QuickStudy: Very Long Instruction Word (VLIW) Microprocessors

When Transmeta Corp., a 5-year-old Santa Clara, Calif.-based start-up in the CPU business, revealed its new Crusoe family of processors last month, experts weren't surprised to learn that the chips are based on Very Long Instruction Word (VLIW) technology.

For one thing, Transmeta's patent disclosures had tipped the secretive firm's hand more than a year ago. But beyond that, VLIW has become the prevailing philosophy of microprocessor design, eclipsing older approaches such as RISC and complex instruction set computing (CISC).

All microprocessor designs seek better performance within the limitations of their contemporary technology. In the 1970s, for example, memory was measured in kilobytes and was very expensive. CISC was the dominant approach because it conserved memory.

In a CISC architecture such as Intel Corp.'s x86, which was introduced in 1978, there can be hundreds of program instructions — simple commands that tell the system to add numbers, store values and display results. If all instructions were the same length, the simple ones would waste memory. Simple instructions require as little as 8 bits of storage space, while the most complex consume 120 bits.

Variable-length instructions are more difficult for a chip to process, though, and the longer CISC instructions are especially complex. Nonetheless, to maintain software compatibility, modern x86 chips such as Intel's Pentium III and Advanced Micro Devices Inc.'s Athlon must still work with

all troublesome CISC instructions that were designed in the 1980s, even though their original advantage — memory conservation — isn't as important.

In the 1980s, RAM chips got bigger and bigger in capacity while their prices dropped. The emphasis in CPU design shifted to raw performance, and RISC became the new philosophy. Examples of RISC architectures include SPARC from Sun Microsystems Inc.; the MIPS Rxxxx series from Mountain View, Calif.-based MIPS Technologies Inc.; Digital Equipment Corp.'s Alpha; the PowerPC, which was jointly developed by IBM and Schaumburg, Ill.-based Motorola Inc.; and Hewlett-Packard Co.'s PA-RISC.

RISC chips use a rather small number of relatively simple, fixed-length instructions, always 32 bits long. Although this wastes some memory by making programs bigger, the instructions are easier and faster to execute.

Because they have to deal with fewer types of instructions, RISC chips require fewer transistors than comparable CISC chips and generally deliver higher performance at similar clock speeds, even though they may have to execute more of their shorter instructions to accomplish a given function.

The simplicity of RISC also makes it easier to design superscalar processors — chips that can execute more than one instruction at a time. This is called instruction-level parallelism, and it's the Holy Grail of CPU architects. Almost all modern RISC and CISC processors are superscalar. But achieving this capability introduced significant new levels of design complexity.