• 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!

AHB2 Mono v Stereo - what is audible?

pjug

Major Contributor
Forum Donor
Joined
Feb 2, 2019
Messages
1,775
Likes
1,561
The impedance of the output stage and it's interaction with the load. You can feed a speaker 2.83V from a voltage source or develop 2.83V across that speaker from a current source. The speaker may sound and behave quite differently across its frequency range.

Imagine if I put a 10 ohm resistor in series with the speaker. And I turn up the amplifier, measure across the speaker 2.83V (1W@8R).

Then I disconnect the 10 ohm resistor, measure and level match across the speaker at 2.83V.

They will not sound the same with music.
Usually you make a lot of sense to me, but I don't follow at all. V is the independent variable and the job of a voltage amplifier is to make the right voltage regardless of load. If you want to add resistance in series with the speaker then of course the voltage at the speaker changes.
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,588
Likes
38,291
Location
Gold Coast, Queensland, Australia
It is an extreme example I used to show how level matching at a spot frequency and capturing a voltage based recording at the speaker terminals is flawed. It doesn't take into account how that speaker will sound when driven by the system as a whole, on musical content, and how two different amplifier configurations that produce the same spot voltage at one arbitary frequency at the speaker terminals, can vary in their overall acoustic response in that same room. A small change in FR could easily activate a room mode and make a large change to the perceived sound. That is a real audible difference.

This discussion is about a BTL amp vs the same amplifier in single ended 2 channel mode and any audible differences into a single pair of speakers in that room. Electrical measurements are absolutely irrelevant for audibility determination in this situation, except for initial level matching, and that needs to be done with access to the loudspeaker's impedance/reactance curve to pick a worst-case load frequency and match to that. And you know me, I love measurements. :)

I never closed my mind like so many others to the notion that amplifiers might sound different to one another, and also devised many experiments and configurations over the years to prove to myself that they actually sounded the same. Trouble is, they didn't.

And we have individuals stating amplifiers sound the same* as if it was some accepted science and completely beyond questioning. Where is the research? Seriously. I have books, magazines etc from the 50s onwards until now, across electronics, HiFi, music, you name it- where is the research? And when questioned, the first thing to do is poo-poo the attempt, but offer no research/experiments of their own.

The amplifiers are at the end of the chain. Everything else gets it easy. No large currents, high voltages or difficult load angles for the rest of the system. The rubber hits the road when the speaker makes acoustic output. And the amplifier does all the work. That interaction is the most complicated and yet is gets the least attention. Funny that.

* when operated at a similar output approaching, but below the threshold of clipping and without gross flaws in FR or audible levels of THD.
 
Last edited:

pjug

Major Contributor
Forum Donor
Joined
Feb 2, 2019
Messages
1,775
Likes
1,561
It is an extreme example I used to show how level matching at a spot frequency and capturing a voltage based recording at the speaker terminals is flawed. It doesn't take into account how that speaker will sound when driven by the system as a whole, on musical content, and how two different amplifier configurations that produce the same spot voltage at one arbitary frequency at the speaker terminals, can vary in their overall acoustic response in that same room. A small change in FR could easily activate a room mode and make a large change to the perceived sound. That is a real audible difference.

This discussion is about a BTL amp vs the same amplifier in single ended 2 channel mode and any audible differences into a single pair of speakers in that room. Electrical measurements are absolutely irrelevant for audibility determination in this situation, except for initial level matching, and that needs to be done with access to the loudspeaker's impedance/reactance curve to pick a worst-case load frequency and match to that. And you know me, I love measurements. :)

I never closed my mind like so many others to the notion that amplifiers might sound different to one another, and also devised many experiments and configurations over the years to prove to myself that they actually sounded the same. Trouble is, they didn't.

