Why is it needed - inverting switching? There are two reasons: firstly, to get rid of the electrolytic capacitor in the OOS circuit, which has a bad effect on the sound; secondly, to weaken the influence of the imperfection of the input differential stage of the microcircuit (in it, the OOS signal is subtracted from the input signal, and if the differential stage is bad, then the OOS does not work well). In an integrated design, the differential amplifier actually turns out to be very good: due to the fact that transistors located on the chip at a distance of 0.05...0.2 mm from each other have almost identical characteristics, and due to the fact that it is possible not to be afraid to use a good circuit with twenty transistors. However, even with such a differential stage, inverting switching will allow you to squeeze the maximum out of the sound quality, getting rid of all its errors altogether.

Very important! Ground sharing resistor R10 can degrade the performance of the amplifier if it is of the wrong resistance! Constant output, unstable operation, increased noise are signs of incorrect resistance. The most common problems are poor solder contact; wrong resistor (1kOhm instead of 1 Ohm). Quite often it happens that the resistor says 1.5 Ohms, but its real resistance is not that much. Or it was overheated when soldering. The resistor can be replaced with a jumper, this will worsen the sound very, very slightly (and if you’re lucky, it won’t worsen at all, but watch the ground loops in the amplifier in general!), but if the resistance is high, or the soldering is bad, it will be much worse!!!

R10 cannot be set to more than 3.3 ohms.

An important point: input capacitor C2 sets the lower operating frequency of the amplifier at a level of -3 dB. Choose the one you want. At least 5 Hz! But remember that no speaker can reproduce this frequency. And if very low frequencies of even a small magnitude are applied to the speakers (and they are present in a real signal, especially coming from an LP player of vinyl records), then the speakers will be overloaded and create large distortions. So C2 works like a subsonic filter, cutting off those frequencies that are no longer reproduced. Typically, the input capacitor is adjusted to a frequency 2...3 times lower than the actual lower operating frequency of the speakers.