Negative feedback bad for audio?

[SIY]

I really don't know why you suppose that Pass can't tell the sound differences among the the many, many different designs he has developed.

And yet he has never demonstrated such. Nor is he ever likely to.

 

[Theriverlethe]

I really don't know why you suppose that Pass can't tell the sound differences among the the many, many different designs he has developed.

Take 3-4 minutes and read my recent amp review here ... https://www.audioasylum.com/forums/amp/messages/23/235714.html ... you will, of course, scoff at my evaluation method and conclusions.

Probably because it contradicts a century of psychoacoustic research, doesn’t show up on measurement equipment far more sensitive than human hearing, and has never been demonstrated scientifically.

 

[maty]

If good sound quality is already available in the room, there is a very easy subjective test. Play recordings that get us excited, make us believe that we are in the concert hall or that the musicians are in ours. If the new amplifier, with spectacular measurements, is not capable of generating those emotions, then it is not what we need.

If we already had 3D sound, with sound layers, check if they increase.

This will be much easier with acoustic and electrical instrumentation, with unadulterated voices. If everything is artificial, synthetic then it is another story, and it is not worth spending a lot of money on the equipment, either.

 

[Feanor]

When claiming something, we usually resort to sticking to the null hypothesis (the one who does not claim any change) and test if there is a change. Until Pass proves himself in a third-party, scrutinized and repeatable test with enough positive evidence that he can hear the difference, we believe he cannot detect the differences.

... Or you could actually listen to a Pass amp or two for a few hours.

 

[KeithPhantom]

... Or you could actually listen to a Pass amp or two for a few hours.

Are my ears better than an AP analyzer? I don't think so (even though I'm able to hear 20 kHz {I'm only 21}). Also, why do you shift the burden of proof onto me? Whomever is making the positive claim (Pass can hear a difference {the alternative hypothesis}) must prove their hypothesis correct.

Why is it so hard for you to accept that humans have limits? Our hearing isn't better than machines at judging objective qualities you can measure in equipment, were pretty bad assessing multiple types of distortion and noise, something a cheap oscilloscope can do orders of magnitude better than us. There is personal preference, but the deltas needed to be able to clearly distinguish between devices have to be huge (and for electronic gear, they have to be fundamentally broken compared to the standard). There's no magical properties in sound that cannot be measured, we can measure all of the physical properties of audio. We can measure fidelity to the source.

 

[Feanor]

Probably because it contradicts a century of psychoacoustic research, doesn’t show up on measurement equipment far more sensitive than human hearing, and has never been demonstrated scientifically.

... So this sounds like some medical doctors I've heard of. Describe your symptoms and if s/he does doesn't recognize the cause or course of treatment, s/he tells you it's psychosomatic or simply you imagination.

Actually some measurements show up differences that correlated to human preferences. 2nd & 3rd harmonic distortions are pleasant sound whereas harmonics of 4th order and higher are increasingly discordant and unpleasant. Designers such as Pass and Curl, (so disdained hereabouts), know this and have designed equipment accordingly; (see Amir's test of the Parasound JC2 preamp). On the audiophile extreme are SET lovers. These tube amps produce copious amounts of 2nd order distortion in particular and very little higher order distortion. SET's aren't to my taste, but to me that only proves the relevance of subjectivity.

 

[Feanor]

Are my ears better than an AP analyzer? I don't think so (even though I'm able to hear 20 kHz {I'm only 21}). Also, why do you shift the burden of proof onto me? Whomever is making the positive claim (Pass can hear a difference {the alternative hypothesis}) must prove their hypothesis correct.

Why is it so hard for you to accept that humans have limits? Our hearing isn't better than machines at judging objective qualities you can measure in equipment, were pretty bad assessing multiple types of distortion and noise, something a cheap oscilloscope can do orders of magnitude better than us. There is personal preference, but the deltas needed to be able to clearly distinguish between devices have to be huge (and for electronic gear, they have to be fundamentally broken compared to the standard). There's no magical properties in sound that cannot be measured, we can measure all of the physical properties of audio. We can measure fidelity to the source.

Oh no, sir, I don't claim my ears are better than an AP analyser. In fact I'm pretty much stone deaf above 10 kHz. But whereas I might have bit of a hearing deficit, I don't have a listening deficit.

The argument that reasonably designed, properly functioning amplifiers of adequate power all sound the same has been around for many decades. As a 50 hi-fi veteran, to me it's very "old" in the sense of archaic.

