What is a Personal (Digital) Computer?

by voidstar (2021)

There is a lot of opinion on what is a “Personal Computer” and what model/device first represents that definition. These same types of discussion apply to pianos, cars, airplanes; nearly all engineered products end up with these types of debates on which specific model was “first.”

In an interview, Scott Miller (developer of the Kroz game series for MS-DOS) stated he considered the Wang 2200 (c. 1973) as the first personal computer. While it does have all the criteria of such a device (monitor, keyboard, RAM, programmable), it retailed on release for over $7000.

Older folks who were in college in the 1960s would argue that the PDP-1 was the first personal computer. This is valid as well, as a single person could manage operating that system: setting up the input, loading and executing programs, and observing results (in lights, audio, or the 1024×1024 “screen”). But again, this system was over $120,000.

One could argue the abacus or the piano are also forms of “personal computers.” These are more far stretched arguments, but they emphasize a point: many devices have been built that can “compute” (given an input they retain a value, and then provide a subsequent output). So a distinction must be made: older mainframes from early 1960s/1950s were not “personal” since they required many technicians to keep operational, and users were generally not free to just do anything they wanted – the cost of operating the system meant users had focused objectives. In addition, to me, an aspect of a “Personal Computer” is affordability to both the home and (small) office.

Based on my criteria, I consider the Commodore PET as the “first” personal digital computer because of two aspects: single board design and price under $1000 (one led to the other — the low cost wasn’t just cheap hardware, it was a refinement in mass-producing the processor chips that led to the reduction in cost). Even the IBM PC 5100 was an earlier personal computer (c. 1975), but they retailed for over $20,000. This price reflected the fact that these machines used “one-off” components that were difficult to acquire in large quantity. Expansion was limited and maintenance was impractical. They were all impressive machines (such as the Wang 2200 and IBM 5100), but they did not yet represent an affordable household appliance.

Various clever engineers had the foresight to recognized a need for making this type of equipment more affordable. The resulting goal was to create a “single board” computer that was cheaper and easier to build in bulk. The Apple 1 essentially reached this goal in 1976, along with the KIM-1 prototype. Both were still “kit computers,” needing a few additional components to make them part of more useful systems.

Arcade machines of the early 1970s were essentially “embedded processors” with specific hardware and software, and a singular purpose: ROMs that rendered specific graphics, processed specific inputs, and were not general purpose to be re-programmed for other purposes. Like “pocket calculators” of the early 1970s, these devices had a screen, primitive keyboard (or input device, even if it was just a couple buttons), and a central processor and memory.

But the key was simplifying these devices to a single electronics board, to make them easier to produce on a mass scale (to be affordable and practical as a desk-sized appliance). The second key was making them “general purpose” by being re-programmable to serve a variety of purposes (accounting software, cook books, digital globe, flight simulators, word processing, etc.). So two things had to come together: efficient hardware design and board layout, and a concise programming language. BASIC served as that programming language, since early base versions could fit within 2K of ROM.

While the Altair was another kit computer available in 1975, and popular since it was affordable – it was very difficult to operate, giving it very limited utility. Indeed it was a computer, and it could operated by an individual (making it personal, its resources were not shared by other operators or users). But like the Wright Flyer or the early Tricycle like “motor cars”, or the earliest pianos that often suffered from “broken hammers” – for each of these, the technical design needed to be refined for some decades following the initial “proof of concept.”

Another aspect of the “personal” computer was that key concept of being under full control of an individual operator. Around the 1950s – 1960s, computers were expensive and “time sharing” was used to share those hardware resources across users. These was a reasonable concept for many application, but for more intense development no single user had exclusive access to all the available hardware. Moreover, what the machine was to be used for was also dictated by the owner (or sponsor) of that equipment. Making a machine “personal” meant that users had full control over what it is used for.

It wasn’t until 1977 that enough “industry” contacts were in place, to produce a suitable amount of chips (RAM, video ROMs, CPUs, and supporting “system chips”), making affordable personal computer appliances a possibility. The concept of a personal computer was well established, but it required coordinating factories to mass produce the necessary chips to bring the concept to the mainstream consumers.

Two key events also probably motivated this transition to more home/office personal computers: the end of the Apollo missions and end of the Vietnam War. Prior to this, only the government could afford $20,000 “portable” computers with specialized hardware. But the engineering know-how to build these digital computing systems had been established, then becoming more mainstream exactly around 1975. The concept became refined (fitting on a single electronic board, with established expansion port protocols), resulting in the first practical home/office personal computer to be realized in 1977. This exactly fits the profile of Wozniak being the son of an engineer, and Chuck Peddle being a military trained engineer, each working tirelessly to achieve that vision of a practical home computer. The utility of such a machine was then left to others, which led to the rise of many various software shops, each specializing in a variety of types of software application (simulation, games, education, productivity, asynchronous digital communication, etc.).