And we have individuals stating amplifiers sound the same* as if it was some accepted science and completely beyond questioning. Where is the research? Seriously. I have books, magazines etc from the 50s onwards until now, across electronics, HiFi, music, you name it- where is the research? And when questioned, the first thing to do is poo-poo the attempt, but offer no reseach/experiments of their own.

The amplifiers are at the end of the chain. Everything else gets it easy. No large currents, high voltages or difficult load angles for the rest of the system. The rubber hits the road when the speaker makes acoustic output. And the amplifier does all the work. That interaction is the most complicated and yet is gets the least attention. Funny that.

* when operated at a similar output approaching, but below the threshold of clipping and without gross flaws in FR or audible levels of THD.
System as a whole? It's a simple amplification of V, and then what flows into the load follows. I had the independent and dependent variables drilled into my head in school, do I need to go back? I am not closing my mind about amps sounding different; I just want to get at it properly. Such a simple thing to do to measure voltage compared to using microphone. Just get the output voltage matched reasonably before recording and then trimming can be done after recording.
 
OP
Coach_Kaarlo

Coach_Kaarlo

Active Member
Joined
Apr 27, 2020
Messages
196
Likes
222
Location
Sydney
@Coach_Kaarlo Yes, it's typical behaviour isn't it? People making excuses about this and that, analyzing the files to 'look' for differences instead of putting their audiophile balls on the line and ABXing the files first.

It's not hard. I did it, heard a difference and focused on it- my results show once I heard the difference, it was easy. That was one test, no peeking at files or cheating. Just trusted my ears and what I know about BTL amplifiers and how stereo amplifiers change their sound when you bridge them into a monoblock.

People seem to think their season ticket to the Audiophile playground will get revoked if they publicly admit something not popular, or they fail some "test". It's hilarious.

Yep, reasons / excuses abound, but miss the actual point - which was the point actually.....
 
OP
Coach_Kaarlo

Coach_Kaarlo

Active Member
Joined
Apr 27, 2020
Messages
196
Likes
222
Location
Sydney
And for the record the recordings were outside which if half the respondents had bothered to read the prior posts would have been clear.
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,386
Likes
24,752
Location
Alfred, NY
The impedance of the output stage and it's interaction with the load. You can feed a speaker 2.83V from a voltage source or develop 2.83V across that speaker from a current source. The speaker may sound and behave quite differently across its frequency range.

Imagine if I put a 10 ohm resistor in series with the speaker. And I turn up the amplifier, measure across the speaker 2.83V (1W@8R).

Then I disconnect the 10 ohm resistor, measure and level match across the speaker at 2.83V.

They will not sound the same with music.
In those cases, the recorded electrical signals will be different. And if audibly different, you will have established that the amplifiers sound different.

If the electrical signal at the speaker terminals is essentially the same, the sound will be the same. If it's different, the sound MAY be different, regardless of the mechanism of the difference.
 

pjug

Major Contributor
Forum Donor
Joined
Feb 2, 2019
Messages
1,775
Likes
1,561
Yep, reasons / excuses abound, but miss the actual point - which was the point actually.....
So you don't want to record voltage? I just used a voltage divider in an old sub amp and got a recording of speaker voltage going in a few minutes. This was a 1000R/100R divider in the plate amp. I would rather use something like 100R/10R, or maybe lower still, but it is very simple.
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,588
Likes
38,291
Location
Gold Coast, Queensland, Australia
In those cases, the recorded electrical signals will be different. And if audibly different, you will have established that the amplifiers sound different.

If the electrical signal at the speaker terminals is essentially the same, the sound will be the same. If it's different, the sound MAY be different, regardless of the mechanism of the difference.

Recorded differences in electrical signals and audible differences are not interchangeable. One does not assume the other, nor vice-versa.

Surely you must accept that matching levels at an arbitary frequency, electrically, at the speaker terminals as suggested and espoused by many (including yourself?) does not ensure the loudspeaker, nor the ultimate response, to be matched, nor behave the same, in terms of acoustic output across the frequency spectrum when reproducing music or any other waveforms, when driven by different amplifier configurations.

