Jazdoc, Now that we know the Emia is not an LCR type, do you know anything more about the topology of the RIAA that you could reveal without violating any confidentiality agreement with the two designers? I am still wondering how they effect direct coupling and whether the accuracy of the RIAA equalization would be affected as the gain stage tube ages, and its output Z drifts accordingly. I think Hiho said it first; the virtue of the high value resistor at the output of the gain stage in typical RC RIAA networks is that it ameliorates the negative effect of tube aging. No matter; the Emia sounds excellent in the home of my neighbor, and he is thrilled with it.
I thought more about the Allen Wright RTP. As Hiho mentioned, the RIAA equalization depends upon the constancy of the input impedance of that 50K attenuator seen in the schematic. But at the same time, the signal must pass through a 2.2uF capacitor between the plate of the gain stage and the input of the attenuator. Those are at least minor drawbacks. My MP1 has a second gain stage downstream from the dual-differential cascode input stage. THAT second gain stage then drives the attenuator, thus isolating RIAA equalization from the attenuator. However, there's no free lunch; you still need a largish coupling capacitor between the output of the second gain stage and the attenuator. I use a smaller value capacitor (0.68uF) and an attenuator with a higher input Z, 100K ohms, to achieve roughly the same low frequency cut-off.
I thought more about the Allen Wright RTP. As Hiho mentioned, the RIAA equalization depends upon the constancy of the input impedance of that 50K attenuator seen in the schematic. But at the same time, the signal must pass through a 2.2uF capacitor between the plate of the gain stage and the input of the attenuator. Those are at least minor drawbacks. My MP1 has a second gain stage downstream from the dual-differential cascode input stage. THAT second gain stage then drives the attenuator, thus isolating RIAA equalization from the attenuator. However, there's no free lunch; you still need a largish coupling capacitor between the output of the second gain stage and the attenuator. I use a smaller value capacitor (0.68uF) and an attenuator with a higher input Z, 100K ohms, to achieve roughly the same low frequency cut-off.