91 plus US minicomputer company attempts from 1968 to 1982*

Appendix of 50 super-minicomputers, "multis", RISC-based architectures, mini-supercomputers, scalable computers, and supercomputers (c1983-1995).

The following list includes all general and special-purpose minicomputers for, real-time, communications, business etc., sold through QEMs, end users, and bundled for process control and testing, for example. It does not include scores of military, AT&T, European, and Japanese computers. At a later time, Tandem formed and array processing systems were developed for niche markets.

The minicomputer companies of the 1980s era characterized their offerings as Super-Minicomputers to differentiate them from their early smaller relatives. In the early 1980s a score of companies started up to build various supercomputers and mini-supercomputers (also called Crayettes after the Cray-style vector design), but all (e.g. Alliant, Convex that was sold to HP, Cydrome, Multiflow ) failed. See the following section.

49 started up and retained autonomy

2 grew at significant rates and continue to grow

8 grew at diminished or declining rates, or found small niches

39 ceased to manufacture

10 started up and merged with larger companies

2  grew at significant rates and continue to grow

2 continued and manufacture niche products for some time

6 stopped manufacturing minicomputers in the merged division

8 existing computer companies built minicomputers

2 made successful minicomputers and grew rapidly

2 continued with diminishing success in minis

4 stopped manufacturing minicomputers

25 existing non-computer companies built minicomputers for backward Integration or special system niches

1 acquired an embryonic company in the design state and formed a division to become a highly successful supplier

3 continued to build and now supply minicomputers for niche markets All of which subsequently disappeared as general purpose supplies

21 discontinued building minicomputers

*Bell, C. Gordon, “Mini and Micro Industries”, Computer, October 1984, pp 14-30. With comments January 2002.

50+ Mini-supercomputers and Scalable Scientific Computers
1983-1995 prompted by Killer Micros and ARPA's SCI (Scalable Computing Initiative)

The startups building a variety of high performance, sometimes scalable, scientific computers covering a wide price range from $50K to $50M that also failed either before or after first product delivery included:

33+ Super-minicomputers including those startups to exploit the RISC ideas, Mini-supercomputers and Personal Supercomputers

10 Scalable computers

7 High Performance Vector Architectures or R&D Efforts