agree with doug self?

[lt cdr data]

do you agree/disagree with his arguments? I find them compelling, I am no design expert, however, his findings are scientific and repeatable for all to see.

at times I find him TOO objective, I listen to music to enjoy it, and his arguments about all music going via 100s of 5534 chips hold no sway, as he says thus the sound of other opamps don't matter simply bcos your signal has passed via these, its like saying bcos of that, you can't hear the effect of altering valves.

he also questions resevoir capacitance, I am as yet undecided as to size and effect

it has been known for decades, the effect that a tube characteristic has on distortion, ie asymetrical giving 2nd harmonic distortion and symetrical giving odd, ie cancelling even, however, I feel he deserves nigh on a nobel prize for his work in deconstructing each stage and showing the type of distortion each produces. he also argues cogently that distortin should be reduced ad infinitum as long as the price is negligible, bear in mind moving coil speakers distort perhaps 5% or more this swamps even 0.1% distortion so one could argue like he does that one can't hear that for the speaker non linearity.

but full marks to doug for his landmark work.

I have only one real problem with it, it makes amplifiers boring, as in if we followed it, we would all end up with the same amplifier.

finally, would you like anything to be added to the books? ie areas you think are not covered sufficiently? or you would like to see more of?

[Leon08]

I would say that his books are difficult to read and understand for beginners, and information wich he gives is very common and not going into details.

[DF96]

I think his book makes a valuable contribution. He covers points which sometimes get ignored. However, like all authors, he has assumptions/beliefs/prejudices so it depends to some extent on whether you share these. If you do, then you will follow his advice. If you have the opposite ones, you will probably want to ignore him.

I would love to have a perfectly boring amplifier, which adds nothing and takes away nothing. Too many people seem to confuse amplifiers with effects boxes. For me at present that means a reasonably neutral valve amplifier (no "tube sound"!).

Self was not writing for beginners, and the information he gives is not all common although some of it may be.

[Bob Cordell]

Originally Posted by lt cdr data

do you agree/disagree with his arguments? I find them compelling, I am no design expert, however, his findings are scientific and repeatable for all to see.

at times I find him TOO objective, I listen to music to enjoy it, and his arguments about all music going via 100s of 5534 chips hold no sway, as he says thus the sound of other opamps don't matter simply bcos your signal has passed via these, its like saying bcos of that, you can't hear the effect of altering valves.

he also questions resevoir capacitance, I am as yet undecided as to size and effect

it has been known for decades, the effect that a tube characteristic has on distortion, ie asymetrical giving 2nd harmonic distortion and symetrical giving odd, ie cancelling even, however, I feel he deserves nigh on a nobel prize for his work in deconstructing each stage and showing the type of distortion each produces. he also argues cogently that distortin should be reduced ad infinitum as long as the price is negligible, bear in mind moving coil speakers distort perhaps 5% or more this swamps even 0.1% distortion so one could argue like he does that one can't hear that for the speaker non linearity.

but full marks to doug for his landmark work.

I have only one real problem with it, it makes amplifiers boring, as in if we followed it, we would all end up with the same amplifier.

finally, would you like anything to be added to the books? ie areas you think are not covered sufficiently? or you would like to see more of?

Doug has done a lot of really good work and measurements over the years, but he and I have not always had the same point of view. I am less comfortable stating that amplifiers will sound the same if they both measure well under traditional laboratory conditions. I am not one to throw out measurements, as many of you know, but I do believe that much more needs to be done in the way of measurements and in terms of how amplifiers behave in the real world, as I discuss in several chapters in my book (CordellAudio.com - Home). I believe that one reason amplifiers sound differently is that they misbehave differently.

There have been times when Doug has poo-poo'd some things that I thought needed to be taken more seriously. For example, he was late to the party with DC servos and the benefits of output Triples. I also disagree fairly strongly on the merits of CFP output stages (I don't like them). And of course then there are MOSFETs, on which we totally disagree.

But this is all what makes amplifier design so interesting. We have different design philosophies and interpret measurements differently sometimes. There is plenty of room for different opinions and approaches in the amplifier design business. Doug has done a very good job of uncovering and de-mystifying a lot of amplifier design issues over the years, and for this he deserves a lot of credit.

Cheers,
Bob

[AKSA]

Very nicely put, Bob!

