Category: Timeline Stories

In early 1978 some of the greatest technology minds of Xerox Corporation PARC in Palo Alto, California—Adele Goldberg (b. 1945) (the initial idea was her), Douglas Fairbairn (b. 1948) (he became a chief hardware designer), and Lawrence “Larry” Tesler (1945-2020), who used to work on the famous Xerox Alto project (which pioneered the graphical user interface), started their work on the Xerox NoteTaker, an early portable computer, which strongly influenced the design of some later computers like Osborne 1 and Compaq Portable.

The Xerox NoteTaker relied heavily on the earlier Dynabook project of Alan Kay (developed in 1972), and just like it, it did not enter production, and only around ten prototypes were built. However, in contrast with the Dynabook, which was a concept for a transportable computer that was impossible to implement with available technology, the NoteTaker was intended to show what could be done.

The NoteTaker computer (see NoteTaker System Manual) weighed some 22 kg (dimensions: 2 1/2 x 21 1/2 x 7 1/2 in) and was built using what was then highly advanced technology, including a built-in touch-sensitive monochrome display monitor, a 340K bytes floppy disk drive, and a mouse. It fitted into a case similar in form to that of a portable sewing machine; the keyboard folded out from the bottom to reveal the monitor and the floppy drive. NoteTaker used a version of the Smalltalk-78 operating system that was written for the Xerox Alto computer.

NoteTaker featured some very advanced hardware: It was battery-powered; had a central CPU with 5 MHz Intel 8086 processor and 4K 16-bit words local memory; a minimum of 128 kB of 16-bit RAM; 7″ diagonal CRT displaying 640 dots horizontally and 480 dots in the vertical direction; a 300 bps modem; a 2″ speaker for audio output; a transparent overlay tablet for pointing on the screen; an analog-to-digital converter with an 8 input multiplexer on the input, and a two-channel digital-to-analog converter; interfaces—Ethernet, EiA, and IEEE bus interface.

The architecture of NoteTaker is such that a small number of processors can be operated in parallel, thus the system offered a useful platform for experimenting with multiprocessor architectures.

The NoteTaker had never been produced commercially, but including so much advanced technologies, it would likely have cost in excess of USD 50,000.

The Xerox Alto computer, developed at Xerox PARC (Palo Alto Research Center) in 1972, was a remarkable machine. The Alto greatly influenced the design of personal computers in the following decades, notably the Macintosh and the first Sun workstations. It was not a commercial product, but several thousand units were built and were heavily used by the US government, PARC, other Xerox facilities, and several famous universities for many years.  The Alto was the first system to bring together all of the components of the modern Graphical User Interface (GUI), and it was the first to implement the LAN technology, named Ethernet. The first WYSIWYG word processor was produced for the Alto.

The Alto was first conceived in a December 1972 memorandum written by the PARC scientist and manager Butler Lampson (see the Memorandum for Xerox Alto), who requests support from the Xerox Corporation for the construction of a number of Alto personal workstations.

Alto was inspired by the On-Line System (NLS) produced by Douglas Engelbart at Stanford Research Institute (SRI), and Alan Kay’s Dynabook, and was designed primarily by Charles “Chuck” Thacker (1943-2017), the project leader. Manufacturing was subcontracted to Clement Designlabs, whose team included Carl J. Clement, Ken Campbell, and Fred Stengel. An initial run of 80 units was produced by Designlabs (initially an Alto cost some $10000 to build, and it was not typically purchased by someone working based on a personal budget), working with Tony Ciuffini and Rick Nevinger at Xerox El Segundo, who were responsible for installing the Alto’s electronics. Due to the success of the pilot run, the team went on to produce approximately 2000 units over the next ten years.

The original Altos (see the nearby image) incorporated:
• Bit-mapped black and white display sized 606×808 (the same dimensions as a regular (8.5″x11″) sheet of paper, aligned vertically)
• 5.8 MHz CPU
• 128KB of memory (at the cost of $4000)
• 2.5MB removable cartridge hard drive
• Three-button mouse
• 64-key keyboard and a 5-finger key set
Development proceeded for the Alto for most of the 1970s, contributing progressive new features in hardware and software. The PARC Altos were connected together in a local area network, using a new networking technology named Ethernet. Early software for the Alto was written in the BCPL programming language, and later in the Mesa and Smalltalk programming languages, which were not widely used outside PARC, but influenced several later languages, such as Modula (see the Users Handbook of Xerox Alto).

The following are a few of the programming facilities and software applications, available for the Alto:
• Many programming languages, including BCPL, LISP, Smalltalk , Mesa, and Poplar
• Bravo and Gypsy—the first WYSIWYG word processors
• Laurel and its successor Hardy—Network E-mail clients
• Markup and Draw—Painting and graphics manipulation (bitmap editors)
• Neptune—File manager
• FTP and chat utilities
• Games—Chess, Pinball, Othello, and a Alto Trek game by Gene Ball
• Sil—vector graphics editor, used mainly for logic circuits, printed circuit
• Officetalk—an experimental forms-processing system
The brain of Alto was a bit-slice processor based on the Texas Instruments’ 74181 chip, a ROM control store with a writable control store extension and 128 (expandable to 512) kB of main memory. It had a hard disk that used a removable 2.5 MB single-platter cartridge (manufactured by Diablo Systems, a company Xerox later bought), all housed in a cabinet about the size of a small refrigerator. The Alto’s CPU was a very innovative microcoded processor, which used microcode for most of the I/O functions, rather than hardware. The microcode machine had 16 tasks, one of which executed the normal instruction set, with the others used for the display, memory refresh, disk, network, and other I/O functions.

The input devices of Alto were a custom detachable keyboard, a three-button mouse, an optional 5-key chord keyset, as well as several other I/O devices—a TV camera, the Hy-Type daisywheel printer, and a parallel port, although these were quite rare. The mouse and chord keyset had been introduced by SRI’s On-Line System; while the mouse was an instant success among Alto users, the chord keyset never became popular.

