further: nonsense: "new" oscilator core

[Martin Czech]

:::Yes , with V1, V2 = 1V. But at the lower side there will be 1mV,
:::so 10% untrimmed error!  Ok with V1max ~ 5V, 1:1000 range
:::and trimming we might get to 30uV/5mV ~ 0.6% error, that's
:::still a lot,and then a lot of triming is required for the
:::reference voltages as well...
:::
:::This could get very tough for low control voltages...

I may add that the lower the frequency, the higher the relative
error, but it translates into beat frequency like:

20kHz, 5V -> 6ppm error (assuming only offset errors) -> 0.12Hz
20 Hz, 5mV -> 0.6% error -> of course also 0.12Hz

this is what it is all about, linear detune :->>

Well, not really, unsymmetry are of course also there.

Of course problems get larger if you try to go deep sub audio,
the oscillator may get stuck then, for very wide range vcos
this kind of linear core is certainly not recommended,

but we knew all this from the beginning.

m.c.