The Personal Computer: From Kit to Commodity

Photo: khezez | خزاز / Pexels
In January 1975, a photograph of a metal box covered in toggle switches and LED lights appeared on the cover of Popular Electronics magazine. The headline called it "the world's first minicomputer kit to rival commercial models." There was no keyboard, no screen, and no software for it. Buyers assembled it from parts. Over 10,000 people sent in $397 for one. The personal computer had arrived — ugly, difficult, and completely transformative.
What "Personal Computer" Actually Meant
Before 1975, computers were institutional objects. A university had one. A large corporation leased time on one. The concept of a single human owning and operating a computer was, for most people, somewhere between absurd and alarming.
Mainframe — a large, centrally managed computer system, typically filling a room, shared by many users through terminals. Running a mainframe required a dedicated team of operators and a climate-controlled facility.
Minicomputer — a smaller, cheaper computer introduced in the 1960s (by companies like DEC and Data General) that could fit in a large cabinet and be operated by a small organization without a full data center. Still institutional. Still not personal.
The personal computer changed both words in that phrase. It was personal — owned, operated, and controlled by one person. And it was, by the standards of 1975, a computer: programmable, capable of storing and retrieving data, and running software written by its owner.
Source: Computer History Museum, "The Personal Computer" (2024); Altair 8800 history, Wikipedia.
The Story Behind the Discovery
Ed Roberts and the Altair
Henry Edward Roberts was not a visionary. He was an engineer running a small electronics company in Albuquerque, New Mexico, called MITS (Micro Instrumentation Telemetry Systems), which had made a decent living selling calculator kits before Texas Instruments released cheap calculators that made the kit business uneconomical. MITS was nearly bankrupt in 1974.
Roberts bet the company on Intel's new 8080 microprocessor, which had been released in 1974 for $360. His plan was a computer kit built around the 8080 — a machine that hobbyists could assemble and program. He contacted Popular Electronics, which agreed to put the machine on its January 1975 cover.
Roberts named it the Altair 8800 — a name reportedly suggested by his daughter, who was watching an episode of Star Trek in which the crew visited the Altair star system.
Source: Altair 8800, Wikipedia; MITS history, Computer History Museum.
The machine had no software. Two young programmers in Boston — Bill Gates and Paul Allen — saw the Popular Electronics cover and called Roberts to offer a BASIC interpreter for a machine they had never seen. They wrote it for a simulated 8080 on a Harvard mainframe. When Allen flew to Albuquerque to demonstrate it, neither he nor Gates had ever run their interpreter on a real Altair. It worked. Gates and Allen named their partnership Micro-Soft.
Source: Paul Allen, "Idea Man" (2011); Microsoft corporate history, Microsoft Archives.
The Altair launched the hobbyist computer movement and — crucially — an ecosystem. The Homebrew Computer Club in Menlo Park, California, began meeting in March 1975 to share programs and hardware modifications. Its 500 members included, among others, two young engineers from the Santa Clara Valley.
Wozniak, Jobs, and the Apple I
Steve Wozniak was an engineer's engineer: he designed circuits the way other people did crossword puzzles, for the pleasure of finding the most elegant solution. By 1976, working at Hewlett-Packard, he had designed a complete single-board computer that he showed off at Homebrew meetings. It used the MOS Technology 6502 processor (which cost $25, compared to the 8080's $360).
Steve Jobs recognized that Wozniak's board was not just a clever engineering exercise — it was a product. They built 50 units in Jobs's childhood bedroom, sold them to a local computer shop for $666.66 each, and called the company Apple Computer. The Apple I was sold as a bare circuit board; buyers still needed to add a keyboard and power supply.
Source: Apple I history, Computer History Museum; Walter Isaacson, "Steve Jobs" (2011).
The Apple II, released in April 1977, was the first personal computer designed to be used by someone who was not an engineer. It came in a beige plastic case, had a keyboard, produced color graphics, and could run VisiCalc — the first spreadsheet application, released in 1979. VisiCalc sold the Apple II to businesses: for the first time, an accountant could run projections without waiting for the mainframe department.
