Throughout the 1980s and 1990s, computer power seemed to increase so drastically that pretty much any computer you could own rapidly became obsolete. Such were the strides in available power and performance. (This seems to have tailed off somewhat now.)

I guess this is basically due to Moore's law — the prognostication that the number of transistors per unit die area roughly doubles roughly every 18 months or so.

My question: What happened before microchips? Was the march of progress still as rapid? Was it a bit slower? Or was it a lot slower? Did price/performance ratio improve as much, or was it more a case of companies just building more and more expensive computers?

  • 3
    The 1940s saw the use of machinery to crack the German and Japaanese codes. The 1950s saw revolutions ini surance and air defense. The 1960s saw revlutions in banking and airplane manufacture. None of this would have been possible without the computers of thhe era. Nov 14, 2021 at 23:32

2 Answers 2


TL;DR Progress was much slower before microchips/integrated circuits. In fact, computer technology was so completely hamstrung by available component technology that it only proceeded at a "glacial pace" measured in decades and centuries.

To be historically accurate, you first have to acknowledge that "computing machinery" has been around for many centuries. The vast majority of this history only saw clever innovations in the design of and uses for mechanical computing devices. So, advances in this technology need to be understood as taking centuries, and would not be fully obsoleted until only the most recently concluded - 20th century. It was during this long time-frame that the most important innovation of all actually occurred - this being programmability.

The search for practical electronic components that could be reasonably applied to computing machinery only began around the middle of the 20th century. At this time, progress was more rapid and can be measured in decades, but mainly in terms of better components with which to build the computer. It can be roughly summarized as:

While this work to find a suitable electronic component progressed, the most important gain was in practicality of the computer, and not so much in performance or capacity, which is how Moore's Law is generally observed. Inventing and refining better computer construction components allowed the nascent industry to gradually move from being practical for governments in an existential crisis (World Wat II), to practical for academic research and enormous corporate entities, to practical for regular enterprises with large information processing needs.

Only when integrated circuit technology became the basic component for computer construction did the technology become both practical enough and sufficiently scalable, thanks to the truth of Moore's Law, to initiate the "Computer Revolution" that we mostly think of today.

  • 2
    Going far enough back, and covering a long enough duration, progress may seem like a snail :))
    – Raffzahn
    Nov 14, 2021 at 21:58
  • 4
    Don't forget that "exponential growth" starts off slowly. An analogue of Moore's law was operating during the 1940-1970 timeframe. Nov 15, 2021 at 1:57
  • 6
    @WalterMitty The eerie property of exponential growth is that its speed is always the same, for contemporaries. Only in hindsight does it seem slow ;-). That's one of the reasons why there won't be a singularity -- the e curve simply doesn't have one. It's the same everywhere. Nov 15, 2021 at 5:03
  • Interestingly, your opinion appears to be the exact opposite of Raffzahn's! Nov 15, 2021 at 5:05
  • 4
    From a mathematical perspective (I.e., not with reference to Moore's law or, to anything else in the physical world) Peter is absolutely correct. Exponential growth does not "start out slowly" and then suddenly pick up speed. A graph of the exponential function looks exactly the same at any scale. Change the height scale, change the time scale, change both, doesn't matter. So long as the scales are linear, the curve that you plot always is the same shape. Nov 15, 2021 at 17:32

To a certain extent, yes, there was rapid obsolescence. This is why, for example, many computers were leased rather than purchased.

Two of the computers I used at university, the EE KDF9, and the ICL 1906A, had operational lifetimes by my estimation of about 7 to 8 years. They were both obsolete by the time they were replaced.

Looking at ICL as a particular example, the 1900 range was introduced in 1964. Its replacement, the 2900 range, was introduced in 1974, though of course had been in development for several years before that.

So that gives the range a lifetime of approximately a decade. However, models within the range get replaced at a faster rate. This chart for the 1900 series shows that the market lifetime of a given model was from 2 to 5 years. The newer systems were upwardly compatible with older systems, as a consequence of a well-defined architecture.

The idea of a computer's "architecture" was introduced by IBM in the System/360, an idea latched on to by other manufacturers, and at least meant that a customer could replace an existing machine with a newer and presumably better implementation, without having to rewrite applications. We're used to that today, of course, since Intel has been pumping out variations on x86 for several decades.

Some inertia accrued from the mere fact that the physical replacement of a large machine in an air-conditioned room took weeks if not months.

  • "meant that a customer could replace an existing machine with a newer and presumably better implementation" - and you still can to this day, emulation of such programs on POWER systems remains. Nov 15, 2021 at 15:43
  • The ICL New Range (2900 onwards) is an interesting example of that. But that wasn't so much a thing in the early days, with occasional exceptions such as Honeywell's "Liberator" (IBM 1401 emulator) and the rumoured microcoded emulator for ICL 1900 running on S/360.
    – dave
    Nov 15, 2021 at 15:46

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .