In June 1993 the first TOP500 list of the world’s fastest computers was published. The top four machines on it were all built by the same company, and so were five of the top ten. Fourteen months later that company filed for bankruptcy protection.
Thinking Machines Corporation built one of the most striking computers anyone has made, a black cube filled with blinking red lights that went into museums and into Jurassic Park. It employed Richard Feynman, on and off, for five years. It was, briefly, the fastest thing in computing. And it never really had a market, which is the whole story.
What follows is assembled from federal audit reports, bankruptcy filings, the company’s own hardware manuals and the recollections of the people who were there. Thinking Machines was privately held and never filed public accounts, so the revenue and profit figures below are the company’s own statements as reported at the time and cannot be checked against an audited filing. Figures taken from the federal audit are identified as such.
A machine with sixty-five thousand tiny brains
Thinking Machines was founded in May 1983 and set up in an old mansion outside Boston, the Robert Treat Paine estate in Waltham. The architect was Danny Hillis, whose doctoral work at MIT was the machine itself. The company and the dissertation were built in parallel, which is why the thesis reads like a design document.
One correction before going further, because it appears in nearly every account. Hillis’s doctoral supervisor was Gerald Sussman, not Marvin Minsky. Minsky was deeply involved in the company’s formation and recruited Howard Resnikoff as a research director, and accounts differ on whether to call him a founder, but he did not supervise the thesis. The front matter says Sussman, on both the signature page and the abstract.
The idea was to invert how computers work. A conventional machine has one fast processor that fetches data, operates on it and puts it back, over and over. Hillis proposed giving every piece of data its own processor, so that instead of one clever thing working very fast you had an enormous number of stupid things all working at once.
The CM-1 as built carried 65,536 processors, each one bit wide with 4,096 bits of memory to itself, running at four megahertz inside a cube about 1.3 metres on a side and drawing roughly twelve kilowatts. Sixteen processors shared a router chip, and the 4,096 routers were wired as a twelve-dimensional hypercube, so no router sat more than twelve hops from any other.
That last detail is worth stating carefully, because it is routinely mangled. The processors are not the corners of the hypercube. The router chips are. Descriptions of a twenty-dimensional machine refer to a million-processor design that was never built, and it survives in the literature only because of a story about Richard Feynman.
Richard Feynman and the router
Hillis wrote the definitive account himself, in Physics Today in February 1989, a year after Feynman died. It is the source for everything in this section and it is worth reading in full.
Hillis mentioned over lunch that he planned to start a company to build a parallel computer with a million processors. Feynman’s response, in Hillis’s telling, was “That is positively the dopiest idea I ever heard.” For Feynman a crazy idea was an opportunity to prove it either right or wrong, and by the end of lunch he had agreed to spend the summer at the company.
Told his job would be advising on the application of parallel processing to scientific problems, he said that sounded like a bunch of baloney and asked for something real to do. He was sent out to buy office supplies. While he was gone the team decided the part of the machine they were most worried about was the router, and when he came back with the pencils they gave him the router to analyse.
By the end of that summer he had an answer, and he presented it as a set of partial differential equations. Treating a set of boolean circuits as a continuous differentiable system is a natural move for a physicist and a very strange one for a computer designer. His variables were things like the average number of one bits in a message address.
Five buffers or seven
The company’s own discrete analysis said the router chip needed seven buffers. Feynman’s equations said five. They decided to play it safe and ignore Feynman. That was September, and by the following spring the chips they had designed were slightly too large to manufacture, and the only way to fix it was to cut back to five buffers.
He was right. The machine worked, and the first program it ran, in April 1985, was Conway’s Game of Life. Feynman tested the machine against quantum chromodynamics, modelling a section of space-time as a discrete four-dimensional lattice, and since the only language he really knew was BASIC he invented a parallel version of it and simulated the program by hand to estimate the speed.
The quantum computing claim, and why it does not hold
In a 2016 oral history interview with the Computer History Museum, Hillis said that the project Feynman was working on was quantum computing, and that he was quite sure Thinking Machines was the first company ever to work on it. For a publication about quantum computing that is a tempting claim, and it does not survive checking.
Hillis’s own account, written in 1989 while the memory was fresh, lists what Feynman did at the company. The router, Hopfield associative memories, a logarithm algorithm from his Los Alamos days, quantum chromodynamics, cellular automata and fluid simulation with Stephen Wolfram, and simulated evolution using Fokker-Planck equations. Quantum computing is not mentioned once.