 

[KeithPhantom]

Actually some measurements show up differences that correlated to human preferences. 2nd & 3rd harmonic distortions are pleasant sound whereas harmonics of 4th order and higher are increasingly discordant and unpleasant. Designers such as Pass and Curl, (so disdained hereabouts), know this and have designed equipment accordingly; (see Amir's test of the Parasound JC2 preamp). On the audiophile extreme are SET lovers. These tube amps produce copious amounts of 2nd order distortion in particular and very little higher order distortion. SET's aren't to my taste, but to me that only proves the relevance of subjectivity.

So why use an amp to add distortion? I can build something that can add those kinds of distortion and make the adjustable, heck, they even are used in music mixing as plugins. You can control them as you like, and you don't need to have the fixed amount a designer allows you to have. You can have this for far cheaper than any of the "audiophile" amps, and you can set them at your pleasure, not trusting in what this guy thinks sounds good. By the way, you can measure all of that as well.

 

[AudioStudies]

I know someone who has designed tube amps for decades, and he won't touch OTL or the use of SETs. Thus, even many tube designers shy away from those things because of the flaws.

 

[thefsb]

So why use an amp to add distortion? I can build something that can add those kinds of distortion and make the adjustable, heck, they even are used in music mixing as plugins. You can control them as you like, and you don't need to have the fixed amount a designer allows you to have.

Right! You're talking about Guitar amps. That's what tubes are for! And it's where distortion is valued.

You can have this for far cheaper than any of the "audiophile" amps,

Well, depends on the amp. Some aren't exactly inexpensive.

Personally, I do not use that kind of tech. I use digital modeling and speaker cabinet impulse responses from York Audio with PA amps (which are usually calss D) and cabs.

 

[SIY]

Oh no, sir, I don't claim my ears are better than an AP analyser. In fact I'm pretty much stone deaf above 10 kHz. But whereas I might have bit of a hearing deficit, I don't have a listening deficit.

The argument that reasonably designed, properly functioning amplifiers of adequate power all sound the same has been around for many decades. As a 50 hi-fi veteran, to me it's very "old" in the sense of archaic.

Great, looking forward for you to finally be the one to offer counter-evidence.

 

[KeithPhantom]

Right! You're talking about Guitar amps. That's what tubes are for! And it's where distortion is valued.

Do you strictly need a tube amp to add distortion? I bet software (in combination with hardware) can do what tubes can while tubes cannot do what software can. VST plugins can (at least possible in theory, just a need of a correct implementation) emulate all classic and modern guitar amps you can think of with better accuracy than getting the real deal. You don't need hardware to add distortion, software can do it and have more flexibility.

 

[Tom C]

Culled from around the web. Sorry it’s sideways. I can’t seem to correct it. According to the technical notes, the Western Electric 91-B amplifier became available in June of 1936. The design is quintessential SET, using a single 300A tube as the output stage. The other uncovered tube in the photo is the rectifier for the power supply. Two shielded, high-gain pentodes are visible on the left side of the chassis. While the 300A is no longer manufactured, the 300B is, and the only difference between the two tubes is the configuration of the base, and the pin assignments.
As can be seen from the schematic, negative feedback was employed from the plate of the output to the grid of the input, as it was discovered to improve performance.

 

[NTK]

I just don't get this thing about adding second harmonic (H2) distortions will make the reproduced sound sounding more pleasant. How come nobody says anything about the accompanying euphonic intermodulation distortions (IMD)?

Mathematically, a simple harmonic generator is the Chebyshev polynomial of the first kind (as used by some harmonics generators). If you want to generate H2, run the signal through the 2nd order Chebyshev polynomial function and out comes the H2 with the same amplitude.

So, if you want to add 1% H2 distortion, simply calculates: f_new(t) = f(t) + 0.01*T_2(f(t)), where f(t) is the original signal and T_2(x) is the 2nd order Chebyshev polynomial function.

Below are three examples. The top graph is a pure 250 Hz tone plus 1% H2 (see the 500 Hz peak with a -40 dB lower peak height). The middle graph is the same for 450 Hz. In the bottom graph, the input is the sum of the 250 Hz and 450 Hz tones. We can see the exact same H2 peaks, plus two new intermodulation peaks at 700 Hz (= 450 + 250 Hz) and 200 Hz (= 450 - 250 Hz). What is interesting is that these non-harmonic IMD have higher amplitudes than the H2. What are the effects of these non-harmonic distortions? Are they also "euphonic"? Since music typically has more than one fundamental frequencies, there will be an abundance of these IMD. I have not seen any talks about the benefits of these IMD, only H2 and maybe H3. We know if there are H2 and H3 distortions, the IMD will be there too. Why only tout H2 and H3, and not IMD?

