At least we're not still stuck in the days of germanium transistors, where you had to ratchet pliers onto the component legs upstream of the solder joint to stop the junction disintegrating each time you made a joint!
These are actually being phased out of the designs now, as C0G are proving to have better temperature stability and less liable to drift during soldering ;)
To get a quite honest figure for dynamic range (of the phono preamp), you just add the overload margin onto the SNR with the inputs connected to a low-impedance source.
Many manufacturers do this, the idea being to find the highest number possible for a press release.
Yes, at the input only the signal level would be changing, with frequency response and distortion being much the same.
However, depending on the performance of the signal chain, reducing the operating level may also alter sound quality. The preamp you reference uses a valve section and specs...
The sound is changing because you are making slight level changes as a result of the insertion loss varying, with over 3dB for a 5 ohm cartridge brought down from 100 to 10 ohms. This has much the same effect as making minor adjustments to the volume control on your amplifier, with the major...
I wouldn't connect a capacitance meter to the S20's input, as this would probably overload it and therefore give a false reading (or potentially even cause damage to the IC).
I'd use a standard capacitance meter to measure the cable capacitance when disconnected.
You could try RG59 coax with RCA connectors soldered on the ends for 60pF/m. I use RG58C, which is a bit higher around 80pF, or URM76 at 100pF. Thick ground conductors also help a lot with ground-current conversion into common-mode interference.
I wouldn't recommend going over 250pF for the AT...
The LED can be changed out, but generally modification is discouraged. If it gets broken in the process I'll be the one having to deal with it!
TBH, I haven't tested it with the 'line-input energiser' yet...