What are the advantages/disadvantages of different amp topologies?

[morricab]

Here is something for you to consider:

Back when Mr. Recklinghausen said that the only distortion specs were IMD and THD. IMD means quite a lot since its really audible. But THD not so much- in fact really low THD figures have often been associated with bad sounding amps.

But what they didn't do back then was measure distortion vs frequency, which was unfortunate because there is a strong correlation between the musical nature of an amplifier and its distortion vs frequency curve.

For example, if I have this right you like SETs. It might interest you to know that all zero feedback amplifiers have a ruler flat DvsF curve across the audio band. Its not even right to say its a curve. This is IMO/IME essential to getting an amp to sound musical. To explain to you why that is so gets pretty technical as you have to talk about feedback and Gain Bandwidth Product, and why if those things aren't right the amp can sound messed up. I think I might have covered that already in this thread anyway.

Amps that don't have feedback don't have to worry about Gain Bandwidth Product so how musical they are depends on their distortion spectra. There is a member here who goes by 'morricab' who is an SET aficionado and he harps on how you have to have exponential decay of amplitude of higher ordered harmonics as the order of the harmonic is increased (turns out as long as the decay of higher orders is on an exponential curve, the actual exponent isn't important but that's another topic) in order for the amp to sound right. He's not wrong.

That's clearly a set of measurements!

This being the internet you can say what you want but that does not make it true. Engineering makes all audio (including SETs) possible as well as cars, the internet, airplanes and so on. Its not like you get to pick and choose; measurements are essential in all engineering disciplines.

Geddes found no better correlation with IMD than he did with THD and sound quality…none.

 

[DasguteOhr]

Measurements are only a guide to an exciting rendition of music that will closely resemble the experience of the live performance.

Do conductors use measurements to judge the performance of their orchestra? No of course they don't!

Do orchestras sound the same when they play in different concert halls? No of course they don't.

But likewise all amps should not sound the same, which in theory, they would if they all measured perfectly. They don't, and neither should they measure perfectly. The best measuring amp I've ever owned happens also to be just about the dullest. A bit a "character" in amps and speakers will make as big a difference to one's enjoyment of the music as choosing the best concert hall with the brightest orchestra, led by the most engaging conductor.

I think it's the wrong approach, an amplifier should make sound quality and the differences between signal sources audible. E.g how do you want to distinguish between different DACs when the amp put their own stamp on the music. It doesn't matter whether the executing element is tube or transistor it can be implemented with both. If you go the other way you compensate for a mistake with another mistake my opinion.
Your room and your speakers change the sound much more than an amplifier. Isn't it nice to have a constant in the equation?

 

[acousticsguru]

I think it's the wrong approach, an amplifier should make sound quality and the differences between signal sources audible. E.g how do you want to distinguish between different DACs when the amp put their own stamp on the music. It doesn't matter whether the executing element is tube or transistor it can be implemented with both. If you go the other way you compensate for a mistake with another mistake my opinion.
Your room and your speakers change the sound much more than an amplifier. Isn't it nice to have a constant in the equation?

Agree wholeheartedly! Also, philosophically speaking, it would seem pointless to try and build a system on the concept of compensation.

Greetings from Switzerland, David.

 

[Republicoftexas69]

Not all engineers rely only on measurements... Maybe I should put that a different way:

Some engineers are also audiophiles and care a lot about how things sound. But its a simple fact that no matter what kind of amp you have, you have to take measurements to make sure the amp works. Voltage tests are measurements- if they are off from nominal values you start looking for why, like a bad tube, wiring error or the like... measurements are essential and anyone thinking otherwise is trying to live their life according to a made up story; a myth.

The pragmatic engineer has to do a bit of research to be able to draw a line between what we can hear and what we can measure.

It turns out that in the last 30 years our ability to measure things has dramatically improved! But the understanding of what the measurements are showing us is considerably more rare- even to those doing the measurements.

Here's an example: I don't think anyone would argue that harmonics are what define the individual sound of musical instruments- for example the difference between a trumpet and a clarinet. But the measurement guys get up in a huff when you try to tell them that the distortion of the amplifier is modifying that sound of those instruments and that you can hear that as a change in their tonality.

I'm not one of those; I'm the in-between guy that has sorted out an obvious and simple fact: that the 'sound' of any amp, its 'sonic signature' is exactly two things, the first might be frequency response error but the major aspect being how the amp makes distortion. Further, I understand that when you put a musical signal through an amplifier, the harmonic distortion the amp makes isn't a separate signal that might be so many dB down and perhaps 'not audible' because its masked... because the amp makes distortion it adds harmonics to the musical instruments and thus modifies their tonality.

This is a simple realization and its weird to me that most people haven't figured this out; if they have they certainly are not vocal about it!