In the early mice, the buttons were three narrow bars, arranged top to bottom rather than side to side. The motion was sensed by two wheels perpendicular to each other. These were soon replaced with ball-type mice. These were photo-mechanical mice—first using white light and then using IR to count the rotations of wheels inside the mouse.

The keyboard was interesting in that each key was represented as a separate bit in a set of registers. This characteristic was used to alter where the Alto would boot from (it could be booted from either a local disk or the network.) The keyboard registers were used as the address on the disk to boot from, and by holding specific keys down while pressing the boot button, different microcode and operating systems could be loaded. This gave rise to the expression “nose boot” where the keys needed to boot for a test OS release required more fingers than you could come up with. Nose boots were made obsolete by the “move2keys” program that shifted files on the disk so that a specified key sequence could be used.

By the beginning of 1978, Altos were being utilized in four test sites: the White House, the U.S. House of Representatives, the Atlantic Richfield Company, the Santa Clara, California, and the offices of Xerox’s copier sales division. Xerox donated a total of 50 Altos to outstanding universities—Stanford, Carnegie Mellon, MIT, and the University of Rochester, including IFS file servers (the file server was a common application for the machine) and Dover laser printers. Xerox management rejected creating a commercially obtainable version of the Alto for many years. The Xerox Star, the first commercial product to use many of Alto’s ideas was released in 1981, just prior to the first IBM PC, at the cost of $16000, but it was too late.

In December 1979, Apple Computer’s founder Steve Jobs visited Xerox PARC, where he was shown the Smalltalk-80 programming environment, networking, and most importantly the WYSIWYG—the mouse-driven graphical user interface provided by the Alto. He was not impressed by the first two, but was excited by the last one, and promptly integrated it—first into the Apple Lisa and then into the Macintosh, inviting several key researchers from PARC to work in his company.

Lee Felsenstein (born 27 April 1945 in Philadelphia) is an American computer engineer who played a significant role in the development of the personal computer, designing two important machines—Sol-20 and Osborne 1 (one of the first portable computers), as well as another computer hardware, like the Pennywhistle modem (1973).

In November 1975, the electronics engineer Robert M. (Bob) Marsh, who had just founded his own company, Processor Technology in Berkeley, California (to produce ROM, RAM, and I/O cards for the Altair), and Leslie (Les) Solomon (technical editor of Popular Electronics magazine) discussed the just emerged concept of the personal computer. The famous Altair 8800 was just presented (in the January 1975 issue of Popular Electronics) and seemed to be very profitable, although it had a small amount of memory and its programming was an extremely tedious process, as a keyboard wasn’t provided. Solomon agreed to carry a construction article on an intelligent terminal on the cover of the magazine if a working model could be supplied in thirty days.

The work on the project started immediately in Processor Technology, and Marsh ask for advice his friend Lee Felsenstein, whose workshop was in the same room as his.

Lee Felsenstein, a BSc in Electrical Engineering and Computer Science at the University of California, Berkeley, (1972), and a friend of Marsh from the University (Marsh was an alumnus of the class of ’68) and Homebrew Computer Club (an early computer hobbyist users’ group in Silicon Valley, CA), already worked for Processor Technology on the project for building a plug-in video terminal board for Altair. He became the main designer of Sol-20, working in cooperation with Marsh himself (who designed the power supply and the audio cassette interface) and Gordon French (project manager and mechanical designer).

As the design progressed, the team realized that they were building a general-purpose computer rather than just an intelligent terminal, but the decision was made to soft-pedal the fact until the last possible moment. Once published, all the fuss possible was to be made about its general-purpose nature; but until it actually saw print, it was to be treated first as a terminal.

A friend of Marsh could supply walnut side panels for the case, and if the computer was designed low enough, they can use center-cut pieces of wood, which were ordinarily almost thrown away, and thus were almost free to get. The height problem was solved by mounting the expansion boards horizontally rather than vertically (see the nearby image). With only room in the case for five boards, most of the computer functions, including the processor, video, I/O ports, and cassette interface, were mounted on a single large board positioned on the bottom of the case.

When the prototype machine was completed, Marsh and Felsenstein headed for New York to demonstrate it to Les Solomon. The demonstration in the offices of Electronics was successful, in spite of the fact that when the computer was initially turned on, it did not work. Murphy’s Law is confirmed! Felsenstein needed almost a whole day to trace the trouble to a wire, damaged during the transportation to New York.

The machine had been named The SOL after the biblical king Solomon when Felsenstein proposed: “Let’s advertise it as having the wisdom of Solomon” (and yet Lee was born in the “Strawberry Mansion” neighborhood in Philadelphia to Jewish immigrant parents). The computer was launched to the market in June and was presented in the July 1976 issue of Popular Electronics.

The article in Popular Electronics offered a kit version (just like Altair, Scelbi-8H, and Imsai computers) at a relatively low price ($995) and free schematics to all who asked. Processor Technology was soon deluged with orders, and it took almost a year to fulfill them.

The SOL computer was designed in three variations:
1. Sol-PC has a single circuit board without a case or power supply. It was sold as a $475 kit, which was assembled by the purchaser, or fully assembled for $745.
2. Sol-10 was a Sol-PC as the motherboard, but include a case, keyboard, and power supply.
3. Sol-20 was the same as Sol-10, but include a vertical expansion backplane, enhanced keyboard, and a bigger power supply, and cost about $200 more.

The full working version of Sol-20 ($2129) was introduced in August 1976 at the Personal Computing Show in Atlantic City and it became the hit of the show.

