Re: most efficient RISC instruction

From: Larry Loen <lwloen@us.ibm.com>
> I was only trying to object to the idea someone expressed that RISC
> architecture couldn't do two relevant things in one instruction.  No more,
> no less.

fair enough. It is interesting that all RISC architectures feature
"complexity creep" with time. The original premise behind RISC
was to ruthlessly remove complexity such that each instruction
was as barebones as possible thereby allowing cheap and
fast hardware implementation. Complexity should be done in
software not in hardware. My point was really that RISC
instructions *should* not do more than one thing at a time.