> It seems that chip is more versatile than I first thought.
It's a special-purpose chip which can be used for a lot of audio.
Of course it is a 9V 1/4W 16r audio power amp for pocket radios. You can even squeeze 0.4W in 8r.
It has low idle power (for pocket battery life), so no big sin to use it for low/no "power output" jobs.
It "is an opamp" but with dedicated NFB resistors. For DC you let both inputs fall to V-, the output will set at half-supply, done. For AC/Audio the naked gain is about 20, extra pins let you set it to 200, a nice range.
Here the Gain pot trims from say 40 to 200.
Drawback in g-world: the input is only 50K, awful low for G-pickups. The "Ruby Amp" adds that JFET for a high input, and a volume knob.
Detail critiques: the 220uFd output cap is for a pocket loudspeaker, and MUCH bigger than needed to slam tube grids and mix-resistors. But hey 220u@16V costs pennies, and they are not my pennies, and using it means it could be re-purposed to drive a speaker without finding a big cap. The <1r output of the '386 will totally SWAMP the output of the tube preamp, we need switching or mixing here, but we can fret that later. The two-12AX7 preamp has far more gain than the '386 preamp; I wonder if the tube-side even needs that much gain (even for insensitive self-split output). So it may be "dangerously high gain", too easy to hit FULL POWER SPLATTT, but I do not think Jericho will fall, and gain-tweaks are easy to do after smoke-test.