The dates settle it. Feynman’s founding paper of the field, Simulating physics with computers, appeared in the International Journal of Theoretical Physics in June 1982, roughly a year before Thinking Machines was incorporated, and it came out of a 1981 conference at MIT. His quantum computing work was not Thinking Machines work.
The truth is better than the claim anyway. The man who wrote the founding paper of quantum computing spent five of his last years hand-simulating boolean circuits and writing a parallel dialect of BASIC for a very classical machine.
The lights, and what they actually did
The Connection Machine is the only supercomputer most people can picture, and it is because of the lights. The usual assumption is that they were decoration. The usual correction is that they were purely functional. Both are wrong, and the documentation is unusually clear about it.
Status lights are ordinary things to put on a circuit board, showing whether power is on or a chip is seated. Carl Feynman, Richard’s son and an employee, described a fantasy of the machine as a vast cloud of lights flickering as they sent messages, like neurons firing. The designer, Tamiko Thiel, made those board-level lights visible through the skin of the cabinet.
So they were real status indicators, deliberately exposed. And a program could seize them. The CM-2 instruction set contains an operation that determines whether the lights display internal status or are driven by the user program, and another that latches an arbitrary pattern onto them. A machine with nothing to do went dark, which is why finished programs sometimes carried extra operations purely to keep the face blinking for visitors.
The CM-2 that carried them arrived in April 1987, not 1986 as is often written. It kept the 65,536 processors but gave each one sixteen times the memory, 512 megabytes in total, and added an optional floating-point accelerator using Weitek chips. Alongside it came the DataVault, five gigabytes spread across thirty-nine disk drives. Each 64-bit chunk arriving from the machine was split into two 32-bit words, and each of those took seven error-correcting bits to make 39, one bit to each drive, with three spare drives standing by. It is an early commercial disk array, and Thinking Machines called rebuilding a failed drive healing the database.
Who was actually buying
This is where the story turns, and the source is a federal audit rather than a memoir. In May 1993 the General Accounting Office reported to the House Armed Services Committee on the high-performance computing programme run by ARPA, the Pentagon’s research agency.
The report is blunt. ARPA had facilitated the placement of 44 systems made by Intel and 24 made by Thinking Machines, and had not supported procurements from several major rivals with which it had no development contracts. Government officials, researchers and vendors told the GAO that this had distorted the market for massively parallel machines. The excluded competitors are named as Kendall Square Research, nCUBE and MasPar.
Adding up every disclosed cost in the GAO’s placement table gives about $30.2 million across four funding sources, ARPA, the Navy, NASA and the Strategic Defense Initiative Organization. That is a floor rather than a total, because several rows are grouped and some show no figure at all, and it covers procurement rather than research funding. A figure of $55 million circulates widely; it traces to a single magazine article citing a 1991 newspaper investigation, and it should be attributed rather than asserted.
What the inquiry actually found
One founding detail explains a great deal. Among the co-founders was Marvin Denicoff, who until August 1983 had directed the Office of Naval Research’s Information Sciences Program. He is recorded as a co-founder in 1983 and as having directed that programme until August 1983. Nothing in the audit connects him to any particular procurement decision, none of the sources alleges he did anything improper, and we make no such suggestion here.
It is often implied that an investigation caught ARPA and Thinking Machines doing something improper. It did not. The GAO made three headline management recommendations, that the placement programme be broadened to more vendors, that a database of project status be published, and that more emphasis go on system software. There was no finding of illegality, fraud or breach of procurement regulations, and ARPA officials generally agreed improvements could be made.
The GAO also noticed something the company would rather it had not. The CM-2 and CM-200 machines being placed were, in its words, essentially refinements of the original CM-1 design rather than significant departures from it. A programme meant to get novel prototypes into researchers’ hands was buying updated versions of the same computer.
The CM-5, and the quiet surrender
The CM-5 was announced in October 1991 and it abandoned almost everything that made the Connection Machine distinctive. The custom one-bit processors went, replaced by ordinary SPARC microprocessors of the kind Sun put in workstations. The hypercube went, replaced by a fat tree. The operating system was a version of SunOS.
Thinking Machines never framed this as a retreat. Its own technical summary calls the CM-5 a universal architecture combining the best features of existing designs, able to run nodes in lockstep or independently as the problem required. The marketing answer was that it was neither one thing nor the other but both.
Judge it by what it did rather than what it said. The company that had bet everything on custom silicon was now buying processors from Sun, and the technical summary names the topologies it had left behind, listing the hypercube and the two-dimensional mesh used in the CM-1 and CM-2 as the designs its fat tree improved upon. Internally the switch reportedly took eighteen months of argument before Hillis came round.
