Dispatch
Three chips for a cell was the wrong optimization.
125 plus 138 looked clever until routing. 74251 is the same logical capacity without the chance of contention, and stronger drive than 74151.
In routing the ALU I came across a realization: the 125 + 138 combination defeats the optimization I was looking for. I need three chips for a single ALU cell. The 74251 offers the same logical capacity without the physical chance of contention.
Over the 74151, the 74251 offers stronger drive — which was my initial worry when designing with 74151. Logical changes are none. Routing, area, and performance by bus length are better.
This is the kind of entry the journal is for. The truth table did not move. The board did.