"There is plenty of room for different opinions and approaches in the amplifier design business."
I'm glad to see that my statement sparked some spirited debate on both books and audio design itself. After being an engineer for many years in several electrical engineering disciplines, I have learned that one-size-fits-all does not apply even in highly developed engineering disciplines. Engineers are driven to different designs and different design philosophies by many different factors. Indeed, I have often said that one thing that largely affects how an engineer designs a circuit is what he fears the most. This is another way of describing risk management. No engineer knows everything and no engineer has experience with all the various approaches to accomplishing something. They make the best of what they have in terms of knowledge and experience and this leads legitimately to different outcomes. Similarly, in audio, some give more credibility to anecdotal information and are more curious about seeing if something non-intuitive is true.
Anyway, I have high respect for Doug, and his book paved the way for mine. Indeed, I have three editions of his amplifier book on my shelf. Reading his book also made me think about numerous issues that I might not otherwise have thought of. I have also learned a lot from those whom I've had disagreements with in both the recent and distant past. Disagreements make you think, as long as you don't totally dismiss the other person's view. Although I have not always agreed with him, I have learned a huge amount from Matti Otala, for example.
I hope that anyone who has read Doug's book will enjoy and learn from my book, and vice-versa. In some ways they are complementary, each giving different amounts of emphasis to different topics and each representing a different writing style.
Cheers,
Bob
I'm glad to see that my statement sparked some spirited debate on both books and audio design itself. After being an engineer for many years in several electrical engineering disciplines, I have learned that one-size-fits-all does not apply even in highly developed engineering disciplines. Engineers are driven to different designs and different design philosophies by many different factors. Indeed, I have often said that one thing that largely affects how an engineer designs a circuit is what he fears the most. This is another way of describing risk management. No engineer knows everything and no engineer has experience with all the various approaches to accomplishing something. They make the best of what they have in terms of knowledge and experience and this leads legitimately to different outcomes. Similarly, in audio, some give more credibility to anecdotal information and are more curious about seeing if something non-intuitive is true.
Anyway, I have high respect for Doug, and his book paved the way for mine. Indeed, I have three editions of his amplifier book on my shelf. Reading his book also made me think about numerous issues that I might not otherwise have thought of. I have also learned a lot from those whom I've had disagreements with in both the recent and distant past. Disagreements make you think, as long as you don't totally dismiss the other person's view. Although I have not always agreed with him, I have learned a huge amount from Matti Otala, for example.
I hope that anyone who has read Doug's book will enjoy and learn from my book, and vice-versa. In some ways they are complementary, each giving different amounts of emphasis to different topics and each representing a different writing style.
Cheers,
Bob
I've read Doug Selfs textbook and learned a lot from it. Firmly objectivist, which is a good thing, but at the same time gives hard evidence on, say, why certain capacitors sound different and even an anecdote of non-linear loudspeaker terminals. My main interest is loudspeakers, but understanding something about amplifier design is quite useful to build a sane loudspeaker load.
A lesson I learned at MIT a long time ago: Get two or more textbooks in the subject. They may teach the same thing from different angles, and given the tuition these days, the second book has paid for itself if it helps you graduate 15 minutes faster.
As a result of this thread, I just ordered your book from Amazon, Bob, and will read it equally thoroughly. The abstracts posted on your web site look very interesting.
A lesson I learned at MIT a long time ago: Get two or more textbooks in the subject. They may teach the same thing from different angles, and given the tuition these days, the second book has paid for itself if it helps you graduate 15 minutes faster.
As a result of this thread, I just ordered your book from Amazon, Bob, and will read it equally thoroughly. The abstracts posted on your web site look very interesting.
I don't think so. It's just that in most fields of human endeavour there are many more ways to be wrong than to be right, so those of us who are right are quite likely to agree with each other!
To get back to Doug Self, other things being equal 0.01% distortion is clearly better than 0.1% distortion. The problem is that other things often are not equal, so as in all engineering there is a balance to be struck and different people strike the balance in different ways and different places. I do reserve the right, though, to laugh at people who have manifestly not got a balance (one end of their see-saw is clearly scraping the ground) yet they believe theirs is more balanced than anyone else.