How do you realistically level match a pair of amplifier configurations and a loudspeaker, when both the amplifiers and the loudspeaker are reactive and non-linear sources/loads to one another? 1kHz, measured on a DMM/ACmV meter/AP at the speaker terminals is not the answer is it? For one single spot frequency and power level, maybe. For everything else, not so much.
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,835
Likes
16,498
Location
Monument, CO
Recorded differences in electrical signals and audible differences are not interchangeable. One does not assume the other, nor vice-versa.

Surely you must accept that matching levels at an arbitary frequency, electrically, at the speaker terminals as suggested and espoused by many (including yourself?) does not ensure the loudspeaker, nor the ultimate response, to be matched, nor behave the same, in terms of acoustic output across the frequency spectrum when reproducing music or any other waveforms, when driven by different amplifier configurations.

How do you realistically level match a pair of amplifier configurations and a loudspeaker, when both the amplifiers and the loudspeaker are reactive and non-linear sources/loads to one another? 1kHz, measured on a DMM/ACmV meter/AP at the speaker terminals is not the answer is it? For one single spot frequency and power level, maybe. For everything else, not so much.

I think, and believe @SIY is saying, that if you measure the voltage (magnitude and phase) over frequency (i.e. do a full sweep), then the voltage plotted at the speaker terminals will be quite different for the two cases (amp direct vs. amp + 10-ohm or whatever resistor). I completely agree a single-point measurement may not be revealing. And speaker level matching is, uh, challenging at best. But again if you match the voltage across the terminals at one frequency, then do a sweep or play music and the sound is different, then the voltage at different frequencies should also be different.

Bridging doubles'ish the output impedance of an amp so I'd expect some variance with audibility depending upon the speaker load impedance (etc.)

Ignore if does not make sense, I am drooling over that old N77ES!
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,386
Likes
24,752
Location
Alfred, NY
Recorded differences in electrical signals and audible differences are not interchangeable. One does not assume the other, nor vice-versa.

Surely you must accept that matching levels at an arbitary frequency, electrically, at the speaker terminals as suggested and espoused by many (including yourself?) does not ensure the loudspeaker, nor the ultimate response, to be matched, nor behave the same, in terms of acoustic output across the frequency spectrum when reproducing music or any other waveforms, when driven by different amplifier configurations.

How do you realistically level match a pair of amplifier configurations and a loudspeaker, when both the amplifiers and the loudspeaker are reactive and non-linear sources/loads to one another? 1kHz, measured on a DMM/ACmV meter/AP at the speaker terminals is not the answer is it? For one single spot frequency and power level, maybe. For everything else, not so much.
Pink noise is a much better signal for that.
 

RichB

Major Contributor
Forum Donor
Joined
May 24, 2019
Messages
1,946
Likes
2,611
Location
Massachusetts
Recorded differences in electrical signals and audible differences are not interchangeable. One does not assume the other, nor vice-versa.

Surely you must accept that matching levels at an arbitary frequency, electrically, at the speaker terminals as suggested and espoused by many (including yourself?) does not ensure the loudspeaker, nor the ultimate response, to be matched, nor behave the same, in terms of acoustic output across the frequency spectrum when reproducing music or any other waveforms, when driven by different amplifier configurations.

How do you realistically level match a pair of amplifier configurations and a loudspeaker, when both the amplifiers and the loudspeaker are reactive and non-linear sources/loads to one another? 1kHz, measured on a DMM/ACmV meter/AP at the speaker terminals is not the answer is it? For one single spot frequency and power level, maybe. For everything else, not so much.

I have been level matching at full-scale 1kHz tone to level match amplifiers.
Using quick-switching, there were identifiable difference between the AHB2 and AT522NC mostly at high frequencies.
The natural conclusion is the amplifiers are not behaving the same with music across all frequency ranges into a reactive load.

