Can you hear a difference in Sound between Audio Amplifiers?

[fmw]

A dozen years ago we did some properly administered bias controlled tests with a panel of 10 audiophiles. We tested about a dozen amps ranging from unbelievably expensive (Mark Levinson and Krell) to run of the mill (Onkyo two channel receiver.) Only one produced sonics that differed from the others. It was a single ended triode amp with about 7% measured harmonic distortion. All the amps with flat frequency responses and inaudible distortion sounded the same. Even one of my expensive amps (a tube amp made by Audio Research) was indistinguishable from the solid state amps. That was impressive to me. The minute you allow bias to enter a listening test, you will get an amazing range of results. People are biased and bias certainly affects the way we hear - or better, the way we interpret what we hear. Needless to say, I don't spend the kind of money I used to on amplifiers. Audio Research is a distant memory.

 

[jinjuku]

Well, you should at least spell his name right in the article: it's Tom Nousaine, (as mtrycraft wrote), not 'Nosaine'.

Also you might want to cite/discuss another famous article on 'amp sound', by E. Brad Meyer

The Amp/Speaker Interface: Are Your Loudspeakers Turning Your Amplifier into a Tone Control?" E. Brad Meyer, Stereo Review, June 1991, page 54,

The Amp/Speaker Interface | Sound and Vision Magazine

My take: if one or both of the amps in your comparison is driven to clipping, or if the test is not double-blinded (either literally, or effectively via ABX switching) , randomized, level-matched to within 0.2 dB (at, say, 1 kHz at least), or lacks sufficient trials (16 would be a good target) then the test is seriously flawed. Ideally also the subjects should be trained with a series of differences that go from large to small.

The link in the S&V article for the PDF doesn't work...

 

[krabapple]

The link in the S&V article for the PDF doesn't work...

That's a pity. It's well worth reading.

 

[AcuDefTechGuy]

A dozen years ago we did some properly administered bias controlled tests with a panel of 10 audiophiles. We tested about a dozen amps ranging from unbelievably expensive (Mark Levinson and Krell) to run of the mill (Onkyo two channel receiver.) Only one produced sonics that differed from the others. It was a single ended triode amp with about 7% measured harmonic distortion. All the amps with flat frequency responses and inaudible distortion sounded the same. Even one of my expensive amps (a tube amp made by Audio Research) was indistinguishable from the solid state amps. That was impressive to me. The minute you allow bias to enter a listening test, you will get an amazing range of results. People are biased and bias certainly affects the way we hear - or better, the way we interpret what we hear. Needless to say, I don't spend the kind of money I used to on amplifiers. Audio Research is a distant memory.

Bias is significant with everything including amps, preamps, speakers, etc.

But some people believe they are immune to some bias, especially if they feel that they are experts.

 

[RichB]

Bias is significant with everything including amps, preamps, speakers, etc.

But some people believe they are immune to some bias, especially if they feel that they are experts.

Oh crap, does this mean i have to get rid of my bumper sticker?

My Bias is Class-A



- Rich

 

[fmw]

Bias is significant with everything including amps, preamps, speakers, etc.

But some people believe they are immune to some bias, especially if they feel that they are experts.

No doubt about that. But none of us are immune. Even after all those tests I have a healthy cynicism about the sound of HiFi components but I still have all the bias like everyone else.

 

[ronwande]

Differences Can Be Measured

I am a retired BSEE who worked in circuit design for some 43 years though not in audio as such. I don't believe in magic. I do believe that if two amplifiers sound different there must be some difference in the signals applied to the speakers. (The alternative to that is that the amplifiers are sending us signals telepathically). After reading this article I think I understand much better, that perhaps those who claim no difference in amplifier sounds have oversimplified things. We are not measuring the right parameters!

A friend had me design and build a stereo amplifier with single ended vacuum tube triode output stage. I had built a very low distortion solid state amplifier based on the three stage Lin architecture. We compared the sound in an A/B listening test with a switch that switched the speakers and a dummy load alternately to the two amplifier outputs. After matching the levels and being sure we were not driving the low powered SET amplifier into clipping, we clearly heard a difference between the two. The tube amplifier sounded brighter, particularly on passages with brass instruments such as trumpets.