SOL-20 became known as the most reliable machine on the market at the time. By 1977, it was the dominant personal computer in the industry. It was in production until 1979, as about 10000 of them were produced (some 5000 as kits, some 5000 as pre-builts) when it has been had replaced by the legendary Apple ][ as leader of the industry.

The processor of Sol-20 was Intel 8080, working at 2 MHz. Internal memory: RAM 1KB up to 64KB max, ROM 1 or 2KB, 1KB video. Display: 64×16 text. External storage: cassette tape, 5.25″ and 8″ disk. Parallel and serial I/O, 5 expansion S-100 slots on the main board. OS options: CP/M, BASIC, NorthStar DOS, and a variety of others. See The System Manual of SOL-20.

Sol-20 was one of the earliest computers to include a keyboard interface and support circuitry for the full implementation of every 8080 function. It was a pioneer in modern video output boards by having a design that actually put up alphanumeric characters on the screen, using a form of distributed processing that didn’t lean on the CPU for all processing.

In July 1974 Dr. Robert Suding (1937-2018), a former Latin teacher and a self-taught electrical engineer, saw the Mark-8 Minicomputer of Titus on the cover of Radio Electronics magazine and then built one. In just a few weeks Suding became a Mark-8 “expert”, who started to design his own upgrades and improvements for the system, including a cassette drive interface and a boot PROM. Suding had one of the few working Mark-8 computers in the Denver, Colorado area, and like-minded hobbyists came from far and wide just to see his system and to get help for theirs.

At the time, Richard Bemis ran The Digital Group Clearinghouse, a Denver-based newsletter for Mark-8 computer enthusiasts. Bemis was impressed with what Suding had accomplished and convinced him that they should join forces and start a company, so in August 1974 Bemis, Suding, and their wives incorporated The Digital Group to market and sell Suding’s improved Mark-8 designs. Bemis was to be the president, and Suding the resident genius and designer, as they eventually sold about 300 Mark-8 kits.

In 1975 Bemis was convinced that The Digital Group (DG) have to launch its own machine, so Suding created a new, reasonably priced, multiple-board computer system with these advanced features:
• Intel 8080, Motorola 6800, or MOS 6500 CPU card
• Video display/cassette interface card
• 8-bit parallel input/output ports
• 2KB dynamic memory
The 3-board Intel or Motorola CPU kit cost $425, while the MOS 6500 CPU kit cost $375. Later, the Zilog Z-80 CPU was added to the selection for $475.

DG computers were among the most advanced microcomputer systems available at the time and were designed to be more user-friendly than other computers. Its products were based on a system of interchangeable boards and components (DG offered both kits and assembled systems, and quite an assortment of peripherals for their systems, including disk drives, a dot-matrix printer, and even a speech synthesizer) which allowed users to upgrade to different CPUs without having to replace their peripherals. Loading software on an audio-based system was as simple as pressing play on a cassette tape player and then the reset button on the system to load a selected application or operating system. Disk system (Diskmon) or data tape system (Phimon) startup was as simple as pressing the reset button, and requesting the desired application by name with the keyboard.

The software applications available were numerous and varied from business applications to hobbies and gaming. The popularity of the DG platform was such that programmers were eager to write applications for it, and the standardization of the hardware allowed for an immediate and large potential customer base.

The DG business was growing fast, perhaps too fast (by 1978 they had 105 employees). Quality suffered, systems were shipped late, upwards of 80% of systems were returned as nonfunctional, and expenses were exceeding income. They continued like this for a year or so, but could not correct or recover from the situation. Popular until the very end, DG went to bankruptcy in August 1979 due to management troubles, not a lack of customer interest or product orders, as it had thousands of product information requests and orders waiting to be filled.

Biography of Robert Suding

Robert Thomas Suding was born on 10 June 1937 in Auburn Heights, Oakland, Michigan. He was the son of a Catholic family from Indiana—August Joseph Suding (1901-1970), an ancestor of German immigrants, and Clara Henrietta Suding (1900-1989).

Robert had been a serious ham radio operator since 1953, always designing & building his own transmitters, amplifiers, antennas, and receivers. In 1955 he graduated from Sacred Heart Seminary (a private Roman Catholic seminary in Detroit, Michigan), and later received MA from the University of Denver, and Ph.D. from Florida State University (1974).

Suding began his career teaching Latin at a junior high school in Michigan but transitioned into a self-taught electrical engineer. In 1967, he left teaching and joined IBM in Denver as a field engineer and became a specialist on the IBM 360/20 computer in hardware and software. Suding worked for IBM until 1975 when he left to dedicate his time to the newly created company the digital group, which he founded together with his colleague Dick Bemis (a software support for IBM in Denver), and their wives (Dick reserved 51% of the shares for himself and generously gave 10% to Suding. He thought this was rather unfair since it was all his designs, but Dick informed him that he was a financial wizard so he deserved 5 times as much as him 🙂 For the next four years, Suding worked on dg microcomputers, usually in his basement. He designed them 16 hours a day, 7 days a week, and got used to having 4 hours of sleep a day for four years.

After the bankruptcy of dg in the early summer of 1979, Suding moved to Virginia and became Chief Scientist at GTE Telenet (an American commercial packet-switched network). Later he started a company called Ultimate Chargers, which supplied equipment for the radio industry.

Suding was known as a mad scientist who passionately invented anything for his hobbies: computers and periphery, amateur radios, binocular telescopes, radio-controlled model airplanes, and pipe organs.

Robert Suding was married to Mary Jane Suding (b. 1938), and they had four children: Linda, Dennis, Ann, and Paul. Dr. Robert Thomas Suding died aged 80, on 6 January 2018 in Conifer, Colorado.

The humble Mark-8 was introduced as the “Personal Minicomputer” and is known as one of the first computers for the home. It was designed in 1972-1973 by Jonathan “Jon” Titus, a Virginia Polytechnic graduate student in chemistry and a computing hobbyist who had the desire to build his own computer. Titus already had a BS from Worcester Polytechnic, and an MS from Rensselaer Polytechnic, and his graduate work was focused on electrochemistry and the development of instrument-to-computer electronics and software.