If you've heard differences between various tube amps, consider this: the differences between various class D amps is more extreme. Some are so bad they are unlistenable to me, some are boring as it gets (lifeless), some are wimpy and don't play bass and some are outright musical by any metric.

When the viewpoint is myopic, such as a seemingly zealous/religious adherence to only one approach to audio, that's when you get in trouble. For decades, people thought of me as that guy that makes OTLs. No-one expected we would produce our own class D amp since tube guys must not know anything. All I can say to that is if you have an engineering degree, if its to be useful you also have to be pragmatic. The head engineer at HH Scott, Daniel Recklinghausen, once said "If it measures good and sounds bad, -- it is bad. If it sounds good and measures bad, -- you've measured the wrong thing." If you think measurements are unimportant you are missing the truth of this very insightful statement.

Thank you Ralph.

 

[morricab]

Agree wholeheartedly! Also, philosophically speaking, it would seem pointless to try and build a system on the concept of compensation.

Greetings from Switzerland, David.

Well, hate to break it to you…all systems are built on at least some degree of compensation…with each and every component. All have a signature of some sort that pulls them away from true accuracy. So, compensation will ensue…also to suit one’s own imagination of what the real thing sounds like…if that is even their goal.

 

[morricab]

I think it's the wrong approach, an amplifier should make sound quality and the differences between signal sources audible. E.g how do you want to distinguish between different DACs when the amp put their own stamp on the music. It doesn't matter whether the executing element is tube or transistor it can be implemented with both. If you go the other way you compensate for a mistake with another mistake my opinion.
Your room and your speakers change the sound much more than an amplifier. Isn't it nice to have a constant in the equation?

Agree differences should be easy to hear but that is not the same as having no character …they all have some (or a lot).
While I agree that speaker and room make bigger quantitative changes to the sound, I think electronics have a more damaging effect on the believability of that sound.

 

[DasguteOhr]

Agree differences should be easy to hear but that is not the same as having no character …they all have some (or a lot).
While I agree that speaker and room make bigger quantitative changes to the sound, I think electronics have a more damaging effect on the believability of that sound.

with your second statement +1
A neutral sounding amplifier doesn't need any character. It must have bandwidth, process impulses quickly and should not change the timbre of an instrument. An good example for that would be the Bryston, a really good amp with DAC and phono mm + sut

https://www.stereophile.com/integratedamps/407bryston/index.html

 

[morricab]

with your second statement +1
A neutral sounding amplifier doesn't need any character. It must have bandwidth, process impulses quickly and should not change the timbre of an instrument. An good example for that would be the Bryston, a really good amp with DAC and phono mm + sut

https://www.stereophile.com/integratedamps/407bryston/index.html

There is no such thing as a neutral sounding amplifier...they all have character...Bryston is a particularly poor choice as an example, IMO. They have a rather unnatural sound...at least the examples I have heard. I also owned a Bryston active crossover and had to sell it because of the overlay of haze it put on the sound.

iMO, you are operating from an unrealistic theory that there is such a thing as neutral amplifier...or neutral anything for that matter.

 

[Hear Here]

Your room and your speakers change the sound much more than an amplifier. Isn't it nice to have a constant in the equation?

Yes, I agree totally there. Variations in speakers has by far the largest influence on what you hear. However if you choose one type of speaker, the matching ideal amp may be far different from the ideal matching amp for a different type of speaker. Even once a particular speaker is chosen, the variation in amp will also contribute towards the enjoyment or otherwise of what you hear at your listening chair.

When I decided to move from tube amps to solid state, I needed to find one that offered as good or better sound than the SETs I'd been using previously. The first ss I bought (on rave recommendation from a fellow Avantgarde speaker owner) was very accurate and dead silent, but drearily dull. Great for "elevator music" as I think background music is described in the US) but not good if you want to listen to the music and be thrilled by what you hear!

I was expecting a Class A amp to win out from the 12 amps I bought or borrowed for home tests and indeed the Accuphase A-36 was pretty impressive. However in the end it came down to an AB (GamuT D200) or a Class D (NAD M33). Sadly the GamuT has no surge protection when turned on or off so it caused a frightening thump through my 107 dB speakers. I still greatly enjoy the M33 after 3 years.

I have less interest in choosing a DAC as they have so much less "character" compared with speakers in particular, but also amplifiers. My upgrade path is likely to be the forthcoming NAD M66 (streamer / DAC / preamp) and the Atma-Sphere Class D mono amps. The M66 includes a better DAC than the M33 but I'm not expecting that to make significant change in the characteristics of the M66. It will hopefully offer a little more finesse! If I could find an ever better streamer / DAC / preamp than the M66, I'd gladly consider it, but its control app must be robust and there are other features that NAD offer that I'd be reluctant to lose.