If I may be indulged in quoting myself from about a week ago in another thread....
"IMHO whilst Self's articles are very good and generally well researched he is guilty of propagating a "there is only ONE correct way of implementing any piece of circuitry and that's Doug Self's way" attitude.
In spite of this I am still very grateful to Self for those areas in which he has increased understanding for many people, including myself. Even if I don't always agree with his viewpoint on matters such as Mosfets... "
I will add to that by observing that, from personal experience, some of Self's topologies may well be close to optimum for measured performance but they certainly are NOT optimum in terms of subjective performance
"IMHO whilst Self's articles are very good and generally well researched he is guilty of propagating a "there is only ONE correct way of implementing any piece of circuitry and that's Doug Self's way" attitude.
In spite of this I am still very grateful to Self for those areas in which he has increased understanding for many people, including myself. Even if I don't always agree with his viewpoint on matters such as Mosfets... "
I will add to that by observing that, from personal experience, some of Self's topologies may well be close to optimum for measured performance but they certainly are NOT optimum in terms of subjective performance
I truly appreciate Mr. Self's writings. His 5'th edition was the reading that got me started a year and a half ago. The first "frugalamp" I made with the mje340/350 VAS was from the book. I was so amazed that it worked and that it sounded better than the sony it replaced. My latest AX voltage module is a tribute to his "blameless" and is both amazing in it's simplicity and low THD (below- .0003% /1khz). For any beginner , or even those a little further on with things , the "blameless" is a required build that gives by far the most "bang for the buck" in such a simple package. Thank you , Mr. self , I really like the blameless ... it adds nothing (sterile) to the sound. There are other topologies that may sound slightly better , but simplicity has it's advantages.
OS
OS
Attachments
Debate. Yes, if we wanted a debate then good job. Unfortunately what we need is some agreement.
It seems that a number of people want to disagree with me, if only on my choice of analogy.
I should have noted this, from the OP...
Amplifiers are boring. We're all going to end up with the same amplifier at some point in the future, once all the hoo-ha dies down. An amplifier can only get so good, then it's perfect. In terms of music reproduction it doesn't even need to be perfect. It doesn't matter mosfet, bjt, single ended, push-pull or class D, they're all the SAME. They're probably all the same already as far as human hearing is concerned. The progressively improving numbers for THD, etc., in modern products have nothing to do with performance, they're about nothing other than advertising copy, and such advertising copy can only be useful where there is debate.
I note in another thread somebody who wants a 1ppm amplifier. It's this kind of obsession with numbers that makes us the victims of salesmen. Not so long ago there was a move in this country to limit motorbikes to 100BHP. Those of us who owned motorbikes all asked 'what's magic about 100?'
For those who didn't like the medical example, let me give you another one.
Spectacles.
Spectacles are good enough. No-one is out there making more accurate spectacles. This is because spectacles are good enough. In fact we can make lenses quite a lot better than we put in spectacles. We use them for cameras, Hubble telescopes (you thought) and the like. We make and use (comparatively) inferior lenses for spectacles because they are cheaper, and in that sense, better. Everybody gets the same lenses (in terms of quality) in their spectacles. All the opticians in the world agree there's no point in using anything better. There's just no debate.
We have almost certainly gone past the point where there are meaningful distinctions between amplifiers in terms of sound reproduction. The meaningful distinctions are in terms of cost, efficiency, size, weight, portability and (although it comes hard from me, having none to speak of) style.
w
Let me ask all these subjectivists a question. Do you think the companies selling you these expensive amps are doing what they do because they want to produce the ultimate amp for the benefit of mankind? Of course not. They're doing it to make money to make a living for themselves and profits for their investors. That's fair enough, but let's not pretend it's otherwise.
An analogy - you'd think after 150 years they would have come up with the ultimate shampoo yet every year tens of thousands of people are employed worldwide to dream up new shampoos. Why? So gullible people will buy their brand instead of their competitors because of its super duper new miracle ingredient and keep their companies in business. The fact that they all just use the same liquid soap doesn't seem to matter. What's more, they dream up exactly the same kind of pseudo scientific subjectivist crap that high-end amp makers come up with to justify what they do.