In 1971, Titus had his first computer experience with a PDP-8 minicomputer from DEC. It was a big and expensive machine, so Jon dreamed—what if I have own computer to work with at home!? Thus the idea for Mark-8 was conceived. When Intel came out with the 8008 microprocessor in 1972, the future looked brighter. Jon procured the 8008 manuals and began conceptualizing a design. He had a computer prototype ready by the fall of 1973. As a grad student, he had no financial means to turn the Mark-8 into a commercial venture, and the idea of establishing a computer company didn’t even occur to him. Jon just wanted to share his design with other hobbyists, so he wrote a couple of letters to Popular Electronics and Radio-Electronics, two well-known hobbyist magazines, asking whether they would be interested in running a how-to-build-it article on the Mark-8. Popular Electronics turned him down, considering the Mark-8 more of an educational project than a truly useful computer, but Radio-Electronics was intrigued.

So, the Mark-8 was introduced as a ‘build it yourself’ project in Radio-Electronics’s magazine July 1974 cover article (see the article), offering a US$5 booklet containing circuit board layouts and DIY construction project descriptions, with Titus himself arranging for $50 circuit board sets to be made by the New Jersey company Techniques for delivery to hobbyists. Prospective Mark-8 builders had to gather the various electronic parts themselves from various sources. About 7500 booklets and some 400 sets of boards were eventually sold.

The Mark-8 was about the size of a large breadbox. It consisted of six circuit boards (CPU Board, Register Display, Output Ports, Memory Latch, Address Latch, and Input Multiplex), one of which held the 8008 and related chips, another the RAM chips, and so on. At the very least, it required memory of eight 256-bit RAMs (in other words, 256 bytes or words), but the memory could be expanded up to 16K by adding more RAM memory boards. There wasn’t any ROM, Jon would have had to pay Intel thousands of dollars to make ROMs for the Mark-8, which meant that every instruction (Mark-8 was programmed in assembly language only) had to be entered by the user and that the programs were lost when the machine was shut off. (RAMs retain their data only as long as the power is on.) In the basic configuration, all those programs had to be entered one bit at a time by flipping a set of toggle switches on the face of the machine—a painstaking and error-prone procedure (although there was a possibility to connect an external ASCII keyboard). The results were displayed on a panel of lights next to the switches.

Biography of Jonathan Titus

Jonathan A. Titus was born in Washington, D.C. in 1945. His father was a lawyer, who was in the Service during World War II in the intelligence branch, doing a lot of code-breaking and deciphering communications. Jon grew up in Huntington, New York, a comfortable middle-class suburb on Long Island’s north shore, where his father worked as an attorney, and his mother was a librarian. Jon has two brothers, both of them younger (Bill is three years younger and Chris is five years younger).

Jon got his start in “computers” by teaching himself about Boolean logic and numbering systems at Harborfields High School in Greenlawn, New York, when he built a 4-bit binary adder and a punched-card reader from his own design, with 24-volt relays and 6-pole switches. Then he and a friend built some relay-based “learning machines”. At this time, his dad bought a Geniac computer, which Jon used to study circuits.

After taking a B.S. in chemistry at Worcester Polytechnic Institute in 1967, Jon went on to Rensselaer Polytechnic Institute, where he picked up a master’s in 1969. Along the way, he became deeply interested in scientific instrumentation. Titus knew a great deal about electronics, he loved to tinker with gadgets in his spare time. In 1978 he got a Ph.D. in chemistry from Virginia Tech.

After Mark-8, in 1975, Jon designed the Mark-80, a computer based on the Intel 8080 microprocessor. Then he and his brother, Dr. Christopher A. Titus, designed several other pieces of computer equipment. In 1976, they designed the “Dyna-Micro” computer, also featured in a Radio-Electronics cover story (Vol. 47, issues May and June 1976). It used an Intel 8080 chip and it had a small keyboard for the entry of octal opcodes.

Jonathan Titus has been a prolific writer on various subjects related to computing. He authored and co-authored over a dozen books. Most of them cover various aspects of designs using Intel 8080 microprocessors. Others are related to microcomputer interfacing and programming. Titus also worked many years as an editor and contributor to several magazines about computers and electronics.

For his contribution to computing with Mark-8 design, Jon Titus was honored in 2002 with The George R. Stibitz Computer & Communications Pioneer Award.

It was in April of 1972 when the USA company Intel Corp. announced its first 8-bit microprocessor, the 8008. In just a few months, the prototypes of the first general-purpose computers powered by the 8008 chip were already working on-site at Réalisations et Études Électroniques in Paris and at Micro Computer Machines (MCM) with headquarters situated on the outskirts of Toronto. So, in the first half of 1973, the first microprocessor-based computers appeared—the French Micral-N and the Canadian MCM/70.

The remarkable MCM/70, a product of Micro Computer Machines, one of three related companies set up in Toronto in 1971 by the entrepreneur and technical wizard Merslau “Mers” Kutt (born 1933), is one of the first microcomputers in the world, the second to be shipped in completed form, and the first portable computer. Kutt, a professor of mathematics at Queen’s University in Kingston, Ontario during the late 1960s, noted that the efficiency of computer users there was hampered by the long wait times involved in submitting programs in punched card form for batch processing by a shared mainframe computer. In 1968, Kutt founded a company and began to produce a data-entry device named Key-Edit. This was a low-cost terminal, with a one-line display device, which bypassed the need for keypunching.

In 1971, Kutt began planning a machine to support software development in the recently developed by Kenneth Eugene Iverson programming language APL. APL was best programmed using a custom keyboard and these were very rare at the time. He initially named his design the Key-Cassette; similar in design and concept to Key-Edit, it would offer editing ability and support for either two cassette decks or one cassette and an acoustic coupler to upload programs to other machines.

