Another thing about the crosstalk of D90SE.
This is crosstalk of D90SE with E1DA @ 4.5 V, 1.66 kOhm:
This is crosstalk of D90SE with E1DA @ 6.7 V, 2.46 kOhm:
The result is the sum of the signal and the noise at the corresponding frequency, so at 4.5 V the graph is slightly lower in the middle.
But after 10 kHz there is a difference which is not the noise, but the signal from another channel.
The recorded test signals were level-matched before analysis.
This is 30 kHz -0.1 dBFS tone in the left channel recorded @ 1.66 kOhm:
(this one is -0.6 dBFS actually because the input were slightly overloaded, but the result is the same)
This is 30 kHz -0.1 dBFS tone in the left channel recorded @ 2.46 kOhm: