There was a time during the late 1980s and much of the 1990s when computers were tremendously exciting because they were getting better at an unbelievable pace. Though from an intentionally comedic perspective, the later portions of Weird Al’s parody song All About the Pentiums captures some of this spirit (link because they don’t allow embed).

You’ve gotta be the dumbest newbie I’ve ever seen
You’ve got white-out all over your screen
You think your Commodore 64 is really neato
What kinda chip you got in there, a Dorito?
You’re usin’ a 286? Don’t make me laugh
Your Windows boots up in what, a day and a half?
You could back up your whole hard drive on a floppy diskette
You’re the biggest joke on the Internet

And again, later:

My new computer’s got the clocks, it rocks
But it was obsolete before I opened the box
You say you’ve had your desktop for over a week?
Throw that junk away, man, it’s an antique
Your laptop is a month old? Well that’s great
If you could use a nice, heavy paperweight

When this song was written we had had about a decade of computers being approximately twice as fast every year, though I think that CD-ROM played a big part in this general sort of impression, too. To see what I mean, I want to run through a very brief history of processors.

In 1982, Intel released the 80286, more popularly called the 286. It was used in a variety of things at a variety of speeds, so for simplicity I’m going to focus on where I met it: the IBM PS/2. It had 1MB of RAM and the CPU ran at 10MHz.

The 286 was succeeded by the 386, though at the time it often took years for these things to work into consumer hardware. We had our 286-based PS2 in about 1988, while the 386 came out in 1995. We got a computer with a 386 somewhere around 1990. That ran at about 20MHz, and was faster at executing instructions even at the same clock speed. The 486 was released in 1990 and started at 25MHz in the computers available to us some time later. In 1992 the DX2 was launched, which was a 486 running at 50MHz. Shortly thereafter a version at 66MHz was launched. These were over twice as fast as the initial 486, since being on the same architecture the clock speed mostly tells you the performance.

In 1993 the DX4-100 as well as the next-generation Pentium were launched, and again performance lept up, somewhere around doubling. The Pentium Pro would follow in 1995 and the Pentium II in 1996. By the time of the Pentium II, not only was the architecture faster at the same clock speed, but clock speeds had reached up to 450MHz (lowest was 233MHz). The Pentium III was released in 1999, and would have clock speeds that, in the high-end models, reached up to 1.4GHz (the lowest was 400MHz). By now there were becoming issues with fetching data from RAM that meant that increased clock speeds didn’t linearly increase processing power, but the increased processing power around now was still almost beyond comprehension compared to what it was a mere ten years before.

However, it was still small compared to how big an improvement CD-ROMs were. To see this, we have to consider what (in a sense) they replaced: the floppy disk. While there were several formats of floppy disk, the dominant one in the mid to late 1980s was the 3.5″ floppy disk. Here’s a decent picture of them from Wikipedia:

(For those wondering about the name, the floppy disk was inside of the hard plastic case. The metal door slid to expose the floppy disk to the drive that read it.)

This stored 1.44MB of data. Hence the Evil Overlord rule, “all important data will be padded to 1.45MB”—so the hero couldn’t copy the critical file onto a floppy drive and escape with it, as often happened in movies from a very specific time period. The 1.44MB 3.5” floppy disks were not overly expensive, but they were not super cheap and they were cumbersome. They also were not fast. It would take close to a minute to read all of the data on a floppy disk, though there was some variability. Software would come in boxes that had multiple floppy disks. Early games might be on three floppies, while later games might be on seven or eight. I remember the first time I played around with installing Linux, it took about twenty floppy disks.

Then some time around the later days of the 486 came the widespread availability of affordable, if still expensive, CD-ROM drives. These had a capacity of 650MB and a read speed of 150 Kilobytes per second, making them almost eight times faster than a floppy drive, but more importantly, they had the capacity to store as much data as 450 floppy drives. Hard drives of the day were often in the range of only 200 or 300 MB, so these amazing disks had as much capacity as two to three hard drives, which were the enormous storage vessels we had been copying floppy disks to in order to install software for years. It was an amazing revolution. Computers were getting faster, but doubling speeds at every 18 months (though it seemed faster at the time) by 1992 they were less than 100 times faster than they were in 1985. With the CD-ROM, we had something that could store over 400 times as much data in a portable format as the year before they were introduced. It was a revolution in what could be distributed with software. Games quickly began shipping with far more art, for more screens, and actual music. I remember getting a demo copy of the entire Encyclopedia Brittanica on a single CD, complete with pictures.

DVDs would come out in 1996, though those were movies that came out and it took a year or so for DVD-ROM to really start being used. They were a revolution in movies for various reasons I won’t get into here, but in computers they only represented an increase in storage, and the change was not nearly as great. A single-layer DVD could store 4.7GB and a dual-layer DVD could store 8.5GB. Taking the bigger of the two (which were common), that was thirteen times the capacity of a CD-ROM. That’s a big improvement, but nothing like the CD-ROM’s 450x improvement over the floppy disk. I think that DVD-ROM also made less of a splash because, while almost nothing fit on a single floppy disk, quite a large number of things fit on a single CD-ROM and even more things fit on two CD-ROMs, which were not much harder to package than a single one. Still, it was more than a ten-fold improvement in capacity.

(As a tangent, where we are on the s-curve of technology can be seen in Blu-Ray disks, which had capacities of 25GB and 50GB for single an dual-layer disks. This represents only a six-fold increase in storage capacity over DVDs.)

So while computers got faster, the explosion in storage, for a time, outpaced the explosion in speed. While we tend to forget how much of a revolution CD-ROM was in its day, I do think that it made a large contribution to the heady sense of technological improvement accelerating almost beyond comprehension.

Of course, these days, a two year old computer is still pretty new and quite capable. I’m buying a new laptop this summer to replace my five year old laptop, but only because I’m tired of the palm-detection not working and because I’d like a bit larger screen. It still has enough storage and is fast enough for what I want to use it for. It’s rare to bother with physical disks for data storage these days (except for buying movies), but while WiFi keeps getting faster, it’s hard to care, even for tech nerds like me, because it’s been plenty fast enough for everything everyone does with it for years now. WiFi 6 is cool and all, but the total amount of time it saves in a day is, perhaps, a few seconds.

This is, by the way, something that people who looked at pure technological capacity tended to miss, back when they expected unbridled growth to continue. Eventually, technology reaches a point where it can do most of what people want it to do, so the additional benefits of improved technology become smaller. At the same time, at some point improved technology becomes more expensive (typically after improvements are figuring out how to do what one is already doing better, and when they come from new, more difficult but now possible ways of doing what one is doing). At some point, the additional benefits aren’t worth the additional cost, and while the technical ability to improve technology exists, the money to do that improvement isn’t there. This is rarely a sudden stopping-point; more often it’s a gradual slowing down of improvements.