Jay, I'm disappointed to hear you say that. One of the main problems that computer engineers have is our egos -- we think we're always right and aren't fans of using new or other peoples' ideas. I am often the same way.

Unless you write the code and produce the hardware yourself, the very fact that there are few good assembly language programmers should raise a flag: Who's going to maintain the code after you're gone? Someone's going to have to, and you just said there aren't many people capable of doing it. Your tight, low-cost solution is going to come back to haunt your customers when they can't find you because you've retired to the Bahamas.

Assembly language is great, and I use it whenever I can. Even if someone doesn't use it daily, it's a good foundation for programming in a higher-level language. But to say assembly language is always the solution is like saying that an F-16 is always the best mode of transportation. If you're going from USA to China, I won't argue. But I'll bet I make it to the corner drugstore walking before you even start the engines. :)

In our competitive market, a tight solution isn't always the best, even if it saves 50 cents on unit production costs. Getting the product to market before the competition can be worth much more than 50 cents per unit. And maintainability of the code after-the-fact is worth gold.

My last full-time boss was an assembly language fanatic while others in the department programmed in 'C' and, God forbid, Visual Basic. I later quit and went into consulting. He later quit and is making good bucks too... as a mananger (not program manager) for another company. Assembly language is less in demand every day. We can argue the merits, but that is the ugly reality.

Good luck,
Craig Steiner