Thanks for the information. No, but I thought there was an argument for keeping the INPUT GND out of the feedback signal (which I think is what SIG GND also comprises). I think I even saw some discussions on here previously that tying it into the feedback ground might not be a good idea, since it wasn't a nice looking ground (full of some nasty, albiert smaller, currents). So I thought there might be an argument for keeping it separate as well.
Yep. All that is covered by Douglas Self's article series in Wireless World from the mid/late 1990s, which later became the foundation for his book on power amps. If I recall correctly the article series was called Distortions in Power Amplifiers. You might be able to find it online. You can also see Post #14.
Tom
Tom