Bit slicing

This article is about processor construction technique. For bit slicing as bit plane separation, see bit-plane.

Bit slicing is a technique for constructing a processor from modules, bit slice processors, each of which processes one bit field or "slice" of an operand. The grouped bit slices then have the capability to process the chosen full word-length of a design. Bit slice processors usually consist of an arithmetic logic unit (ALU) of 1, 2, 4 or 8 bits and control lines (including carry or overflow signals usually internal to the processor). For example, two 4-bit ALUs could be arranged side by side, with control lines between them, to form an 8-bit ALU. A microsequencer or Control ROM executes a program to provide data and control signals. Examples of bit-slice microprocesor modules can be seen in AMD's Am2900 Family or the National Semiconductor IMP-16 and IMP-8 family.

What was the motivation for bit slicing? In the era when these were popular (early 1980's), it was not clear what what was the right width, or whether RISC or CISC would win. 32-bit architectures were being bandied about, but few were in production. 16-bit processors were common but expensive, and the 8-bit Z80 was still selling "like hotcakes". The new concept of RISC processors was being discussed by innovators Patterson and Sequin (RISCvsCISC) -- which one should I build? Bit slice products allowed engineers and students to create their own computers much in the same way a mason could construct a different project every time, even though the building blocks remained the same. The complexity of creating a new computer architecture was greatly reduced when the details of of the ALU were already created (and Debugged) for you.

This article is based on material taken from the Free On-line Dictionary of Computing prior to 1 November 2008 and incorporated under the "relicensing" terms of the GFDL, version 1.3 or later.