We can easily settle this argument once and for all with a series of properly conducted double blind tests. If they show, as I am 100% sure that they will, with high statistical confidence, that all amps sound the same unless driven into clipping etc. then that's the end of the matter and all these fancy dan audiophile companies can shut up shop and lay everyone off.
Somehow I don't think there will be a stampede of volunteers.
An analogy - you'd think after 150 years they would have come up with the ultimate shampoo yet every year tens of thousands of people are employed worldwide to dream up new shampoos. Why? So gullible people will buy their brand instead of their competitors because of its super duper new miracle ingredient and keep their companies in business. The fact that they all just use the same liquid soap doesn't seem to matter. What's more, they dream up exactly the same kind of pseudo scientific subjectivist crap that high-end amp makers come up with to justify what they do.
We can easily settle this argument once and for all with a series of properly conducted double blind tests. If they show, as I am 100% sure that they will, with high statistical confidence, that all amps sound the same unless driven into clipping etc. then that's the end of the matter and all these fancy dan audiophile companies can shut up shop and lay everyone off.
Somehow I don't think there will be a stampede of volunteers.
Last edited:
I am probably what you would call a subjectivist.
I agree... but would go on to say that some of these companies produce outstanding products that really deliver the goods sonically. The fact that they may be overpriced or turn in a less than perfect technical score is another matter.
I am not qualified to say or judge the matter... all I know is my hair feels "scratty" and horrible after some and not others
Again price charged is a different matter I think.With the greatest respect, that statement suggests to me that you may not be receptive to what are in reality "small" differences in subjective performance between designs, and there's no shame in that at all, it's just the way it is for some folk, but to someone who can pick up on those minute differences, it makes all the difference in the world between an amp that is trully captivating to listen to and one which is just mediocre.
As an electronics engineer I hate to say that... all my instincts tell me to strive for technical perfection... my ears, when I sit down to listen tell me different and that it's not the whole story.
Mooly
Double blind tests are the only way of doing this because the human mind is able to fool our judgement so easily. Our auditory memory is fleeting - how can you say that something you just heard is better or worse than something you heard 5 minutes or 5 days ago? Switching between amps as the same source is playing music and then deciding 1. whether a switch has really been made or not and 2. whether the switched amp sounds better than before is the only foolproof way of doing it. If you can't tell a difference that's because there isn't one! Simples.
We can make the blind test unimpechable by just inviting professional musicians and golden eared subjectivist gurus to the listening panel. Lots of them to make it statistically valid.
The subjectivists could then chose the best speaker system in the world to play the amps through, connected to the amps by their ultimate cable of choice. The amps would be powered by a separate line to the local 230v substation - made of the subjectivists choice of ultimate mains cable of course.
If they then could not identify whether it was a £20k or a £200 amplifier playing would you then accept that they all sound the same?
I know, maybe it was the thickness of the gold plating on the A-B selector switch that was hiding the differences all along.
So how come the high-end companies have never done that exercise? Hmmm?
Double blind tests are the only way of doing this because the human mind is able to fool our judgement so easily. Our auditory memory is fleeting - how can you say that something you just heard is better or worse than something you heard 5 minutes or 5 days ago? Switching between amps as the same source is playing music and then deciding 1. whether a switch has really been made or not and 2. whether the switched amp sounds better than before is the only foolproof way of doing it. If you can't tell a difference that's because there isn't one! Simples.
We can make the blind test unimpechable by just inviting professional musicians and golden eared subjectivist gurus to the listening panel. Lots of them to make it statistically valid.
The subjectivists could then chose the best speaker system in the world to play the amps through, connected to the amps by their ultimate cable of choice. The amps would be powered by a separate line to the local 230v substation - made of the subjectivists choice of ultimate mains cable of course.
If they then could not identify whether it was a £20k or a £200 amplifier playing would you then accept that they all sound the same?
I know, maybe it was the thickness of the gold plating on the A-B selector switch that was hiding the differences all along.
So how come the high-end companies have never done that exercise? Hmmm?
Last edited:
Mr Self is very good at vulgarizing audio amps design,
in short, a good summary of works that has been largely
made by japanese amps designers, but unfortunately,
their names are often if not always unknown in this side
of the world.