I had previously built a variable second harmonic generator using an analog multiplier chip. I inserted that between the preamp and my solid state amplifier and again carefully matched the output levels of the two. Then I tweaked the second harmonic level a few times and to me the two amplifiers sounded identical! I don't pretend to have a golden ear and my friend has a hearing aid and can hardly hear a violin solo in the high audio range, BUT I clearly heard the second harmonic present in the tube amplifier as adding an "edge" to the sound. Doug Self advocates calling a second harmonic processor a "niceness control".

As an experiment, I set up my amplifier with a 500 Hz sine wave input from a low distortion signal generator, the output of an HP 339A distortion test set, close to 0.001% THD. I added the distortion generator and adjusted until I could just hear the second harmonic. The distortion analyzer said 0.08%. I am positive that I couldn't hear that level of distortion in music.

I have been interested in audio since my college days, having built several vacuum tube amplifiers back then. Thank you for this interesting article. I have another anecdotal story that might be of interest but I'll save it for another time. Meanwhile I'll read the whole thread.

 

[3db]

I am a retired BSEE who worked in circuit design for some 43 years though not in audio as such. I don't believe in magic. I do believe that if two amplifiers sound different there must be some difference in the signals applied to the speakers. (The alternative to that is that the amplifiers are sending us signals telepathically). After reading this article I think I understand much better, that perhaps those who claim no difference in amplifier sounds have oversimplified things. We are not measuring the right parameters!

A friend had me design and build a stereo amplifier with single ended vacuum tube triode output stage. I had built a very low distortion solid state amplifier based on the three stage Lin architecture. We compared the sound in an A/B listening test with a switch that switched the speakers and a dummy load alternately to the two amplifier outputs. After matching the levels and being sure we were not driving the low powered SET amplifier into clipping, we clearly heard a difference between the two. The tube amplifier sounded brighter, particularly on passages with brass instruments such as trumpets.

I had previously built a variable second harmonic generator using an analog multiplier chip. I inserted that between the preamp and my solid state amplifier and again carefully matched the output levels of the two. Then I tweaked the second harmonic level a few times and to me the two amplifiers sounded identical! I don't pretend to have a golden ear and my friend has a hearing aid and can hardly hear a violin solo in the high audio range, BUT I clearly heard the second harmonic present in the tube amplifier as adding an "edge" to the sound. Doug Self advocates calling a second harmonic processor a "niceness control".

As an experiment, I set up my amplifier with a 500 Hz sine wave input from a low distortion signal generator, the output of an HP 339A distortion test set, close to 0.001% THD. I added the distortion generator and adjusted until I could just hear the second harmonic. The distortion analyzer said 0.08%. I am positive that I couldn't hear that level of distortion in music.

I have been interested in audio since my college days, having built several vacuum tube amplifiers back then. Thank you for this interesting article. I have another anecdotal story that might be of interest but I'll save it for another time. Meanwhile I'll read the whole thread.

You experienced a difference in soudn between a tube amp and a SS amp? Am I reading this properly?

 

[jinjuku]

Also the difference was measurable to the point of knowing how and reproducing it. Certainly along my opinion of amps can be made to sound different.

 

[fmw]

Normally, when we talk about amplifiers sounding the same, we are referring to solid state amplifiers with flat frequency response and inaudible distortion. Those amps will sound the same as long as they aren't overdriven. Tube amps are a somewhat different issue. I once had a high end tube amp made by Audio Research that had inaudible distortion like the amp Ronwande discusses above. It fell right in line with the inaudible solid state amps in our bias controlled tests. Nobody could tell it from a solid state amp. We also tested a single ended triode that measured over 5% harmonic distortion. It was overwhelmingly identified by the listening panel. We also tested a very old transformerless tube amp from the 1960's that was also identified enough to call it different sounding. But the goal, is not to do that. What we should want in an amplifier is transparency. We should want it to enlarge the waveforms without changing their shape. The great majority of amplifiers do that at least at the level of audibility. Those that do not are either incompetently designed or designed to have some specific sonic signature.