 

[acousticsguru]

Well, hate to break it to you…all systems are built on at least some degree of compensation…with each and every component. All have a signature of some sort that pulls them away from true accuracy. So, compensation will ensue…also to suit one’s own imagination of what the real thing sounds like…if that is even their goal.

You're missing point (mine as well as that of @DasguteOhr) resorting to truisms and arguing in absolutes, throwing out the baby with the bathwater. The question is not whether 100% neutrality (aka transparency to the source, i.e. the recording, if not live event) is a real-world option at this point in time, but about pragmatism. Provided experiencing music as if one were there with the musicians is considered a goal (maybe not yours, but some music lovers' and audiophiles'), it's strictly illogical to try and build systems maximizing, but only logical to try and minimize the overall amount of and need for compensation, in short trying to minimize individual component flaws is a goal, whether or not perfection can be achieved notwithstanding. If that weren't true, then the use of the terms "better" and "worse" for musical playback wouldn't even make sense, and this forum should be renamed "personaltasteforum". If that were your argument, I'll buy it (for the simple reason that I don't believe most audiophiles have much interest in truth), but I seem to remember you keep arguing about some things sounding "better" or "worse" than others, and why.

Greetings from Switzerland, David.

 

[Atmasphere]

with your second statement +1
A neutral sounding amplifier doesn't need any character. It must have bandwidth, process impulses quickly and should not change the timbre of an instrument. An good example for that would be the Bryston, a really good amp with DAC and phono mm + sut

https://www.stereophile.com/integratedamps/407bryston/index.html

Emphasis added; I agree. But I'm having trouble believing the Bryston is a good example. If you look at the measurements as seen at the link you provided you'll see that distortion rises with frequency beginning at 1KHz, which suggests insufficient Gain Bandwidth Product to support the feedback used. This can result in unmasked higher ordered harmonic content, which is interpreted by the ear as brightness and possibly also harshness.

 

[morricab]

You're missing point (mine as well as that of @DasguteOhr) resorting to truisms and arguing in absolutes, throwing out the baby with the bathwater. The question is not whether 100% neutrality (aka transparency to the source, i.e. the recording, if not live event) is a real-world option at this point in time, but about pragmatism. Provided experiencing music as if one were there with the musicians is considered a goal (maybe not yours, but some music lovers' and audiophiles'), it's strictly illogical to try and build systems maximizing, but only logical to try and minimize the overall amount of and need for compensation, in short trying to minimize individual component flaws is a goal, whether or not perfection can be achieved notwithstanding. It that weren't true, then the use of the terms "better" and "worse" for musical playback wouldn't even make sense, and this forum should be renamed "personaltasteforum". If that were your argument, I'll buy it (for the simple reason that I don't believe most audiophiles have much interest in truth), but I seem to remember you keep arguing about some things sounding "better" or "worse" than others, and why.

Greetings from Switzerland, David.

I got your point just fine. First of all you know my audio goals pretty well, so no need to write much of what you write, not to mention my positions are quite clearly laid out throughout the forum. Second, way too many ifs in your arguments (if this then that etc.). My point was very clear, there is no true neutral and most of what masquerades as neutral really has a signature that is unmusical. Your point is nice in theory but not really what happens in practice even if you think that’s what you do.

 

[acousticsguru]

I got your point just fine. First of all you know my audio goals pretty well, so no need to write much of what you write, not to mention my positions are quite clearly laid out throughout the forum. Second, way too many ifs in your arguments (if this then that etc.). My point was very clear, there is no true neutral and most of what masquerades as neutral really has a signature that is unmusical. Your point is nice in theory but not really what happens in practice even if you think that’s what you do.

No worries, I got the gist all along that you’re purposely ignoring the point: is it or is it not a goal to try and build systems minimizing the need for compensating individual component flaws? Now to the “ifs”: IF that is true, the whole rest of your argumentation collapses to the inconsequential disagreement that said Bryston unit (which I don’t know) may or may not be a good example of the point @DasguteOhr was trying to make. Much ado about nothing…

Greetings from Switzerland, David.

 

[DasguteOhr]

Emphasis added; I agree. But I'm having trouble believing the Bryston is a good example. If you look at the measurements as seen at the link you provided you'll see that distortion rises with frequency beginning at 1KHz, which suggests insufficient Gain Bandwidth Product to support the feedback used. This can result in unmasked higher ordered harmonic content, which is interpreted by the ear as brightness and possibly also harshness.

