Revision as of 07:02, 27 May 2017

Fairchild 100K (F-100K; 100xxx) was a family of very high-speed discrete logic chips and 8-bit bit-slice chips introduced by Fairchild Semiconductor in the mid-1970s but did not reach high availability until the early 1980s. This series was implemented using emitter-coupled logic (ECL) making those chips considerably faster than comparable Schottky TTL-based chips. The 100K family were an improved version of the 10K which was originally introduced by Motorola and was later also manufactured by Fairchild. Many early high-speed systems and supercomputers made use of those chips.

Overview

In the mid-1970s Fairchild introduced the 100K which was an improved version of the original 10K family. The new family had a lower supply voltage of around ~4.5 V from 5.2 V as well as considerably faster propagation delay (e.g., down to 0.75ns from 2ns). Additionally, the 100K family made use of a larger package (DIP-24 vs DIP-16 for most components).

100K-based parts found their way into many early high-performance systems. For example, the Floating Point Systems FPS-264 64-bit floating-point co-processor which was introduced in February 1985 performed 4 to 5 times faster than its predecessor, FPS-164, by simply switching to 100K series ECL chips from Schottky TTL; this was all done without making any architectural changes and maintaining 100% software compatibility.

Making use of ECL meant system designers were confronted with high power consumption which reduced the usefulness of those chips. It's part of the reason why most systems that used those chips only used them where high speed was vitality needed and even then expensive special cooling was often needed.

Members

This section is empty; you can help add the missing info by editing this page.

Bit-slice parts

See also: bit-slice microprocessor and F220

In April 1980 Fairchild introduced an expansion to the standard discrete logic chips, the F220 (10022x) series of micro-programmed 8-bit bit-slice chips. The sub-family was composed of 5 chips featuring 1,000-gate density and sub-nanosecond delays.

Systems

• 1981: Control Data Corporation Cyber 205 used 100K series chips for its microcode memory logic boards
• 1983: UC Berkeley Big RISC (BRISC) supercomputer was made entirely of 100K parts
• 1985: Floating Point Systems FPS-264 FP Coprocessor
• 1980s: Digital Equipment Corporation VAX 8000 minicomputers used 100K series parts for various components such as the clock phase generator

This list is incomplete; you can help by expanding it.