I have a question about the choice of the input impedance of a high frequency amplifier.
In general I know from elementary analog electronics courses that it is good for an amplifier (for instance an Op - Amp) or in general for a voltage meter, to have a very high input impedance. In this way, there would be only a negligible voltage drop on the parasitic resistance of the source, and so no signal loss.
But this seems to me in contrast with the design of RF amplifiers input impedance. For instance, let's consider this circuit of a logarithmic amplifier which works at frequencies between 1 MHz and 8 GHz:
The designer put at the input a parallel resistance (R3) of 51 Ohm in order to realize the impedance matching with the signal generator (which is supposed to have 50 Ohm output impedance) to avoid reflection. But it is completely in contrast with the previous criterion.
I have a similar doubt about the output impedance of a device: in radiofrequency circuits I see that it is chosen to be equal to the input impedance of the following stage, while in analog electronics courses I learnt that it is important for it to be equal to 0 (or anyway it must be very low).