Of course, I was seeing preproduction boards.
When I disassembled production models that had junction connections, I knew that the product would have client issues.
This ain't the 80’s anymore with two and four layer boards, that even I could hack and bypass some limitation.
I would think within the chip sets, it wouldn’t be that difficult in place sleeper embedded systems, never used till activated, and nobody would be the wiser.
Exactly, it takes a computer to design the circuit pathway on these multilayer boards and adding a single component can force changes in the positions of other components.
I recall the instructions for some of the add-on boards that were sold back in the 80s: "After being certain you have grounded yourself to the chassis, if you have a Revision A computer board, clip wire B to Leg 4 of the IC chip in the lower left corner of your circuit board as shown in fig. 1; otherwise, if you have a Revision B computer board, clip wire B to Leg 5 of the IC chip next to the second blue capacitor in the middle of the circuit board as shown in fig. 2. If you have a Revision C or later computer board, or there is a wire soldered to either of those legs on either of these ICs, wire B is not required and should be secured so as to not touch anything on the circuit board."
Of course the instructions seldom told you how to determine whether your computers board was revision A, B, C or later, and you spent a good deal of time looking around to find the often hand-inked indicator of which it was. Ah, the good old days of the golden age of home computing.