 

[panteragstk]

Those that do not are either incompetently designed or designed to have some specific sonic signature.

Interesting statement. I would think an amp designed to have a "sonic signature" would be undesirable. You said above exactly what I would want an amp to do: produce a clean waveform and be as transparent as possible. Intentionally altering the way an amp sounds would seem like a bad practice. Unless of course you are referring to guitar amps.

 

[Steve81]

Interesting statement. I would think an amp designed to have a "sonic signature" would be undesirable.... Intentionally altering the way an amp sounds would seem like a bad practice.

I dunno. Michael Fremer really seems to have enjoyed these bad boys:
Wavac SH-833 monoblock power amplifier | Stereophile.com

I found reading the review and JA's commentary in the measurements section to be mildly entertaining

 

[3db]

I dunno. Michael Fremer really seems to have enjoyed these bad boys:
Wavac SH-833 monoblock power amplifier | Stereophile.com

I found reading the review and JA's commentary in the measurements section to be mildly entertaining

All that means is either Mr Fermer is impressed with the cost of these amps or the tuning of sound may fit his ears the best and others may be turned off by them. if other could get past the cost of these.

 

[Steve81]

if other could get past the cost of these.

Or the size You'd have to pay me to let those things into my home...

 

[fmw]

Interesting statement. I would think an amp designed to have a "sonic signature" would be undesirable. You said above exactly what I would want an amp to do: produce a clean waveform and be as transparent as possible. Intentionally altering the way an amp sounds would seem like a bad practice. Unless of course you are referring to guitar amps.

I couldn't agree more. I think it's OK for a listener to adjust sonics using things like room treatments or equalizers. That's just a matter of preference. But having those sonic signatures built in are a bad idea in my view as well.

 

[AcuDefTechGuy]

I couldn't agree more. I think it's OK for a listener to adjust sonics using things like room treatments or equalizers. That's just a matter of preference. But having those sonic signatures built in are a bad idea in my view as well.

People will do all kinds of things to sell whether it is arguably good or bad, don't they?

Some people just need another excuse to buy.

 

[TigerFlash18]

Well guys...I did it...I read all 60 pages of this thread. And it prompted me to jump on and make an account here.

Some quick background. I am very slowly building a home theater in one of my guest bedrooms, so far I've just been listening to stereo music in direct mode. I am kind of awaiting a tv console and a sofa before I set the other speakers up.

Anyways, I took EE as an undergrad/grad, and now work as a computer engineer. Before I came to this thread my biggest opinion regarding amplifiers was that they were simple, all they do is take a signal from the preamp and make it bigger. How in the world can they possibly sound distinct unless they are doing something wrong... Well, after reading 60 pages of this thread, my opinion has not changed Haha..I'm ever so glad that I didn't dump heaps of money into my amp, and instead am looking at accoustic panels to make my room better. For those that are wondering I got the emotiva XPA 5, and couldn't be happier. I even stuck a current meter up to the thing and there is just no way I can reach the volume level it is capable of without destroying my hearing.

Cheers!

 

[PENG]

I even stuck a current meter up to the thing and there is just no way I can reach the volume level it is capable of without destroying my hearing.

Cheers!

Glad to know I wasn't the only crazy one who did such thing. Actually I've had 3 meters on them once or twice, to read both V and I at the same time. We might have missed a few dB of instantaneous peaks due to the meter's (depend on the type) response time but wouldn't be by much. Again, the question asked in this thread is so subject to interpretation that there could be no right or wrong answers.

 

[lvb1770]

I'm surprised to read a thread at this website asking me what I hear. Usually what one hears, if so expressed in one of your forums, is treated as a hallucination on the hearer's part; unless backed up by rigid double blind testing. So I am in agreement that the question is not specific enough, unless the intent of the OP was to ignite a firestorm. I will say that I, if you forgive me, heard less of a difference when I upgraded a $300 receiver with a $1500 power amp than I did when switching from lamp cord to 12 gauge speaker wire.

 

[mtrycrafts]

Well guys...I did it...I read all 60 pages of this thread. And it prompted me to jump on and make an account here.


Cheers!

Must have missed your post . Welcome What a trooper, 60 pages. I think you will like it here.