Thanks for the effort. But I am not sure that the original poster would be happy if he tried to build a cable. They are not simple straight-trough connections between the male and female end.
I have not tested the wiring of my cables but I know for a fact that the mainframe signals an MSIB error (lights I/O check) if you only plug one end of a cable to the in or out MSIB connector. Note that a loop between these connectors can be made without complain (it is meant to test cables).
So I guess there may at least be a wire bridge or resistor between two pins of the same connector.
Or something more clever. Heck, you could even use TDR techniques to detect unterminated lines. :P
Anyway, I'd like to get the whole book you extracted that page from.
I have a 70800B MSIB Cable, but need to make another cable. Then, I disassembled the original cable and confirmed the pinout as below.
1. Pin #16(GND) and #35(CABLE) are tied (jumper) at each connector ends. 2. Cable outer shield is tied to Pin #19(CGND). 3. Other pins are connected in straight-through manner by "twisted pare". Pairing are: (1,20) (2,21) (3,22) ... (15, 34) (17,36) (18,37)
As you can see in the pin assignments, it seems the signals are differencial.
Pay attention to mechanical aspect also. The screws are in metric. Use thin (less than 16.5 mm) connector-cover, or thick cover interfares with neighbor MSIB connector.
My homemade cable does work in 71210C MMS Spectrum Analyser without problem.