A node could be fitted with four vector units alongside its SPARC chip. By the November 1993 technical summary those gave 160 megaflops of peak 64-bit floating point and 640 megabytes per second of memory bandwidth, up from the 128 and 512 quoted at the October 1991 announcement. The vector units could not fetch their own instructions; the SPARC issued to them. The technical summary calls a single node with vector units a supercomputer in itself, which was true and also the problem, because so was a rack of workstations.
Fourteen months
In June 1993 the first TOP500 list appeared. A CM-5 with 1,024 nodes at Los Alamos was first in the world at 59.7 gigaflops measured. A 544-node machine at the Minnesota Supercomputer Center was second, a 512-node machine at the National Center for Supercomputing Applications third, and FROSTBURG at the National Security Agency fourth. A 256-node CM-5 at the Naval Research Laboratory was seventh.
Thinking Machines filed a voluntary Chapter 11 petition on 17 August 1994, in the United States Bankruptcy Court for the District of Massachusetts. When the resulting lease dispute reached the First Circuit the following year, Judge Bruce Selya noted drily that the environs of the company’s Cambridge building were not sufficiently conducive to fertile thought.
The numbers behind the filing tell the real story, and they are the opposite of the one usually told. Computergram International, reporting on the day of the filing, had the company losing $20 million in 1993 on sales of $82 million, and last making a profit in 1990, when it earned $1 million on sales of $65 million. Peak revenue was therefore 1993, the year of the record loss, not 1990. The same report gives 425 employees, about 140 of whom went at the filing, 112 machines in the field, and six months of unpaid rent on the building.
Why it failed, according to the people who watched
The same economics that ended Seymour Cray‘s companies were at work here. Gordon Bell put it most directly, saying the company had extrapolated from a few sales directly related to government placements that there was a market for huge processing machines, and that there never was a real market there. Lew Tucker, one of the company’s research directors, told Gary Taubes of Inc. that its charter had not been to look at a machine and work out the commercial profit, but to build an interesting machine.
The GAO had named the deeper problem a year before the filing, reporting that while much progress had been made in hardware, software remained too primitive to make massively parallel systems useful. A 1990 assessment by Robert Schreiber, at the NASA Ames institute that ran the agency’s supercomputing research, was harsher still about the hardware, concluding that the floating point had been strapped on rather than designed in, and that without a full custom implementation future Connection Machines would not keep pace with microprocessor-based machines.
Then there was the bet. In 1990 Gordon Bell wagered Danny Hillis that in the last quarter of 1995 the highest sustained performance would come from a machine with fewer than a hundred processors rather than one with more than a thousand. The judges were John Hennessy and David Patterson. By 1996 Hillis was saying that from a changed market standpoint they had both lost, and that the supercomputing market basically did not exist any more as a definable market.
Where the pieces went
The company reorganised in October 1994 as a software-only business and reportedly ran four profitable quarters in that form. The hardware was simply discontinued. Nobody bought the Connection Machine line, which is worth stating plainly because the usual account has Sun buying the hardware business in 1996.
What actually happened, according to trade reporting at the time, is that Sun hired Thinking Machines hardware engineers from late 1994, and then in late 1996 agreed to buy the company’s GlobalWorks assets for an undisclosed sum. GlobalWorks was software, a parallel computing environment that ran on clusters of Sun servers. The distinction matters because it is the difference between a company being acquired and a company being picked over.
The last valuable thing was Darwin, a parallel data-mining toolkit that American Express had been using to sift credit card transactions on a Connection Machine. Oracle bought it in 1999, for a price nobody disclosed. The patents went to a partnership formed under the plan of reorganisation, which then sued IBM. We could not establish how that case ended, so nothing should be read into it either way.
The people turned out to be the asset. Hillis went to the MIT Media Lab, and later to Disney. Others went to Sun and helped turn a workstation company into a web company. Brewster Kahle built the WAIS system at Thinking Machines and went on to found the Internet Archive. Hillis’s own observation is that the best investment anyone could have made would have been to back everybody in that building, whatever they did next.
Asked about the collapse twenty-two years later he said he had been a bad business person who made a lot of stupid mistakes, and that had he known a quarter as much then as he did later, he could have kept the company alive long enough for the web to arrive.
The Jurassic Park question
Connection Machine cabinets stand in the control room in Jurassic Park, in place of the Cray specified in Michael Crichton’s novel. Whether they were working CM-5s or empty front panels wired for their lights is not settled, and we are not going to pretend otherwise.