In May 1972, a technology development company of Kutt, named Kutt Systems, received one of the earliest SIM8-01 kits, featuring an Intel 8008 CPU, 1KB of RAM, and 2KB of ROM memory. The development team of Kutt Systems started to build what was then termed the M/C (for microcomputer). By then, the design had expanded to include a complete keyboard, a chiclet design similar to the ones used on early models of the Commodore PET, and a Burroughs Self-Scan 32-character display. Unlike the earlier Key-Edit system, the M/C would allow entering and executing APL programs.

One of the early prototypes of MCM/70 was demonstrated in May of 1973, the official announcement was made in September in Toronto. The company maintained that the MCM/70 was “of a size, price, and ease-of-use as to bring personal computer ownership to business, education, and scientific users previously unserved by the computer industry”.

The MCM/70 (see its Users Guide), manufactured by Micro Computer Machines (MCM) in Kingston, was encased in a wedge-shaped metal box about half a meter on the side, with a keyboard at the front, a compact audio cassette tape recorder(s) in the middle, one-line plasma display at the top, and an alphanumeric keyboard. An APL interpreter was built into the read-only memory (ROM), and the machine included a battery which allowed it time to save the workspace automatically when it was turned off. The MCM/70 weighed 20 pounds (9 kg) and shipped with up to 8 kilobytes of RAM and zero, one, or two cassette drives.

The first complete systems were shipped in the autumn of 1974. The basic unit, model 720 with an 800 kHz 8008, 2 KB RAM, and no cassette drive sold for $4950 Canadian dollars. The fully equipped model 782 with 8 KB and two drives was $9800 and was the only model that sold well.

At the time, the machine was already officially being called a “personal computer”. The first manuals contain a note from Kutt to future customers, “But the simplicity of the MCM/70 and its associated computer language… make personal computer use and ownership a reality… Enjoy the privilege of having your own personal computer.” In fact, the MCM/70 was too expensive to become a real “personal computer”, and was sold mainly to companies and government institutions (from hospitals and insurance companies to NASA and the United States Army) with the need to make complex calculations and mathematical analysis. Several hundred units of MCM/70 and its upgraded versions (MCM/700, 800, 900, and 1000) were sold, before ceasing production in the late 1970s.

The Sphere 1 computer, one of the earliest microcomputers, was touted by its creator—Michael Donald Wise (1949-2002), the founder and president of Sphere Corporation, a computer company based in Bountiful, Utah, as the first true PC, because it had a keyboard (with a number pad), a monitor, external storage, and did not run on a punch tape (prior microcomputers lacked the user I/O interface built into the Sphere 1).

The Sphere 1 also included a keyboard-operated reset feature consisting of two keys wired in series that sent a reset signal to the CPU triggering a hard reboot. Wise considered this to be the first keyboard-activated reset, a predecessor to the now-common Ctrl-Alt-Delete key combination. The Sphere keyboard has two reset switches and both had to be pressed at the same time. One was in the upper right of the keyboard and one was in the lower left, thus a reset required two hands.

Sphere 1 (see ad 1, ad 2, and newsletter) was created by Michael Wise (with the help of his colleague, the computer engineer Monroe C. Tyler) in early 1975 and announced in November. Initially, Sphere 1 was sold as a kit but later became available to consumers fully assembled. The machine had a limited run of 1300 units (about half were sold as kits and the remainder were sold assembled), with an initial price of $650 (for the kit), and $1400 (assembled).

The Sphere 1 featured a Motorola 6800 microprocessor, 4KB DRAM (expandable to 64 KB), onboard 1 KB EPROM, a full-sized CRT monitor (16 lines x 32 characters), and a keyboard. The peripheral selection includes floppy disks, printers, paper tape punches and readers, additional terminals, digital I/O, etc. The ROM of a basic system contains drivers, BASIC language, a debugger, and an assembler (just imagine how to fit all this in 1 KB, in fact, its BASIC was very slow). When the disk system is purchased, the user received FDOS (disk operating system with an editor, file structure, and full assembler) (for a full description, see the Sphere 1 brochure).

Michael Wise was an inventor and creative genius, not a businessman. His company began advertising (Wise even announced Sphere 2, Sphere 3, and Sphere 4) before the product was fully debugged in order to finance its growth. Enormous, unexpected demand overwhelmed the company, which was literally killed by success (Michael Wise resigned as president in March 1976 and Sphere Co. entered bankruptcy in April 1977.) Competitors quickly filled the void. Nonetheless, the Sphere had made its mark on the history of the personal computer and contributed to both the specs and design of future generations of hardware. Sphere 1 inspired many copycats.

Biography of Michael Wise

Michael Donald “Mike” Wise was born on 22 June 1949, in Wiesbaden, Germany, to Donald Wise (1918-2007), a Major in the Air Force stationed in Germany, and his wife Bonnie Jean Eacho Wise (1921-1988) (married 1939). He was the third of five children (3 girls—Donna Jean, Edith, and Sandy; and 2 boys—Mike and John Keith) born to the family.

Mike attended 24 schools worldwide before graduating from High School in Brewster, Washington, in 1968.

That summer Mike began college at Brigham Young University in Utah and took a course in BASIC using an IBM 360/50. From that point on he quickly became what we would call a hacker. While in college, he wrote device drivers and two operating systems and also managed to go to school. He also started his first programming business, IPCS, in 1970, developing a mailing list program on an NCR Century 50. Other computers he used in school included the SEL-810b, IBM 1130, IBM 370/65, IBM 650, and IBM 1410.

In 1971 Mike began teaching at the College and was on staff for 18 months before leaving for an engineering company that used the PDP-11 computer in Automated Inventory Control (Robotics) applications. Wise quit two years later to start his own Sphere Corporation (together with Monroe C. Tyler) in 1975.

Later Wise was involved in the development of the TRS-80, Commodore PET, and the Macintosh, as well as in the building and selling of the first screen-based microcomputer WYSIWYG word processor, and numerous other “firsts”. Later he had also a successful career in software development starting A-Systems Co. in 1978, the first company to provide job cost accounting software for the PC, as well as his own Internet business (Splor Co.—Internet’s media and commerce network).

Michael Donald Wise passed away on 28 December 2002 (aged 53), in Salt Lake City, after a lengthy battle with diabetes.

In contrast with the first microprocessor-based personal computer—Micral of Gernelle, and the first microprocessor-based computer kit—Scelbi-8H of Wadsworth, the MITS Altair 8800 was an extremely successful market product. The designer—Ed Roberts intended to sell only a few hundred to hobbyists, but he was surprised when he sold thousands in the first month.

Altair 8800 microcomputer was sold by mail order through advertisements in Popular Electronics, Radio-Electronics, and other hobbyist magazines. Both kits and fully assembled machines were available. Today the Altair 8800 is widely recognized as the first spark, that led to the microcomputer revolution of the next few years, because the computer bus designed for the Altair was to become a de facto standard in the form of the S-100 bus, and the first programming language for the machine was Micro-Soft’s founding product—Altair BASIC.

In 1969 an engineer, working at the Air Force Weapons Laboratory at Kirtland Air Force Base in New Mexico—Henry Edward Roberts (1941-2010), together with three other colleagues decided to use his electronics background to produce small kits for model rocket hobbyists. Therefore they founded Micro Instrumentation and Telemetry Systems (MITS) in Roberts’ garage in Albuquerque, New Mexico, and started selling radio transmitters and instruments for model rockets. Rocket kits didn’t achieve market success, thus later on MITS switched to calculator kits, which appeared to be a more successful venture.

The microcomputer industry really took off when Intel introduced the 8080 CPU in April of 1974. The 8080 processor was capable of addressing up to 64Kb of RAM and was powerful enough to build a real computer. Following the line of several improved models of calculator kits and test equipment, Roberts decided to design an Intel 8080-based computer, and the first prototype was ready in October 1974. At the same time, he was contacted by one of the editors of the magazine Popular Electronics, who knew MITS was working on an Intel 8080-based computer project and thought Roberts could provide the project for the always popular January issue. Thus the Altair 8800 (the name Altair was suggested by the editors, not by Roberts) was born (see the nearby image).

The Altair 8800 was launched at just the right time. There was already a sizable customer base who knew about computers and wanted to have one at hand—schools, colleges, electronics hobbyists, etc. Actually, there were Intel 8008-based computer systems available in 1974, but they were not powerful enough to run a high-level language like BASIC, suitable for non-professionals. The Altair had enough power to be actually useful and was designed as an expandable system, that opened it up to all sorts of applications.

Roberts optimistically told his banker that he could sell 800 computers and he knew they needed to sell 200 over the next year just to break even. To his surprise, when readers got the January issue of Popular Electronics, MITS was flooded with inquiries and orders. They had to hire extra people just to answer the phones. In February MITS received 1000 orders for the Altair. The quoted delivery time was 60 days but it was months before they could meet that. Roberts focused on delivering the computer; all of the options would wait until they could keep pace with the orders. MITS claimed to have delivered 2500 Altair 8800s by the end of May. The number was over 5000 by August 1975. MITS had under 20 employees in January but had grown to 90 by October 1975.

The Altair 8800 computer was very profitable and the expansion bus allowed MITS to sell additional memory and interface boards. Altair used a CPU Intel 8080A (rarely 8080), which worked at a speed of 2 MHz (each instruction takes 4 clock cycles). The RAM provided was only 256 bytes (“1024 word” memory) and you had to buy this memory board. The BASIC language, which was announced in July 1975, required one or two 4096-word memory boards and an interface board to be provided. The computer kit cost $439, and 2 types of memory boards were provided—1024 word Memory Board ($176) and 4096 word Memory Board ($264). Later Roberts offered also a Parallel Interface Board ($92), 2 types of Serial Interface Boards, Audio Cassette Interface Board, and Teletype.

Initially, programming the Altair was an extremely tedious process, as a keyboard wasn’t provided. The user must toggle the switches to positions corresponding to an 8080 microprocessor instruction or opcode in binary, then use the enter switch to load the code into the machine’s memory, and then repeat this step until all the opcodes of a presumably complete and correct program were in place (see the lower image of the front panel). Sounds weird, but when the machine was first shipped, the switches and lights were the only interfaces, and all one could do with the machine was making programs to make the lights blink. Nevertheless, many boxes were sold in this form. Roberts was already hard at work on additional cards, including a paper tape reader for storage, additional RAM cards, an RS-232 serial interface to connect to a proper teletype terminal, a video card, and an 8″ floppy drive that used hard sectored floppies and stored 300 KB.

The January 1975 article for Altair excited a Harvard University undergraduate named Bill Gates, and his good friend Paul Allen, and the duo contacted Roberts to write a BASIC language interpreter for the machine. Roberts shows his interest, but… in fact, Gates and Allen had no BASIC yet to offer. When they called Roberts to follow up on the letter he expressed his interest, the two started work on their BASIC interpreter, using a self-made simulator for the 8080 on a PDP-10 minicomputer. They figured they had only several weeks before someone else beat them to the punch, and once they had a version working on the simulator, Allen flew to MITS in Albuquerque to deliver the program, Altair BASIC (see the reference manual), on a paper tape. The first time it was run, it displayed Altair Basic, then crashed, but that was enough for them to join. The next day, they brought in a new paper tape and it ran (thank God 🙂 The first program ever typed in was “10 print 2+2” and after typing “run” it typed back the correct answer: “4“. Allan was offered a position by Roberts as the Director of Software and the only member of the software department 🙂 Gates, who was then still a student, started working for MITS part-time after he left school. Later, Gates and Allen would leave MITS to begin a company called Micro-Soft.

