High damping factor does not have to be a good thing, as it normaly means that there is more negative feedback.
(not saying that this is the case with Audionet amps)
Yes, you are absolutely right. But neither damping factor nor negative feedback are a good or bad thing per se. It depends on the implementation. Damping factor is just an indicator for the amp’s ability to control the woofers.
Ol’ skool says Class A all the way, please no negative feedback. Rationale of Class A is avoidance of crossover distortions through single-ended operation and constant power. Turn-side is the sauna-effect, i.e. most of the energy drawn is dissipated as heat. Not very comfy, not very green. In a push & pull output-stage topology power is generated more efficiently by using multiple output devices in parallel. Also, while a single-ended amp always operates in Class A, a push & pull design can operate in either A or B.
Similarly, while negative feedback can be generally used to reduce distortions, sub-optimal implementation can lead to instability of the amp and cause oscillation. Therefore, the easy answer is to avoid negative feedback. Many great, but traditionally designed amps stick to these two rules as the path to audio Nirvana. It’s a bit like assuming displacement is the only way to create torque in a car engine. In both cases it works, but is a somewhat old-fashioned and inefficient approach (as often observed in US vs. European car engines).
More modern amp designs, like e.g. from Soulution and Audionet, prefer to leverage the efficiency of A/B designs, and are not afraid to use negative feedback to minimize distortions to extremely low levels. In their designs the aforementioned potential problems of using negative feedback are eliminated through their meticulous designs. Distortions are further eliminated through ultra high-frequency designs operating up to 700-800 kHz levels.
In the Humboldt design signals are in addition linearized in real-time at zero crossing, with positive and negative signals separated on component level, even down to separate power supplies for each. There are also no electromechanical (e.g. relais) or passive (e.g. condensator) components in the signal path.
But there are also ways to implement pure Class A more efficiently, as e.g. with the microcontroller based operation approach in the latest Boulder designs, which also adjust the supplied Class A power constantly on a need-to basis. They therefore produce less heat compared to e.g. Pass and Gryphon designs.
Sent from my iPad using Tapatalk