Bryston is clean, fast, punchy and stable under load. I didn't notice the jump from 0.01 to 0.06 thd. Even just before clipping at 135 watts at 8 ohms, distortion is not higher, which I find remarkable.In the bass range it will be higher but they only use 1khz tone

There is no such thing as a neutral sounding amplifier...they all have character...Bryston is a particularly poor choice as an example, IMO. They have a rather unnatural sound...at least the examples I have heard. I also owned a Bryston active crossover and had to sell it because of the overlay of haze it put on the sound.

iMO, you are operating from an unrealistic theory that there is such a thing as neutral amplifier...or neutral anything for that matter.

We drift back down to personal taste level, just listen to it. For me, the Bryston B 100sst and the Musical Fidelity MS6i are currently the benchmark for affordable intregrated amp.

 

[DasguteOhr]

Yes, I agree totally there. Variations in speakers has by far the largest influence on what you hear. However if you choose one type of speaker, the matching ideal amp may be far different from the ideal matching amp for a different type of speaker. Even once a particular speaker is chosen, the variation in amp will also contribute towards the enjoyment or otherwise of what you hear at your listening chair.

When I decided to move from tube amps to solid state, I needed to find one that offered as good or better sound than the SETs I'd been using previously. The first ss I bought (on rave recommendation from a fellow Avantgarde speaker owner) was very accurate and dead silent, but drearily dull. Great for "elevator music" as I think background music is described in the US) but not good if you want to listen to the music and be thrilled by what you hear!

I was expecting a Class A amp to win out from the 12 amps I bought or borrowed for home tests and indeed the Accuphase A-36 was pretty impressive. However in the end it came down to an AB (GamuT D200) or a Class D (NAD M33). Sadly the GamuT has no surge protection when turned on or off so it caused a frightening thump through my 107 dB speakers. I still greatly enjoy the M33 after 3 years.

I have less interest in choosing a DAC as they have so much less "character" compared with speakers in particular, but also amplifiers. My upgrade path is likely to be the forthcoming NAD M66 (streamer / DAC / preamp) and the Atma-Sphere Class D mono amps. The M66 includes a better DAC than the M33 but I'm not expecting that to make significant change in the characteristics of the M66. It will hopefully offer a little more finesse! If I could find an ever better streamer / DAC / preamp than the M66, I'd gladly consider it, but its control app must be robust and there are other features that NAD offer that I'd be reluctant to lose.

There are just as many differences in DACs as there are in tonearms and cartridges. This can create very big differences in the sound because it is at the beginning of the signal chain.But here we are on the topic that the amplifier has to let it through without changing or limiting it.

 

[acousticsguru]

There are just as many differences in DACs as there are in tonearms and cartridges. This can create very big differences in the sound because it is at the beginning of the signal chain.But here we are on the topic that the amplifier has to let it through without changing or limiting it.

True. I do however agree that there’s greater variety in sound among speakers than e.g. DACs. That doesn’t change the fact that nothing in a system is going to make up (compensate) for the flaws of e.g. a source component.

Note I’m not saying one can’t use a particularly “tubey” sounding amp (or, DAC, or both) to make e.g. early mono recordings “palatable”. Basically anything that’ll make someone enjoy music seems to me a perfectly valid alternative.

Having said that, playing back the same recording (examples that spring to mind where I’ve made such comparisons include such recordings as the Melchior/Lehmann/Böhm Walküre Act I, the Beecham Mozart Magic Flute, Schnabel’s Beethoven, along with countless other historical piano recordings) through a more neutral sounding system may bring the listener closer to the essence of the performance, playing or singing of an artist etc. Musical curiosity to me is an equally valid approach.

Unless one has the luxury of multiple systems, the fact remains audiophiles will have to find a compromise that suits their tastes, or, and the latter I happen to find deplorable, they’ll end up among the many gear-orientated audiophiles who will not listen to anything historical or anything they deem sonically inferior quality recordings, disregardIng musical qualities.

Greetings from Switzerland, David.

 

[Audiohertz2]

Agree,

Way more flaws in loudspeakers than there are in source quality , of DAC’s , TT’s et al ..

Regards

 

[Audiohertz2]

Geddes found no better correlation with IMD than he did with THD and sound quality…none.

Ahhhh , the great inventing Audio Guru Geddes...


Cheers

 

[acousticsguru]

I’ve often wondered if people don’t simply abuse the term “neutrality” when it comes up in discussions as an attribute of e.g. bland sounding systems. I think of it as a synonym for the longer “transparent to the source”. It’s a goal rather than a present day reality that should not deter those from pursuing it who are interested in music in the sense that they’re curious what and how an artist plays or played rather than whether it sounds nice. If I were given the choice between an audiophile system and a time machine, I’d have loved to be there in person. Point being: over and over again, whenever I'm in the mood for the real, full experience.

Greetings from Switzerland, David.

 

[Audiohertz2]

Accurate to the source or to the sound one is seeking .. ?

Hi -Fidelity is accuracy to the source ..


Regards