mean that during the burn in period discharge is different for different frequencies and that influence is therefore more pronounced at certain frequencies, meaning distortions will be more present at certain frequencies as well. Would that mean a cable capacitance change? Maybe yes, maybe not, I wouldn't know. What if change over burn-in period is related to varying speed of discharge rather than quantity (again, time domain)?
 
Last edited:

RichB

Major Contributor
Forum Donor
Joined
May 24, 2019
Messages
1,946
Likes
2,611
Location
Massachusetts
Benchmark is selling reasonable speaker cables and interconnects. That should tell you something.

The AHB2 uses two fuses, anyone want to talk about fuses. Oh no, what have I done? :p

- Rich
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,579
Likes
38,280
Location
Gold Coast, Queensland, Australia
If burn in changes the sound (according to my subjective experience, it does and it does change it a lot at least if PTFE is dielectric)

So if burn in is real, and changes the sound, at what point do you decide the sound is optimal and more importantly, when do you notice the performance deteriorating? In other words, how do you know when to throw them out? Surely burn in and burn out are like an inverted bathtub curve I proposed in a thread a while back:

https://www.audiosciencereview.com/...-burn-in-am-i-deluded.5339/page-2#post-118141
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,388
Location
Seattle Area
If you do understand science then you understand you've written a non-scientific claim. Scientific experiments cannot prove non-existence. It either prove existence or don't show such things exist. But mere fact it didn't show doesn't prove non-existence. It may as well mean that methodology used was not well adjusted for the purpose.
Please... We have all heard this line. It means nothing. By that logic we should look to see if you eat enough radishes, you cure your cancer.

We have decades of research into what is or is not significant in audio. We don't start from ground zero every time some lay person thinks of a technical theory that they don't even know what it means.

That aside, I gave you something that demonstrates positive proof: show an audible difference using accepted scientific means. If you cannot, then we are done.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,388
Location
Seattle Area
Any claim needs a scientific proof to be considered a fact, this applies to your claims as well.
I make no claims. I speak on behalf of countless engineering and researchers in audio. I am telling you that based on that collective knowledge, you have not provided anything of value for one to go and chase. People who make up the theories are responsible for providing such proof. Otherwise, they should not invent these theories.

The audiophiles who keep saying we don't know enough about audio, are so good at audio science/engineering as to keep inventing new theories. Which is it? If they don't understand it, they should stay out of making up stuff. And worse yet, asking us to go and investigate what they imagine without a shred of back of up, either in listening tests, or physics of what they say about audio cables.

Show me another example in your life where you dispute science this way. Do you make up your own theories of what is wrong with you task your doctor to prove them???
 

GrimSurfer

Major Contributor
Joined
May 25, 2019
Messages
1,238
Likes
1,484
Scientific experiments cannot prove non-existence. It either prove existence or don't show such things exist. But mere fact it didn't show doesn't prove non-existence.

Two people carrying pamphlets came to my door earlier in the day and said the same thing. I asked them if they were audiophiles on pilgrimage. It turned out they were.

Anywhoo, I decided to join their cult and throw out all my solid state and digital gear. I'm currently sitting cross legged on a straw mat playing a phonograph record with a sharpened bobby pin. It sounds... amazing.

Tomorrrow I get my paper pants and tambourine. We're going to protest the local philharmonic which, we all know, worships the devil.

Life is so much clearer now... rational thought is such a buzzkill.
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,835
Likes
16,497
Location
Monument, CO
Impulses, square waves, and eye diagrams are routinely used to assess the time-domain response of a cable. Vector network analyzers (VNAs), spectrum analyzers, phase noise meters, noise figure meters, etc. can be used to assess system and cable performance.

There are many studies going back decades about the impact of dielectric absorption (DA)/dielectric relaxation (Debye effects), charge traps, surface roughness and loss tangents, skin effect, dispersion, etc. ad nauseum. Nothing I am aware of indicates any such effects are audible. Of course, I don't know everything, so the typical subjectivist response is to claim a lack of knowledge on my part means there is a possibility they are right (actually, they generally they claim they are right, based on what they think they hear -- somehow my bias is always wrong and their bias is always right). At audio frequencies DA is on the order of megaohms to gigaohms depending upon the dielectric and the impact of charge traps is typically in the fV range. Cables do have limited bandwidth, and hysteresis effects, and again these are readily measurable in the frequency and time domain. At work I do it all the time and cables are a significant player in performance for GHz signal transmission. Similarly, I have measured various dielectric absorption effects and noise from charge traps (using a leaf voltmeter), for systems with 140~160 dB dynamic range operating from 10 to 100+ GHz. Audio, not so much. The primary limiter for Gb/s data transmission is bandwidth, not charge traps or other things claimed to be audible, and if those were a problem we would see them since we are measuring fs to ps jitter numbers. Audio frequencies are too low, impedances too small, and cable lengths far too short for RF/mW/mmW effects to matter. Many audiophile companies prey on users by presenting factual data that, while true, is far from being applicable to audio signals.