Source: VisiCalc history, Dan Bricklin's website (bricklin.com); Apple II specifications, Apple Computer (1977).
| Year | Machine | Processor | Price | Significance |
|---|---|---|---|---|
| 1975 | Altair 8800 | Intel 8080 | $397 (kit) | First mass-market personal computer kit |
| 1976 | Apple I | MOS 6502 | $666 | First complete single-board computer |
| 1977 | Apple II | MOS 6502 | $1,298 | First consumer-friendly personal computer |
| 1977 | TRS-80 | Zilog Z80 | $599 | Sold at Radio Shack stores nationwide |
| 1977 | Commodore PET | MOS 6502 | $795 | First all-in-one personal computer |
| 1981 | IBM PC | Intel 8088 | $1,565 | Established the industry standard |
| 1984 | Apple Macintosh | Motorola 68000 | $2,495 | First commercially successful GUI computer |
IBM's Open Architecture Mistake
IBM was the world's dominant computing company in 1980. Its mainframes generated billions of dollars in revenue through leasing arrangements so complex and vertically integrated that IBM owned the building, the computer, the software, and the service contract. The personal computer market was, in IBM's estimation, a hobbyist curiosity.
By 1980, that view was untenable. Apple Computer had $117 million in revenue. IBM's management formed a secret task force — code-named "Project Chess" — and authorized it to build a personal computer in one year, using off-the-shelf components.
The decision to use off-the-shelf components was pragmatic (it was faster and cheaper) but strategically catastrophic. IBM needed an operating system. Microsoft, which had no operating system of its own, bought a system called QDOS (Quick and Dirty Operating System) from Seattle Computer Products for $75,000 and licensed it to IBM as PC-DOS. The critical clause in the agreement: Microsoft retained the right to license the same operating system — renamed MS-DOS — to other manufacturers.
Source: IBM PC history, Computer History Museum; MS-DOS history, Microsoft Archives; Paul Carroll, "Big Blues: The Unmaking of IBM" (1993).
IBM announced the IBM Personal Computer on August 12, 1981, at a press conference in New York. The machine used an Intel 8088 processor, started at $1,565, and was sold through Sears and ComputerLand. Within a year, it had captured one-third of the personal computer market.
The problem became visible almost immediately. Because IBM had published the technical specifications of its hardware (the "open architecture"), competitors could build machines that ran the same software. Compaq reverse-engineered the IBM PC BIOS legally and shipped a compatible machine in 1982, nine months after IBM's announcement. By 1984, the IBM-compatible PC market was larger than IBM's own PC business — and Microsoft was collecting a license fee from every single manufacturer.
Source: Compaq founding history, Computer History Museum; IBM PC open architecture analysis, IEEE Annals of the History of Computing.
How the Personal Computer Actually Works
A personal computer is, at its structural core, the same architecture described by John von Neumann in 1945: a processor that fetches instructions from memory, executes them, and stores results. What changed between 1945 and 1975 was not the architecture but the cost and size of the components implementing it.
CPU (Central Processing Unit) — the chip that executes instructions. It reads a program from memory one instruction at a time, performs the specified operation (add these two numbers, jump to this address, load this value), and writes the result back to memory or a register.
RAM (Random Access Memory) — temporary, fast storage that holds the program and data the CPU is currently working with. Loses its contents when power is removed.
Storage — persistent memory (floppy disk in 1975, hard disk from the 1980s, SSD from the 2000s) that retains data when power is removed.
Bus — the electrical pathway connecting CPU, RAM, storage, and peripheral devices. The Altair's S-100 bus was the first standard for this; the IBM PC's ISA bus became the industry standard.
The analogy: a CPU is like a chef who can only work with what's on the immediate countertop (registers), who fetches ingredients from a nearby shelf (RAM), and who stores prepared meals in a refrigerator (disk storage). The chef works extremely fast but can only cook what's on the current recipe (program).
What personal computers added to this basic architecture was an interface that did not require the user to speak the machine's language. The Altair required users to enter binary code using toggle switches. The Apple II accepted text commands. The Macintosh responded to a mouse and showed pictures instead of text.
Source: Von Neumann architecture, IEEE Computer Society historical documentation.
Xerox PARC and the Interface That Almost Wasn't
The graphical user interface — windows, icons, a mouse pointer, drop-down menus — was invented at Xerox's Palo Alto Research Center between 1973 and 1975. The Alto computer, developed at PARC starting in 1973, had all of it: overlapping windows, a three-button mouse, bitmap graphics, and a what-you-see-is-what-you-get word processor called Bravo.
Xerox sold none of this. The Alto was never a commercial product. The company's leadership could not reconcile the personal computer with their copier-and-mainframe business model.
In December 1979, Steve Jobs visited PARC with a delegation from Apple. The visit had been arranged as part of a deal in which Xerox received pre-IPO Apple stock. Jobs saw the Alto demonstrated and immediately understood the commercial implications. "Why aren't you doing anything with this? This is the greatest thing," he reportedly told the Xerox engineers. "You're sitting on a goldmine."
Source: Stanford University Libraries, "The Xerox PARC Visit" (primary source account); Xerox Alto history, Wikipedia.
Apple spent four years commercializing what it had seen. The Apple Lisa (1983) was the first commercial attempt — expensive ($9,995) and slow. The Macintosh (January 1984) was the first commercially successful GUI computer, introduced with a single television commercial during Super Bowl XVIII. It cost $2,495 and came with a 9-inch monochrome screen, a 128KB RAM, and a 3.5-inch floppy disk drive.
Source: Macintosh introduction history, Apple Computer (1984); CHM "The Lisa: Apple's Most Influential Failure."
The People Involved
Alan Turing's Conceptual Foundation
The personal computer could not exist without the mathematical framework Alan Turing established in 1936. His paper "On Computable Numbers, with an Application to the Entscheidungsproblem" described a theoretical machine that could execute any computable function given the right program — a concept now called a Turing machine. Every computer ever built is a physical instantiation of this abstraction.
Turing did not build personal computers. He built the conceptual architecture that all computers, personal and otherwise, implement.
Bill Gates and the Software Insight
The insight that would prove more valuable than any hardware design: in a personal computer, the operating system is the platform, and the platform owner captures the industry's economics. Gates understood this in 1980 when he negotiated the MS-DOS deal with IBM. He retained the right to license to others not because he foresaw the IBM-compatible clone market, but because he believed software — not hardware — would be where the value accumulated.
By 1990, MS-DOS was running on approximately 100 million machines. By 1995, Windows 95 had established Microsoft's dominance of the desktop market so thoroughly that the Department of Justice initiated antitrust proceedings against the company.
Source: Microsoft antitrust case history, U.S. DOJ (1998); Gates interview, Smithsonian Computer History Project.
The Engineers Who Built in the Margins
Ed Roberts of MITS, Chuck Peddle at MOS Technology (who designed the 6502), Gary Kildall at Digital Research (who wrote the CP/M operating system that MS-DOS replaced), and Rod Canion of Compaq each made decisions that shaped the personal computer market, usually without recognition proportionate to their impact. Kildall in particular has a legitimate grievance: when IBM came looking for an operating system in 1980, his company Digital Research held the dominant position in 8-bit PC operating systems, but the meeting with IBM went badly (the precise reason is disputed), and Microsoft filled the gap.
Source: Gary Kildall history, Computer History Museum; Paul Freiberger and Michael Swaine, "Fire in the Valley" (2000).
Patent and Priority Disputes
The personal computer industry was built on reverse engineering and standard-setting rather than patent warfare — at least in its early decades. The IBM PC's open architecture meant that anyone with a copy of the technical reference manual could build a compatible machine. Compaq's legal team determined that they could re-implement the IBM BIOS from scratch without using IBM's copyrighted code, and the court agreed.
The more consequential IP dispute was Microsoft's relationship with Apple. Apple's Macintosh GUI drew directly from the Xerox Alto. Microsoft's Windows 1.0, released in 1985, drew directly from the Macintosh. Apple sued Microsoft in 1988, claiming copyright infringement of the Mac's graphical interface. The court ruled in 1992 that most of the disputed interface elements were covered by a 1985 license agreement between Apple and Microsoft, and that the remaining elements were not protectable because they were functional rather than expressive.
Source: Apple Computer, Inc. v. Microsoft Corp., 35 F.3d 1435 (9th Cir. 1994).
Xerox, meanwhile, had sued Apple in 1989 for copying the Alto interface. The case was dismissed because Xerox waited too long to file and the statute of limitations had expired. The company that invented the GUI collected nothing from the GUI revolution.