If one is predisposed to believe all amps sound the same, this is upsetting. I posted the results on AVNivana and ran into such an individual.
The discussion even turned to be challenged that .1 dB matching was unnecessary, according to Toole. Apparently, the levels were overly matched.
Sometimes, the point is the argument :p

- Rich
 

pjug

Major Contributor
Forum Donor
Joined
Feb 2, 2019
Messages
1,775
Likes
1,561
Recorded differences in electrical signals and audible differences are not interchangeable. One does not assume the other, nor vice-versa.

Surely you must accept that matching levels at an arbitary frequency, electrically, at the speaker terminals as suggested and espoused by many (including yourself?) does not ensure the loudspeaker, nor the ultimate response, to be matched, nor behave the same, in terms of acoustic output across the frequency spectrum when reproducing music or any other waveforms, when driven by different amplifier configurations.

How do you realistically level match a pair of amplifier configurations and a loudspeaker, when both the amplifiers and the loudspeaker are reactive and non-linear sources/loads to one another? 1kHz, measured on a DMM/ACmV meter/AP at the speaker terminals is not the answer is it? For one single spot frequency and power level, maybe. For everything else, not so much.
I dont see why level matching of the voltage has to be perfect. The amplifier behavior should be captured if they are just somewhat close. Then trim the recording afterwards if you want to see if differences are audible.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,524
Likes
37,057
It is an extreme example I used to show how level matching at a spot frequency and capturing a voltage based recording at the speaker terminals is flawed. It doesn't take into account how that speaker will sound when driven by the system as a whole, on musical content, and how two different amplifier configurations that produce the same spot voltage at one arbitary frequency at the speaker terminals, can vary in their overall acoustic response in that same room. A small change in FR could easily activate a room mode and make a large change to the perceived sound. That is a real audible difference.

This discussion is about a BTL amp vs the same amplifier in single ended 2 channel mode and any audible differences into a single pair of speakers in that room. Electrical measurements are absolutely irrelevant for audibility determination in this situation, except for initial level matching, and that needs to be done with access to the loudspeaker's impedance/reactance curve to pick a worst-case load frequency and match to that. And you know me, I love measurements. :)

I never closed my mind like so many others to the notion that amplifiers might sound different to one another, and also devised many experiments and configurations over the years to prove to myself that they actually sounded the same. Trouble is, they didn't.

And we have individuals stating amplifiers sound the same* as if it was some accepted science and completely beyond questioning. Where is the research? Seriously. I have books, magazines etc from the 50s onwards until now, across electronics, HiFi, music, you name it- where is the research? And when questioned, the first thing to do is poo-poo the attempt, but offer no reseach/experiments of their own.

The amplifiers are at the end of the chain. Everything else gets it easy. No large currents, high voltages or difficult load angles for the rest of the system. The rubber hits the road when the speaker makes acoustic output. And the amplifier does all the work. That interaction is the most complicated and yet is gets the least attention. Funny that.

* when operated at a similar output approaching, but below the threshold of clipping and without gross flaws in FR or audible levels of THD.
I very much agree amplifiers are not all the same into speaker loads. I'm not sure I understand the objection to matching at 1 khz before auditioning for comparison purposes. Yes, the load is reactive. Yes the output impedance or limited current interactions might alter the speaker response an audible amount. In such cases level cannot be fully matched across the whole spectrum, but that is why it will sound different.

I have some Soundlab ESL's which feature very high impedance down low, and something a touch under 1 ohm at 20 khz (almost wholly capacitive). It is an extreme example of reactive loading. Anything with much output impedance will alter response enough to hear. Quite a few amps create a slight resonant peak between 20-25 khz. That is a low enough Q it elevates the upper audible octave enough to hear with some and not with others. I don't know if I've ever heard any two amps sound the same. Yet it is worthwhile to match levels at the terminals. The differences have always been smaller than expected if you do this matching.

