Finding solutions to problems is a large part of the engineering design process. By “solutions”, I don’t mean finding bugs once the design and development are underway; I do mean figuring out the “best” way to accomplish a given task.
That's the design engineer's dilemma. There usually is no single “best” solution or even a single solution; instead, in most cases, there is an array of viable choices. Each brings with it tradeoffs in size, performance, cost, power, reliability, test time, fabrications time, and other design factors.
I sometimes use the example of planning the best route for a space mission from Earth to another planet: Is your priority travel time? Minimizing fuel use? The need to rendezvous at a specific point in space? Or at a specific time? Or is it some combination of all of the these?
That's why navigators will usually say they have worked out “A” solution, rather than “THE” solution to the captain's request. This is not a straightforward problem such as “what is the sum of 1 + 1 but, unfortunately, that's how too many people with limited understanding of the unavoidable nature of tradeoffs perceive design and implementation.
That's one of the many reasons I get so frustrated when critics who know little of what went into design decisions are so quick to demand answers, after the fact. “Why didn’t you do this?” “Shouldn't you have done that?” “Couldn't you have done it this way?”
The answer is “sure, but then we'd have to give up or change other specifications”, which is a fancy way of saying “there is no such thing as a free lunch.” There is a big difference between weighing the tradeoffs and making decisions, versus taking cheap-and-easy shots in hindsight.
Once in a rare while, there are situations where the tradeoff cost is small and clear. For example, boosting your voltage rails to reduce current at a given power level (and thus reduce losses), has only a modest cost, in the need for step-up and step-down stages; that's why nearly every higher-power design uses higher voltages. It's just so much more efficient, that the choice to go with higher voltages/lower currents is almost a “no brainer.”
Sometimes, just sometimes, even nature smiles at us with a no-tradeoff construct. Consider the brachistochrone curve, see here . No matter where you place the ball on the curve, the time it takes for it to reach the bottom is the same. It's somewhat counterintuitive, but it’s both provable in theory and demonstrable in practice. If only more design decisions were like that!
Have you ever faced design tradeoffs that were painful, and were not easily explained? How did you handle the situation? ?