"No one ever got fired for using Intel..." may be the new motto of our age.
Sure it is. But the saying used to be, "No one ever got fired for buying IBM..." a generation ago when IBM was recognized as a monopoly. It's no longer true for them, and we're on the threshold of the time when it won't be true for Intel anymore, either.
The real issue is that CPU design has shifted to the chip companies. A company without its own massive chip R&D and production facilities (for learning curve reasons) does not have the resources to pioneer new ISAs (instruction set architectures).
...
The trend is pretty clear: the industry-standard architecture wins out *unless* there are compelling reasons for alternatives. And these compelling reasons for alternatives rarely withstand serious scrutiny...and another couple of years of the learning curve at Intel and other chip companies. Any minor performance edges are usually lost when the improvements in the mainstream processors come out.
Fred Weber, VP of engineering at AMD, spoke at Microprocessor Forum a couple of years ago and his thesis was that all of the important features of RISCs had been absorbed into the AMD & Intel CISCs, so the performance advantage of RISC architectures like Alpha no longer existed. Today's implementations of the industry-standard x86 architecture have an efficient RISC microarchitecture under the hood, which can be closely matched to AMD's and Intel's process technologies to obtain high performance. The only real drawback of the x86 ISA is the variable length of its instructions, which make it hard to decode & issue multiple operations in the same clock cycle, but only the first time the code is scanned.
So the question is, why should the industry need a new standard architecture at all? The same economies of scale driven by process technology that allowed the original, rather baroque 8086 architecture to win out over more elegant, orthogonal architectures like the 68000, Z8000, and NS32000 will continue to make x86 (including x86-64) processors have better price/performance than a new-fangled non-standard architecture like the IA-64. Intel has always been successful due to process technology, not innovative computer architecture. |