The January 1975 article would also inspire the creation of the Homebrew Computer Club by a group of Altair 8800 enthusiasts, and from this club emerged twenty-three computer companies, including Apple Computer.

In 1977, MITS was bought by Pertec Computer Corp. for upwards of $6 million, and Roberts retired to a life of vegetable farming in rural Georgia before going to medical school (medicine, but not electronics was his true passion) at Mercer University, where he got a medical degree in 1986. Then he worked as a country doctor in Cochran, Georgia, and died on 1 April 2010.

It was in September 1972, when Nathaniel “Nat” G. Wadsworth (30 May 1943-25 Nov 1998), a 29-year-old electrical engineer, decided to leave his job to create a new kind of computer. Working with a few friends (Bob Findley (who became his main engineering and business partner), Fred Lucas, and Frank Zawacki) in his cottage in Milford, Connecticut, soon he started designing his computer around the new Intel 8008 microprocessor.

Nat Wadsworth show interest in electronics as a teenager, when he was a Ham radio operator since the age of 12, and at 14 made early business assembling electronic kits (heathkits). He dropped from high school and ran away from home at age 17 to join the US Navy, where he served as a shipboard radio operator in the western Pacific. Discharged from the service in 1961, he attended night school, then found a job and entered the University of Connecticut to study electrical engineering. His interest in computers began in 1965 when he worked at the electro-mechanical department of Bunker Ramo, an American electronics company. He worked on DEC PDP-5, and PDP-8 at the end of the 1960s. He even managed to buy his own PDP-8 (at the time only large companies and universities could afford that, but Nat was happy to find a cheap used machine) and wrote a lot of programs in his spare time.

In 1970 Nat got his degree (BSEE) Cum Laude from the University of Connecticut. In the autumn of 1972, he attended a seminar given by Intel Corporation, designed to introduce engineers to Intel’s new 8008 CPU-on-a-chip. He quickly became convinced that he could use the 8008 to replace a great deal of the logic chips he was using in the design of a product underway at the firm where he was then employed as an electronic engineer. His enthusiasm for the 8008 however was not appreciated by management, so he decided to leave and begin his own business.

In 1973 Wadsworth and Findley found the company Scelbi Computer Consulting (an anacronym for SCientific ELectronic BIological). Its Scelbi-8H microcomputer (H standing for hobby, as Nat’s plan was to market his computer at a low price to hobbyists through advertisements in amateur radio magazines) is now recognized as being the first microprocessor-based computer kit to hit the market (the earlier Micral wasn’t a kit, as it was only available in fully assembled form, while Scelbi was available both in kit form and as fully assembled).

The first market announcement for Scelbi-8H was a tiny advertisement in the back of the March 1974 issue of QST, an amateur radio magazine. According to the advertisement, Kit prices for the new Scelbi-8H mini-computer start as low as $440! Actually, with 1K of RAM, the price was some $500. Unfortunately, sometime in the middle of 1974, Zawacki’s and Nat’s close friendship ended, and Zawacki left the company, taking SCELBI’s major investors with him. This affected the company, in that future projects had to be financed by revenues, alone.

Scelbi-8H (see a product brochure) was based on Intel’s first 8-bit microprocessor—8008 (launched in April 1972), the predecessor to the Intel 8080 CPU, used in the Altair 8800. The 8008 was capable of addressing 16Kb of memory and started the design of the first series of microcomputers. The Scelbi-8H had 1K of RAM as a minimum, and an additional 15K of RAM could be purchased for $2760. It had a cassette tape interface, as well as Teletype and oscilloscope interfaces.

After the first advertisement in QST magazine, Scelbi-8H (see the nearby image) appeared in Radio-Electronics and later (September 1975) in BYTE magazine.

Scelbi-8H soon had competitors. In July 1974 Radio-Electronics published plans for a similar 8008 machine, called the Mark-8, that skilled hobbyists could fabricate for the cost of parts. Companies like MITS started selling systems based on more capable processors, such as the 8080 used in the MITS Altair 8800. SCELBI responded by introducing the Scelbi-8B model with 16K of memory (the upper limit of the 8008) and more software available for it (see an ad in Byte magazine).

No high-level programming language was available for the Scelbi-8H in the beginning (see the Users Manual of Scelbi-8H). In 1975 Wadsworth wrote a book, Machine Language Programming for the 8008 and Similar Microcomputers, that taught the assembly language and machine language programming techniques needed to use the 8H. The book included a listing of a floating point package, making it one of the first examples of non-trivial personal-computer software distribution in the spirit of what would much later become known as open source. Because of the similarities between the 8008 and the 8080, this book was purchased by many owners of non-SCELBI hardware. In 1976 Wadsworth and Findley authored a book for computer games.

As the Scelbi-8H did not sell well, it was discontinued by December 1974, and the next year an improved business-market version (named Scelbi-8B), was introduced. Some 200 Scelbi-8B boxes were produced in 1975 and sold at about $580 each, but it also did not become a big market success (as the production cost of the kit was about $1000, so money lost on hardware was recovered through software sales).

Scelbi Computer Consulting discovered that they made more money selling software books than hardware, so by the late 1970’s the company had discontinued making hardware and switched to highly documented software published in book form, including many games, a monitor, an editor, an assembler, and a high-level language dubbed SCELBAL (a dialect of BASIC, that incorporated Wadsworth’s floating-point package), to compete against Altair BASIC.