The other thing is I wonder if there are any conditions where capturing the signal at the speaker terminals would fail to capture differences in the speaker output. I'm not sure there are any, but maybe I'm not thinking of the right thing. With a reactive load voltage capture won't show how wildly current flow might be changing, but if the amp isn't current starved have you missed anything? If it is current starved, the voltage will be altered too. Am I missing something in regard to this?
 

pjug

Major Contributor
Forum Donor
Joined
Feb 2, 2019
Messages
1,775
Likes
1,561
The difference in frequency response in the AHB2 mono vs stereo should be somewhat predictible, right? It would swing twice as much from flat when bridged, compared to what is shown in the grey curve, right? Speaker dependent, of course. [https://www.stereophile.com/content/benchmark-media-systems-ahb2-power-amplifier-measurements]
1623879406811.png

Maybe another way to get at this is apply filters to the source file, simulating the grey curve and grey curve X2, and check whether you can hear the difference.
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,588
Likes
38,291
Location
Gold Coast, Queensland, Australia
I'm not sure I understand the objection to matching at 1 khz before auditioning for comparison purposes.

It's not an objection at all. I am happy to do it and did so for years, but depending on the speaker, it can serve to highlight differences that may be less obvious than if you match at between 400-600Hz, where the nominal low point impedance often lies.

I've always looked to make the amplifier comparisons as fair as possible as my goal was to prove they actually sounded the same, not different.
 
Last edited:

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,335
Likes
6,700
@Coach_Kaarlo Yes, it's typical behaviour isn't it? People making excuses about this and that, analyzing the files to 'look' for differences instead of putting their audiophile balls on the line and ABXing the files first.

I heard(felt) a difference, and it was obvious. No peeking. I'm not sure how to set up the A/B but if you show me how, I bet I can pass.

Like I said, I don't really hear a huge difference, but at the right volume it's obvious because file 2 has a ULF(starting in between 3 and 4 seconds) beat that vibrates the body, whereas file 1 is just audible bass.
 
Last edited:

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,524
Likes
37,057
It's not an objection at all. I am happy to do it and did so for years, but depending on the speaker, it can serve to highlight differences that may be less obvious than if you match at between 400-600Hz, where the nominal low point impedance often lies.

I've always looked to make the amplifier comparisons as fair as possible as my goal was to prove they actually sounded the same, not different.
Ok I sometimes use 440 hz for matching.
 
OP
Coach_Kaarlo

Coach_Kaarlo

Active Member
Joined
Apr 27, 2020
Messages
196
Likes
222
Location
Sydney
Do you have neighbors? Maybe one of your neighbors was starting his rotary subwoofer demo(saw that in a youtube video yesterday) in between the 3 and 4 second mark of the second recording?

Airport. Jets have some low frequency rumbles for sure - I see them in my measurements far more often than I want. However, I am fairly sure during the 2 recordings such ambient noises were not audible. I actually made about 6 recordings and selected the cleanest 2.....
 
OP
Coach_Kaarlo

Coach_Kaarlo

Active Member
Joined
Apr 27, 2020
Messages
196
Likes
222
Location
Sydney
If I understand the theme of level matching correctly, it seems to be to find a linear way of trying to match two non-linear responses; bridged versus stereo amplification into a speaker.

Voltage AND current delivery can change the behaviour of the speaker. Measuring voltage at one nominal frequency only does that - it leaves everything else unmatched essentially (to my thinking).

My idea with using pink noise was that I would be averaging any differences (voltage, curent) at various frequencies, and getting closer to like for like AUDIBLY.

And yes, I failed to accurately match the levels - part of doing so many recordings I suspect. But FOOBAR matches the levels, and like I stated previously, if you do the true blind test with the 2 tracks and can hear what the difference is, and post your results - we have a starting point.

It might be possible that after 100 people have tried this, the actual current 100% success rate might drop somewhat.....
 
Top Bottom