and what inputs/outputs it has... From the picture: BT, coax, optical, USB, RCA & TRS.
audiosciencereview.com
Apparently it's not being mentioned on the Amazon page, but it has already been confirmed by Ayima directly.
Dear FMs, There have been few messages shared on new Aiyima A80 in a few different posts, but I found Aiyima A80 on Amazon today and seems like it will be available around Christmas 2024. While @AIYIMA mentioned HERE that A80 has improvements like PFFB, however they haven't mentioned it on their...
www.audiosciencereview.com
Strange. One should think that they would mention it on their own products page. I didn't even look at the Amazon offer. Instead they just comment on ASR. Strange.
Also it seems like the product disappeared from Amazon again?
Could you explain PFFB in a nutshell?
All switching amplifiers (not just the TI TPA chips) need some sort of analogue output filter (a low pass filter) to get rid of the ultrasonic switching frequency, reconstructing a nice and clean signal. A 1st order LP filter consist of at least one coil. In practice the TPA chips and many others need a 2nd order low pass, which consists of one coil in series and a capacitor in parallel with the load, which usually means between the plus (driven output) and minus (signal ground) output terminals.
However, all but the most basic designs use the TPA chips in Tie-Bridged Load (TBL) configuration: Two amplifier sections are fed with opposite polarity signals, one for the "positive" output and the other one for the "negative" output. The result is higher output voltage (thus more power) but there's no common ground anymore. Both outputs are "floating". As a result you need a fully symmetrical output filter, identical for each output, consisting of at least 2 coils and 2 capacitors.
So this is the filter, the first F in PFFB.
The problem with the output filter is that it degrades the quality of the amplifier output in several ways. In particular the coils add ohmic resistance (where we want the output impedance to be ideally zero!) and non-linearities like hysteresis and saturation at the current limit. Capacitors (especially cheap electrolytic ones) have undesireable properties as well, but the coils are even more critical. Good coils are expensive and even the best ones are rather far from the theoretical ideal. So, real world output filters introduce distortion (changing with frequency and with power) and increase the output resistance (also varying with frequency), the latter resulting in the well known "load dependency" with real world speakers (where the impedance varies with frequency, too).
The engineer's primary weapon fighting non-ideal behaviour of amplifiers is: negative feedback (FB, the last two letters in PFFB). Feedback takes a portion of the output signal and feeds it back to the input in a way that deviations from the input signal are reduced. How much reduction is possible depends on the available open loop gain (with no feedback) of the amplifier itself. The TPA325x chips being specialised audio amplifier chips, not general purpose op-amps, don't have too much gain, but in 2017 TI published a white paper that demonstrated how the non-ideal properties of the amplifier chip including its output filter could be reduced by creating a feedback loop from behind the output filter (Post Filter, PF) to the chip's input. This white paper has been revised in 2018 but it still took a couple of years before the first commercial implementations hit the market.
It's a tradeoff, still. Following the recommendations in the white paper you loose about 6-7 dB of amplifier gain to improve distortion and output resistance by a comparable margin. Either you live with the lowered gain or you need an input buffer stage that makes up for the loss. This amplification stage must be of high quality, of course, so it doesn't reintroduce the amount of noise and distortion that's just been gained on the output side. However, the tradeoff seems to work out quite well. There's no easier way to correct for the deficiencies in output filter coils and capacitances.
Edit:
The nutshell turned out to be a windjammer ...