In 1982, Wadsworth sold the SCELBI publishing business to Hayden Publishing and started exploring pocket computer technology. In 1988, he had the design done for a new pocket computer that he thought would revolutionize the industry. He had parts on hand and was ready to build the first production units when his heart stopped again (he had already survived several heart attacks and two heart surgeries). He survived again, but with heavy damage to his body and mind. On 25 Nov 1998, Nat’s heart stopped again, this time forever.

Wang Laboratories was a computer company founded (with $15000) in June 1951 in Cambridge, Massachusetts, by Dr. An Wang (1920-1990). An Wang emigrated from China in 1945 (he became a US citizen in 1954) and had gotten Master’s and Ph.D. degrees in applied physics from 1945 to 1948 at Harvard. Wang later made some key inventions in the development of core memory technology (the predominant form of random-access computer memory between about 1955 and 1975) and pulse transfer controlling devices (implemented in the Whirlwind computer) and floppy disk drives. At its peak in the 1980s, Wang Labs had annual revenues of $3 billion and employed over 33000 people.

By the mid-1960s, Wang Labs had already made a name for itself in building a series of increasingly sophisticated electronic calculators, such as LOCI, the Wang 300 and 700 families, and many derivative products. Seeing that calculators were getting cheaper and developments in LSI technology would soon make them a commodity item, An Wang decided to develop a general-purpose computer. After several failures, finally he found success with the Wang 2200 computer. Within three years, Wang had sold more than 10000 of the machines (some 65000 systems were shipped in its lifetime), a remarkable success.

The first Wang 2200 (see Wang 2200 A/B Reference Manual) was shipped in May 1973. Over time, various kinds of peripherals were developed, and enhancements were made to Wang BASIC with new microcode.

Build before the era of the widespread use of microprocessors, the Wang 2200 processor consists of a couple of hundred TTL chips spread over half a dozen boards and housed in a heavy steel box. It had a capable BASIC interpreter (written in microcode, there was no machine code that a user could access, unlike microcomputers that would come years later), meaning it could be turned on and used within seconds.

The 64×16 cathode ray tube (CRT) display made editing and running programs interactive and immediate, in comparison with the then-standard method of studying printouts on green bar paper. The 2200 was also expandable; eventually, nearly 100 different peripherals were developed for the system.

Over the years, the 2200 evolved to a desktop computer with an ever-more powerful BASIC dialect, to accommodate multiple users simultaneously, to support up to 16 workstations, and utilized commercial disk technologies that appeared in the late 1970s and early 1980s. New models were produced for nearly 20 years before Wang ended the development.

Biography of An Wang

An Wang was born on 7 February 1920, in Kunshan, Shanghai, China, as the eldest of five children. His mother, Zen Wan (Chien) Wang was a homemaker, his father studied at Shanghai Jiao-Tong University (formerly Nanyang Public School) and his family had been practicing Chinese medicine for generations. An lived in Shanghai with his mother’s family when he was a child, while his father taught English at a private primary school in Kunshan. At the age of six, An moved back to Kunshan to go to school. Because there was no first or second grade in that primary school, he began to study in the third grade. As extracurriculars, his father taught him English, and his grandmother taught him Chinese literature and history. An entered junior high school as the top student in the whole Kunshan district and came to the famous Shanghai high school at the age of 13. In 1936, at the age of 16, he was admitted to Jiao-Tong university to study electrical engineering, graduated with a bachelor’s degree in 1940, then taught there for a year, then worked as an engineer at the Chinese National Government Central Radio Station.

After the war and devastation in China and the loss of half his family, Wang moved to the United States in June 1945 as part of a Chinese government program, to attend Harvard University for graduate school, earning a Ph.D. in applied physics in 1948. After graduation, he worked at Harvard as a research fellow in its Computation Laboratory with Howard Aiken on the design of the Mark IV, Aiken’s first fully electronic computer. At the end of the 1940s, Wang co-invented the pulse transfer controlling device with Way-Dong Woo, a schoolmate from China. The new device implemented write-after-read which made magnetic core memory possible. Wang’s patent (US pat. Nr. 2708722) was one of the most important for core memory and IBM paid him $500,000 in 1955 for rights to it.

Harvard reduced its commitment to computer research in 1951, prompting Wang to start his own engineering business, thus he founded Wang Laboratories in June 1951 as a sole proprietorship. The company became one of the world’s most successful computer companies and by 1984, Wang and his family owned about 55 percent of the company stock, and Forbes magazine, estimating his worth at $1.6 billion, ranked him as the fifth richest American. When Wang looked to retire from actively running his company in 1981, earnings at Wang Labs fell from $210 million in 1984 to $15.5 million in 1985. In July 1985, Wang resumed the presidency and saw profits increase to $50.9 million in 1986. In 1986 he retired again, insisting upon handing over the corporate reins to his elder son Fred (An’s younger son Courtney was a vice president). Hard times ensued for the company and An Wang was eventually forced to remove Fred in 1989. By then the company’s fortunes were already sinking, and Wang filed for bankruptcy in 1992.

Wang is one of the most prolific American inventors (he held 40 patents and 23 honorary degrees) and also was given the Presidential Medal of Liberty by President Reagan. He is considered the creator of word processing as well as a pioneer of the electronic calculator.

In July 1949 An Wang married (second time) Lorraine (Chur) Wang (1920-2016), who was also from Shanghai and immigrated to America in the mid-1940s to do post-graduate work in English Literature at Wellesley College. The family lived in Lincoln, Massachusetts, and had three children: Frederick (born in 1951), Courtney (born in 1956), and Juliette (born in 1964).

In his 1986 autobiography, “Lessons,” Wang attributed his success to typical American business daring, his Confucian values and beliefs, and his skill in being able to “go for a long time without shooting oneself in the foot.”

An Wang, one of America’s wealthiest men, who had given away tens of millions of dollars to charities, died of esophageal cancer on 24 March 1990, in Boston, Massachusetts.