The first is simply the lack of citation of any historical or scholarly references, or measurement data. This need not be . . . I think that Atmasphere could actually support his design choices much better by a properly researched, documented, and peer-reviewed paper on the subject.
Kirkus, thanks for your comments. I can produce a lot of data on this, but FWIW the paper I have often linked to is designed to be easy to read, and also easily understood by someone who does not want to go through mathematical proofs that they may well not care about. However I take your point so perhaps a second document that is a bit more in-depth?
The second is the interchangability of pespectives on amplifier output power ratings, output impedance, and circuit design, specifically the use of negative feedback. Obviously, Atmasphere prefers Class A OTL designs with little loop feedback . . . and the consequences are very poor power efficiency and a high output impedance, both of which are key factors in the rated power outputs of Atmasphere amplifiers.
I should clear something up here. Yes, this paper is on our site but it does not specifically point to our gear as the solution. I see this as a bigger issue! I have had a few people accuse me of making this stuff up- what I suggest to them is use Google to look for older Fisher amplifiers with the variable damping control (it has a spot on the dial labeled 'constant power').
I don't go for feedback simply because I've never seen it sound right (I use master recordings as a reference FWIW). Audiophiles are always looking for that 'difference' in gear that is responsible for it sounding like music rather than electronics. I have run into plenty of info on the subject, from such luminaries in the industry as Norman Crowhurst and Nelson Pass and it seems redundant on my part to restate that which has been part of the lexicon now for over 50 years.
The wholly illogical part is the inherent value judgement against amplifiers that have clipping-power-versus-impedance characteristics that are different from Atmaspheres'. Because an examination of this data gives ZERO information about how much current an amplifier will produce, what its output impedance is, or the amplitude of any distortion products in its output.
Again, this is not just about the amplifiers with which I am associated. You will find similar characteristics in SETs and P-P tube amps that feature little or no feedback. Contrast that against the Wolcott, a tube amplifier that behaves as a Voltage source, and the point is made in better contrast. But there are SS examples of zero feedback designs which I have often mentioned and admire.
With regards to why I eschew loop negative feedback, which I regard as a crucial issue, to restate myself: feedback adds odd-ordered harmonic distortion to nearly any (traditional) tube or transistor design and it is very easy to prove that our ears use odd orders to determine the volume of a sound (I can provide a simple test scenario that almost anyone with test equipment can perform to prove this). Essentially, a fundamental rule of human hearing is being violated in order to reduce (apparent) distortion on paper and to reduce output impedance. Its not coincidence that feedback affects distortion the way its does and we are able to hear its effects the way we do.
Regarding distortion, FWIW our gear makes primarily the 3rd harmonic and is absent of the even orders on account of cancellation throughout the circuit, due to its fully differential topology.
Now I think it is understood that I prefer tubes, but the truth of the matter is I would rather work with transistors. They are a lot less work to build than a good tube amp! So far I've not got transistors to sound right, and I have yet to hear a transistor amp that does. Being in the industry, I've had plenty of opportunity to hear a lot of amps- it comes with the territory. We can write that off to preferences, but there is enough standing evidence and literature to the contrary (IOW, that there is more to it than preferences) without *my* input.