Congratulations on your seminal book!

Hugh

[jacco vermeulen]

I think you should read at least one book, then read another for a second opinion.
The beauty of having two books is that they are stackable, and you can reach stuff on the upper shelf in the kitchen.

[pooge]

Originally Posted by jacco vermeulen

The beauty of having two books is that they are stackable, and you can reach stuff on the upper shelf in the kitchen.

As long as you write down the titles of the books you are standing on when writing your audio review.

[benb]

Originally Posted by lt cdr data

do you agree/disagree with his arguments? I find them compelling, I am no design expert, however, his findings are scientific and repeatable for all to see.

I haven't read Self's, Cordell's or anyone else's power amplifier books (I recall from browsing Amazon there are other authors out there), but years ago I was fascinated by his webpage "Distortion In Power Amplifiers" which (as I remember) described what Self saw as the eight most significant sources of distortion in bipolar audio power amplifiers.

I may eventually get these books, but despite being an electronics design engineer with an interest in audio, I don't know if I'll ever design or build my own power amplifier(s). It's not that it's not an interesting project, but I have too many other interesting projects I want to do, and I might actually contribute something worthwhile to the field elsewhere (like here) whereas I doubt I could with power amplifiers.

Originally Posted by lt cdr data

...
it has been known for decades, the effect that a tube characteristic has on distortion, ie asymetrical giving 2nd harmonic distortion and symetrical giving odd, ie cancelling even, however, I feel he deserves nigh on a nobel prize for his work in deconstructing each stage and showing the type of distortion each produces. he also argues cogently that distortin should be reduced ad infinitum as long as the price is negligible, bear in mind moving coil speakers distort perhaps 5% or more this swamps even 0.1% distortion so one could argue like he does that one can't hear that for the speaker non linearity.

I don' think anyone disagrees with that, I feel there's still good reason to reduce distortion as much as practical. Speakers and solid-state amplifiers generate different types of distortion (they tend to generate more high-order harmonics, speakers tend to generate mostly lower harmonics, as well as more stuff I mention below), and one type doesn't completely mask the other type. They each generate a different "spectrum" of distortion. This is a fact that is easily forgotten with the ubiquitous use of THD as the only and all-encompassing distortion figure. I recall mentioning this in a long op-amp thread here.

Even with tube amplifiers, the distortion is for the most part caused by the tube's nonlinear transfer function (though audio transformers give a yet different kind of distortion), whereas much distortions in speakers are caused by all sorts of resonances and diffractions, things one might think of as "ringing, delay and echo effects," things not generated by amplifiers. Again, the distortions are different and don't fully overlap, and the ear can usually hear the presence of one type of distortion even through another type, regardless of the "THD" figures of each type.

[wakibaki]

Originally Posted by Bob Cordell

There is plenty of room for different opinions and approaches in the amplifier design business.

When I go to the doctor for a diagnosis, I don't want to hear about 'room for different opinions', I want to hear 'this is what you've got and you won't hear much disagreement about it'. This is the kind of service doctors would prefer to offer, and there is a constant striving in medicine, as in all but very few professions, to eliminate disagreement, which is basically an indication of an incomplete understanding. It is an approach which is, to all intents and purposes, a sine qua non for a discipline to qualify as a science.

I don't find it at all encouraging that you seem to revel in the situation prevailing in audio engineering.

Perhaps you think it's entertaining or potentially profitable for you to have a little niche where there's a load of disagreement among 'experts', but I can assure you there are many of us out here who look forward to the day when the arguments about audio engineering have been thoroughly beaten to death as they have been in other disciplines.

Doug may have been 'late to the party' in some instances, but as they say, 'better late than never'. A certain caution is appropriate in an engineering professional anyway. His attitude certainly seems to me to look forward to the day when there will be little or no room for differing opinions in an entirely seemly way.

w

[Damon Hill]

That argument assumes that there is only 'one right way', which is too much like religion to me. What I'm seeing in reality is multiple valid approaches to electronic design, given the available knowledge and technology. This may result in many good designs, with a few outstanding designs, depending on the depth of one's finances.

I doubt there will ever be a 'perfect' and 'ultimate' amplifier design, and that future engineering will be exercises in gilding the lily. There are too many variables and requirements, which makes necessary practical compromises. That's just the reality of the matter. Some of us find that fun.