Though, their work is largely visible in all the amps designed
30 years ago, to the point that some of their implementations
in commercial products where later rediscovered and credited
to occidental researcher, as the so called Hawksford cascode
that was common in Luxman s amps of the late 70s...
in short, a good summary of works that has been largely
made by japanese amps designers, but unfortunately,
their names are often if not always unknown in this side
of the world.
Though, their work is largely visible in all the amps designed
30 years ago, to the point that some of their implementations
in commercial products where later rediscovered and credited
to occidental researcher, as the so called Hawksford cascode
that was common in Luxman s amps of the late 70s...
I've read Doug Selfs textbook and learned a lot from it. Firmly objectivist, which is a good thing, but at the same time gives hard evidence on, say, why certain capacitors sound different and even an anecdote of non-linear loudspeaker terminals. My main interest is loudspeakers, but understanding something about amplifier design is quite useful to build a sane loudspeaker load.
A lesson I learned at MIT a long time ago: Get two or more textbooks in the subject. They may teach the same thing from different angles, and given the tuition these days, the second book has paid for itself if it helps you graduate 15 minutes faster.
As a result of this thread, I just ordered your book from Amazon, Bob, and will read it equally thoroughly. The abstracts posted on your web site look very interesting.
Thanks, asbjbo. I always like to triangulate to the best understanding by reading more than one book. I have also learned a lot from Ben Duncan and Rany Slone.
Thanks for buying my book and I hope you enjoy the read.
Cheers,
Bob
There s nothing inflammatory, i think..
One has just to look at schematics to see that by
the end of the 70s, the japanese designers were
at least one decade in the future..
It s true that their device industry did provide them
advanced components that allowed bold designs
impossible to implment with the available devices
in europe and the us..
I think most opticians would disagree with you. Ooops, they are out to make money, I forgot
.I guess you are not old enough to need bifocals, trifocals, or "continuous bifocals". Some are far better than others. From personal experience, I've had some that gave me headaches.
I think that you are just being a little too black and white. But then again, there is plenty of room in debate for various degrees of contrast.
Cheers,
Bob
Try the Hoya brand. Both objectively AND subjectively better! Expensive, though. You get what you pay for.
I'm a subjectivist in the sense that I design amps not to measure primarily but to listen to. The ultimate purpose of an amp is to be connected to a speaker playing music, not to be wired to an AP and a big resistor reproducing sine waves.
Easy to recognise such a textbook debating technique. The false dichotomy - set up two opposed choices where one is obviously correct. Therefore the other must be false. QED. Yes, of course the companies must make profits or they'd not survive. However, Peter Drucker said that profits are rather like breathing - an indicator of life but not the purpose of it. If the only motivation for doing business is money, that business won't last long. So there are a mixture of motives - which may include some aspirations to make the biggest, or most powerful, or lowest distortion amp.Your analogy has certain weaknesses. I prefer this one. You'd think that since Haydn and Mozart's combined total of 147 symphonies, that all that could be expressed in orchestral music had been expressed. They all use pretty much the same ingredients (instruments, players, conductor) with minor variations. So why on earth did Beethoven come along? Surely, just to make money.
Well you'll certainly not be doing this experiment then as you're already 100% sure of the result. You might wish to read up a long article about Bob Carver on Stereophile. Then again, since you seem to already "know" that all amps sound the same, you probably won't want to read it. Why do you wish to introduce 'unless driven into clipping' as a constraint on your listening tests? Given that people do drive amps into clipping (often unknowingly) that's going to make the test unrepresentative of real-world use.
In my case it will be because double blind listening tests are so boring. About as boring as connecting resistors to an amplifier.
Only way? Your human mind has fooled you into thinking too narrowly.
Umm, could it be because I don't judge things as 'better' or 'worse', rather 'different' ? When I listen to Maurizio Pollini's playing of Chopin's Etudes, is my auditory memory so poor as not to remember how Murray Perahia plays them so as to compare the two? How could music critics even operate if their auditory memories were this poor?
Why would a result from two amplifiers be generalisable to all amplifiers?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.