Unintended Consequences
The End of Typing Pools
In the early 1980s, American businesses employed large numbers of typists — workers whose job was to produce documents on typewriters from handwritten drafts. Word processing software, first on dedicated word processors and then on personal computers running programs like WordPerfect and later Microsoft Word, eliminated this job category within a decade. The Bureau of Labor Statistics tracked "stenographers and typists" as a major occupational category through the 1980s; by the mid-1990s, it had effectively ceased to exist as a distinct labor market.
The Spreadsheet and Finance
VisiCalc, released for the Apple II in 1979, and its successors Lotus 1-2-3 (1983) and Microsoft Excel (1985) changed how financial decisions were made. Before electronic spreadsheets, financial modeling required either a team of accountants or a mainframe. After, a single analyst with a PC could run dozens of scenarios in an afternoon. This capability reshaped corporate finance, investment banking, and eventually contributed to the complexity of financial products (leveraged buyouts, securitized mortgages) that defined late-20th-century capitalism.
Source: Dan Bricklin, "Bricklin on Technology" (2009); VisiCalc historical documentation, bricklin.com.
The Democratization of Publishing
Desktop publishing — the ability to compose pages with mixed text and graphics on a personal computer and output them to a laser printer — emerged with the Macintosh, Aldus PageMaker, and the Apple LaserWriter in 1985. The combined cost of these tools was under $10,000, compared to the hundreds of thousands of dollars required for professional typesetting equipment. Small magazines, community newspapers, self-published books, and corporate newsletters exploded in number and quality.
Where the Personal Computer Stands Today
The traditional PC market — desktops and laptops — peaked around 2011 at approximately 365 million units per year, according to IDC, and has since settled to roughly 250–300 million units annually. Smartphones have captured much of the casual computing use that previously required a PC.
Source: IDC PC market tracker (2024); Gartner PC shipment data.
The most significant architectural shift since the IBM PC is Apple's transition from Intel processors to its own Apple Silicon chips, beginning in November 2020 with the M1. The M1 — and its successors M2, M3, and M4 — are ARM-based system-on-chip (SoC) designs that integrate CPU, GPU, and memory on a single piece of silicon. The M4 Pro, released in 2024, delivers performance in a 14-inch laptop that benchmarks comparably to workstations that cost ten times more a decade earlier.
Source: Apple Silicon transition announcement (November 2020); M4 Pro benchmarks, Geekbench Browser (2024).
The personal computer's conceptual evolution from the Altair's toggle switches to the M4 MacBook Pro represents 50 years of an architecture laid out by John von Neumann and made accessible to individuals by engineers who, in many cases, were simply trying to keep their companies solvent.
What People Get Wrong About the Personal Computer
The most persistent myth is that Apple stole the GUI from Xerox. The real story is more complicated. Xerox gave Apple permission to visit PARC (as part of the pre-IPO stock deal) and demonstrated the Alto interface. Apple then spent four years and enormous engineering effort designing an interface suited to hardware that cost a fraction of the Alto's $32,000. The Lisa and Macintosh interface was not a copy — it was a re-engineering under severe cost constraints. The window management, mouse behavior, and application model were significantly redesigned. Inspiration is not theft.
The second myth: that Bill Gates "stole" DOS from Gary Kildall. The true account is that Digital Research, Kildall's company, failed to close a deal with IBM — the reasons are disputed — and IBM turned to Microsoft. Microsoft bought QDOS from Seattle Computer Products and licensed it to IBM. The OS was not stolen from Kildall, though CP/M's influence on DOS is evident.
A third common error: calling the Altair 8800 the "first personal computer." The Altair was the first personal computer to reach a mass market of hobbyists. The Kenback-1 (1971), the Datapoint 2200 (1970), and the Xerox Alto (1973) all preceded it. What the Altair did that its predecessors did not was catalyze an ecosystem — it created the software, the community, and the vocabulary that defined what personal computing would mean.
Frequently Asked Questions
Q: What was the first personal computer? A: The answer depends on definitions. The Altair 8800 (January 1975) is conventionally cited as the first personal computer to reach a mass market. Earlier machines like the Kenback-1 (1971) and Xerox Alto (1973) preceded it, but neither reached large numbers of individual users. If "personal computer" means a machine designed to be owned and used by one person at home, the Apple II (1977) is probably the better answer because it was the first to be genuinely consumer-friendly.