DA is more a problem with capacitors than cables; in capacitors the effects can become audible in some cases but the usual solution is to use different capacitors with less DA (and voltage coefficients, etc.)

As for proof, read AES, IEEE, and other engineering journals, physics journals, electronics texts and reference books, or what have you. Another thing about these sorts of debates is they often lead to demands for evidence and proof of scientific/engineering data that took years to learn and that cannot be readily presented in a post or two. When years of study is not condensed into a simple post understandable by someone with high-school mathematics then the attack moves to the presenter. One reason I usually just give up and let them go on with their misconceptions; just not worth the effort to try to teach someone unwilling to learn or accept the wealth of knowledge already available, nor the personal attacks on knowledge and motives that follow when failing to put those years of study and experience into something they might understand and believe.

I am struggling to see why this digression is in the AHB2 thread.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,388
Location
Seattle Area

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,383
Likes
24,749
Location
Alfred, NY
DA is more a problem with capacitors than cables; in capacitors the effects can become audible in some cases but the usual solution is to use different capacitors with less DA (and voltage coefficients, etc.).

AFAIK, DA is only a problem in timing circuits and integrators. For things like power supplies and coupling, it has no real effect as long as the f3 is well below the bottom of the frequency range you wish to cover.

Like you, I'm not sure of what some brain lint like the wire nonsense has to to with the AHB2 amp, which is the province of real engineering, not regurgitated ad copy mixed with a severe lack of understanding of basic science.

edit: Just saw Amir's post and will desist from further OT commentary.
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,835
Likes
16,497
Location
Monument, CO
AFAIK, DA is only a problem in timing circuits and integrators. For things like power supplies and coupling, it has no real effect as long as the f3 is well below the bottom of the frequency range you wish to cover.

DA is a problem for wideband and low-noise systems. AFAIK, not for cables conducting audio signals. My comments were based upon experience with things like radar, spread-spectrum com, satcom, etc. as well as Gb digital signals (e.g. SAS, Ethernet, etc.)

On-topic, it would be interesting to learn more about the technology in the AHB2, like the signal-tracking SMPS as well as the feedforward circuitry. I assume there is a patent, should quit debating cables and go look it up.
 
Last edited:

Sal1950

Grand Contributor
The Chicago Crusher
Forum Donor
Joined
Mar 1, 2016
Messages
14,072
Likes
16,605
Location
Central Fl
I am struggling to see why this digression is in the AHB2 thread.
Someone left the door open at the WBF looney tunes room.
 

Wombat

Master Contributor
Joined
Nov 5, 2017
Messages
6,722
Likes
6,459
Location
Australia
I make no claims. I speak on behalf of countless engineering and researchers in audio. I am telling you that based on that collective knowledge, you have not provided anything of value for one to go and chase. People who make up the theories are responsible for providing such proof. Otherwise, they should not invent these theories.

The audiophiles who keep saying we don't know enough about audio, are so good at audio science/engineering as to keep inventing new theories. Which is it? If they don't understand it, they should stay out of making up stuff. And worse yet, asking us to go and investigate what they imagine without a shred of back of up, either in listening tests, or physics of what they say about audio cables.

Show me another example in your life where you dispute science this way. Do you make up your own theories of what is wrong with you task your doctor to prove them???




nq160120.gif
 

Wombat

Master Contributor
Joined
Nov 5, 2017
Messages
6,722
Likes
6,459
Location
Australia
DA is a problem for wideband and low-noise systems. AFAIK, not for cables conducting audio signals. My comments were based upon experience with things like radar, spread-spectrum com, satcom, etc. as well as Gb digital signals (e.g. SAS, Ethernet, etc.)

On-topic, it would be interesting to learn more about the technology in the AHB2, like the signal-tracking SMPS as well as the feedforward circuitry. I assume there is a patent, should quit debating cables and go look it up.


Don, Post #3 may provide a stepping-off point: https://www.harbeth.co.uk/usergroup...-a-new-amplifier-topology-ahb2-from-benchmark

General comment: Patent applicants walk a fine line in that enough information has to be provided for the application to succeed but, if possible, not to make life easy for IP pirates.
 
Last edited:
Top Bottom