Thanks for your interest in Ayre products. I am sure that you will enjoy the VX-5 for many years to come. I am not sure what post you are referring to, so it is hard for me to answer any specific questions. To the best of my knowledge there have been no unusual reports of problems with DC offset in any of our products. So I can speak generally about DC offset and if your questions are not answered, please feel free to be more specific.
In the old days, when everything was tubed, it was pretty difficult to make a DC-coupled amplifier. I had an old DuMont oscilloscope when I first started out as a service tech. It came with the service manual (so I had the schematic diagram) and was from the late 1950's and was pure tubes. This is a different situation than an audio amplifier, but I was actually able to use it as an audio amplifier of sorts. Please allow me to explain. The input stage had either 12AX7's or 6DJ8's (I don't recall any more) with the grid referenced to ground with a 1 Mohm resistor, which is what gave it a high input impedance. The plates of the first stage had about +100 VDC on them, which were direct-coupled to the next stage. In a high gain tube like this, the cathode will be within a few volts of the grid, so the second stage was basically referenced to +100 VDC (requiring a floating heater supply biased up to +100 VDC) and then its plates were up around 200 VDC. I can't remember any more whether there were three or four gain stages, but by the end of the signal chain, there was about +400 VDC at the output of the amplifiers.
Normally this would be a pretty big problem. You certainly couldn't hook this up to an ordinary speaker or anything. But the amps in an oscilloscope are connected to the deflection plates of the Cathode Ray Tube (CRT or display). The job of the deflection plates is exactly what it sounds like -- plates to deflect something. In this case the "something" is a beam of electrons generated by an "electron gun", which not only generates free electrons, but focuses them into a small beam (circa 1 mm) and then the beam uses a positively charged screen grid to accelerate the electron beam to the front of the display. When the beam strikes special chemicals called "phosphors", they cause the phosphors to glow. And then there are two sets of deflection plates, one pair mounted horizontally to "steer" the beam to the right and left (to represent time) and one pair mounted vertically to move the beam up and down to form a representation of the input voltage (signal).
In this case, the fact that the deflection plates had +400 VDC on them was exactly what was required, as by adjusting the relative voltage on the pair of deflection plates, the beam would be "steered" on its way to the screen coated with phosphors. Now it turns out that this same trick can be used to drive electrostatic speakers, and for some reason there was an access plate on the rear of the 'scope that allowed one to connect a signal directly to the deflection plates. Instead I used it to take the output of these DC coupled tube amplifiers and connect them to a pair of electrostatic tweeter panels I had bought from RTR, the OEM supplier for JansZen, Infinity, and several other electrostatic speaker companies.
After reveling in some of the most glorious sound I had ever heard for about an hour, I disassembled the entire contraptions because having +400 VDC on wires running around my room was not the safest way to enjoy music... But it was something of a target to shoot for -- all tube, each stage with its own (tube) regulated power supply, all DC coupled, so no coloration caused by capacitors, and fully balanced from input to output.
All of the products that we have built pretty much follow the same recipe, with one exception. We use transistors instead of tubes. And transistors are available in two (opposite) polarities. This is true for both normal bipolar junction transistors (BJT's) and also field-effect transistors (FET's). The wonderful thing about having opposite polarity gain devices available is that it becomes possible to make a DC amplifier and not end up with hundreds of volts at the output of the circuit! You see, if the first stage is made with N-channel (or NPN BJT's) the output voltage will (typically) be several tens of volts more positive than the input. But then if the next stage is of the opposite polarity (P-channel, or PNP BJT's) then its output voltage will be more negative than the input signal. So if the circuit is carefully designed, the output signal can be at the same 0 VDC reference as the input signal, without the need for any coupling capacitors.
Virtually all companies use capacitors in one way or another to ensure that there is virtually no DC at the output. The problem with that is that there is no such thing as a perfect capacitor, and putting one in the amp (whether it is in the signal path, the feedback path, or in an amplified feedback path -- called a "servo loop" by clever marketing wags) will degrade the sound of the the circuit. So all Ayre products (with two exceptions) are true DC amplifiers, with no coupling caps, no feedback caps, and no "servo loops" (amplified feedback paths). The exceptions are our phono stages, as when you need 80+ dB of gain, there is pretty much no way to make it DC coupled without causing some problems. The other is our AX-7 integrated amplifier which uses a zero-feedback integrated circuit in the input. If all sources were balanced we could direct couple everything. But with single-ended sources the unequal impedance on each phase combines with the input bias current of the BJT's in the input stage to require coupling caps at the input of the amplifier. Our more expensive products use discrete circuits that employ FET inputs, which essentially have zero input bias current and don't suffer from this problem.
Designing and building an amplifier this way (as a true DC amp) is not trivial. It's like the old saying goes, "If it were easy, everybody would be doing it."
This is getting pretty long and my hands are getting numb from typing in my wheelchair, so I will take a break here and pick up this topic again soon.
Ayre Acoustics, Inc.