Q: Did IBM invent the term "PC"? A: IBM called its 1981 machine the "IBM Personal Computer," which popularized the term. But "personal computer" was in use before 1981 — Popular Electronics used "personal computer" in its 1975 Altair coverage. What IBM did was legitimize the concept for corporate buyers who had been skeptical of machines made by hobbyist startups.
Q: Why did IBM lose the PC market it created? A: IBM's open architecture decision made it easy for competitors to build compatible machines. IBM could not legally prevent others from manufacturing IBM-compatible PCs, and Microsoft's decision to license MS-DOS to all manufacturers meant that IBM had no exclusive software advantage either. By the mid-1980s, Compaq, Dell, Hewlett-Packard, and dozens of others were selling IBM-compatible machines at lower prices. IBM's PC division lost money for most of the 1990s and was sold to Lenovo in 2005.
Q: What did Xerox PARC actually invent? A: PARC researchers developed the graphical user interface (overlapping windows, icons, mouse pointer), Ethernet networking, laser printing, the object-oriented programming language Smalltalk, and the WYSIWYG (what-you-see-is-what-you-get) word processor — all between 1970 and 1980. Xerox commercialized the laser printer successfully. The GUI, Ethernet, and object-oriented programming were commercialized by other companies.
Q: How many PCs are sold per year today? A: Approximately 250–300 million traditional PCs (desktops and laptops) ship annually as of 2024, according to IDC and Gartner estimates. This is down from a peak of about 365 million in 2011. If smartphones and tablets are included, the number of personal computing devices sold annually exceeds 1.5 billion.
Q: What is Apple Silicon and why does it matter? A: Apple Silicon refers to Apple's custom ARM-based chips, starting with the M1 in November 2020. Unlike Intel chips, which were general-purpose processors Apple purchased from Intel, Apple Silicon integrates CPU, GPU, neural engine, and memory on a single die (system-on-chip). The result is significantly better performance-per-watt — longer battery life and more performance from a thinner, lighter machine. The transition from Intel to Apple Silicon was completed across Apple's Mac lineup by 2023.
Q: Is the personal computer dying? A: The traditional desktop PC is in long-term decline as a consumer product; smartphones now handle casual computing tasks. But high-performance laptops and desktops remain essential for content creation, software development, gaming, data analysis, and any task requiring sustained computation on a large screen. The form factor is changing — the boundary between a tablet with a keyboard and a laptop is increasingly blurry — but the concept of a personal general-purpose computing device is not going away.
Read next:
Follow new articles
Add SciHub101 to your feed reader to see new science and technology stories.
About the Author
SciHub101 is an independent science and technology publication founded and edited by Buğra Sözeri, a software engineer and app developer based in Türkiye. AI tools assist with drafting and research. Source documentation in the existing archive is uneven, and articles are being reviewed and corrected. The Editorial Team byline identifies editorial responsibility; it does not certify specialist or independent human review. Where we compare products, anything built by our own team is labelled as such — see the editorial guidelines and About page for how corrections and disclosures work.
Related Articles

The Petri Dish: The Simple Invention That Made Modern Microbiology Possible
How Julius Richard Petri's shallow glass dish, Fanny Hesse's agar suggestion, and Robert Koch's lab protocols combined to create the tool behind every bacterial discovery since 1887.

Pasteurization: How Heating Milk Changed the History of Childhood Death
Louis Pasteur invented pasteurization in 1864 to save wine. It took another 50 years to reach milk—and the delay cost hundreds of thousands of children their lives.

How Smartphones Changed the World
In less than 20 years, smartphones transformed how billions of people communicate, navigate, shop, and think.

The Microchip: How Two Men Independently Shrank a Room
Jack Kilby and Robert Noyce independently invented the integrated circuit in 1958–59. Here's how a germanium sliver and a silicon wafer changed everything.

The Invention of the Transistor
How three physicists at Bell Labs in 1947 built a device smaller than a paperclip that would become the foundation of every computer ever made.

Largest Libraries and Archives in the World
The Library of Congress holds over 173 million items and the Internet Archive stores more than 99 petabytes. Here's how the world's biggest libraries stack up.