The film’s special effects coordinator said everything in the set was real, but he was itemising hardware loaned by Silicon Graphics and Apple and did not mention Thinking Machines at all. The claim that they were empty panels traces to a forum recollection and an uncited encyclopaedia edit. Both accounts are weaker than their confident retellings suggest.
Questions and answers
What happened to Thinking Machines Corporation?
It filed for Chapter 11 bankruptcy protection on 17 August 1994, fourteen months after its machines held the top four places on the first TOP500 list. It reorganised as a software company in October 1994, sold its GlobalWorks software to Sun Microsystems in late 1996, and sold its Darwin data-mining product to Oracle in June 1999. The Connection Machine hardware line was discontinued and nobody bought it.
What was the Connection Machine?
A massively parallel computer that gave every piece of data its own processor. The CM-1, announced in 1985 and shipping the following year, carried 65,536 one-bit processors with 4,096 bits of memory each, sixteen to a router chip, with the 4,096 routers wired as a twelve-dimensional hypercube. It inverted the usual arrangement of one fast processor working through data one item at a time.
Did Richard Feynman work on quantum computing at Thinking Machines?
No. Danny Hillis said in 2016 that he did, but Hillis’s own account written in 1989 lists Feynman’s work there as the router analysis, Hopfield networks, a logarithm algorithm, quantum chromodynamics, cellular automata and simulated evolution, and never mentions quantum computing. Feynman’s founding paper on the subject was published in 1982, about a year before the company was incorporated.
What did Feynman actually do at Thinking Machines?
He analysed the machine’s router and delivered the answer as a set of partial differential equations. His analysis said five buffers per chip were enough where the company’s own said seven. They ignored him, then discovered their chips were too big to manufacture and had to cut back to five. He was right.
Were the blinking lights on the Connection Machine real?
Yes, and they were also programmable. There is one light per sixteen processors, so 4,096 on a full machine, and by default each reports internal status. The instruction set includes operations to switch them from status display to user control and to latch an arbitrary pattern onto them, so a program could drive them deliberately.
How much money did DARPA give Thinking Machines?
No consolidated public figure exists. The one official tabulation, a 1993 General Accounting Office report, lists 24 Connection Machines placed with about $30.2 million of disclosed cost across ARPA, the Navy, NASA and the Strategic Defense Initiative Organization, and that is a floor rather than a total. The $55 million often quoted traces to a single magazine article and should be attributed rather than stated.
Did an investigation find that DARPA acted improperly?
No. The GAO found the placement programme too narrow, reporting that ARPA had placed 44 Intel systems and 24 Thinking Machines systems while not procuring from their rivals, and that observers said this distorted the market. It made three management recommendations and found no illegality, fraud or breach of procurement regulations.
Was the Connection Machine in Jurassic Park real?
Nobody can currently say. CM-5 cabinets appear in the control room, replacing the Cray specified in the novel. The film’s special effects coordinator said everything in the set was real, but he was listing loans from Silicon Graphics and Apple and did not mention Thinking Machines. The claim that they were empty light panels rests on a forum recollection and an uncited encyclopaedia edit.
Who founded Thinking Machines?
Danny Hillis, whose MIT doctoral work was the machine itself, with Sheryl Handler as chief executive and Marvin Denicoff among the co-founders. Denicoff had until August 1983 directed the Office of Naval Research’s Information Sciences Program. Marvin Minsky was closely involved and recruited the research director, but he was not a co-founder and did not supervise Hillis’s thesis, which was supervised by Gerald Sussman.
The machines, in order
Specifications below come from the Thinking Machines technical summaries where one survives. The CM-200 is included for completeness, but no manufacturer documentation for it could be found, and the figures usually quoted for it are unsourced.
| Machine | Introduced | Processors | Interconnect | What it introduced |
|---|---|---|---|---|
| CM-1 | announced 1985, shipped 1986 | 65,536 one-bit | 12-dimensional hypercube | Data parallelism as a product, one processor per data item |
| CM-2 | April 1987 | 65,536 one-bit | 12-cube with message combining | Sixteen times the memory, optional floating point, the DataVault |
| CM-200 | 1991 | not documented | as CM-2 | A faster CM-2. The GAO called it a refinement, not a departure |
| CM-5 | October 1991 | SPARC nodes with four vector units | 4-ary fat tree | Commodity processors, a Unix derivative, and no more hypercube |
See today’s quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals.