 

[thefsb]

Do you strictly need a tube amp to add distortion? I bet software (in combination with hardware) can do what tubes can while tubes cannot do what software can. VST plugins can (at least possible in theory, just a need of a correct implementation) emulate all classic and modern guitar amps you can think of with better accuracy than getting the real deal. You don't need hardware to add distortion, software can do it and have more flexibility.

No, ofc you don't strictly need a tube amp to add distortion. For example, I use digital modeling in a pedalboard. For VST (and similar) plugins Amplitube and Bias Amp are popular. And there are scads of hardware devices that aim to achieve similar effects.

Nevertheless, for a guitarist accustomed to using the classic arrangement of passive pickups into a tube amp (with its delicate RLC input circuit and controls), the whole thing is a musical instrument with a very particular interactivity. While I prefer to avoid this kind of thing, I can understand why it's still a very popular technology. And the digital tech isn't necessarily cheaper: there are plenty of nice amps available for less than the price of a Kemper Profiler.

 

[Mulder]

...
Actually some measurements show up differences that correlated to human preferences. 2nd & 3rd harmonic distortions are pleasant sound whereas harmonics of 4th order and higher are increasingly discordant and unpleasant. Designers such as Pass and Curl, (so disdained hereabouts), know this and have designed equipment accordingly;

No, Nelson Pass don´t say that.

"Audiophiles have been accused of using 2nd or 3rd harmonic distortion as tone controls to deliberately alter the sound. I suppose that there are people who like it that way, but I don't think this is generally the case. For reasons which will become clearer when we talk about inter-modulation distortion, high levels of any harmonic become problematic with musical material having multiple instruments, and the argument that 2nd or 3rd adds “musicality” doesn't quite hold up."

https://www.passdiy.com/project/articles/audio-distortion-and-feedback

 

[Feanor]

No, Nelson Pass don´t say that.

"Audiophiles have been accused of using 2nd or 3rd harmonic distortion as tone controls to deliberately alter the sound. I suppose that there are people who like it that way, but I don't think this is generally the case. For reasons which will become clearer when we talk about inter-modulation distortion, high levels of any harmonic become problematic with musical material having multiple instruments, and the argument that 2nd or 3rd adds “musicality” doesn't quite hold up."

https://www.passdiy.com/project/articles/audio-distortion-and-feedback

You're right: Pass didn't say that, (I had previously read that reference BTW). OTOH his designs seem to tolerate of 2nd & 3rd order HD; see Stereophile review of the XA60.5 ... https://www.stereophile.com/content/pass-labs-xa605-monoblock-power-amplifier-measurements. Clearly it isn't Pass' intent to minimize distortion to the rigorous degree that folks here at ASR demand. I supposed may mention that his amps do sound great, (apart from the "fact" that "All amplifier sound the same" )

Designers tend not to admit that they aren't minimizing distortion, much less that they are adding distortion. A favorite dodge used by Pass and others is that they are minimizing high-order HD products -- this happens to be the primary justification for minimizing negative feedback that is seen as causing the higher order products.

 

[Mark Paanakker]

Although this post is somewhat old... The fact that people like (some forms) of distortion has actually been demonstrated in multiple studies. For sure there are different flavors. In some indirect way I have argued that feedback in amps can help. The 2 different (apparently conflicting) views brought forward in this thread are somewhat reconciled by it. I'll attach it as an interesting read.

Cheers,
Mark

 

[SIY]

Although this post is somewhat old... The fact that people like (some forms) of distortion has actually been demonstrated in multiple studies. For sure there are different flavors. In some indirect way I have argued that feedback in amps can help. The 2 different (apparently conflicting) views brought forward in this thread are somewhat reconciled by it. I'll attach it as an interesting read.

Cheers,
Mark

I'm continually baffled by why Nelson Pass's claims about distortion have a life of their own. He has, to my knowledge, never ever demonstrated that his assertions are correct, yet they're taken as gospel.

I'd nominate the 300B as the Most Overrated Tube of All Time. 12AX7 can perform well, but not the way it's generally used.

 

[Mark Paanakker]

I'm continually baffled by why Nelson Pass's claims about distortion have a life of their own. He has, to my knowledge, never ever demonstrated that his assertions are correct, yet they're taken as gospel.

I'd nominate the 300B as the Most Overrated Tube of All Time. 12AX7 can perform well, but not the way it's generally used.

That's a repetition of what has been said in this thread over 10 times, with the addition of an unsubstantiated opinion regarding 300B tubes. Very useful.