Apple’s second computer — its first to have a case — launched in 1977, and that boxy beige Apple II was soon everywhere: in classrooms, living rooms and offices. At the vanguard of a generation of personal computers to come, it featured a particular and carefully-chosen beige. But what did that look like? Those first machines — the ones that have escaped landfills anyway — have shifted in color over 40 years. The documented public record is sketchy and confused. But I stumbled upon a way to investigate what Apple Beige was like.
Fascinating bit of sleuthing, and a fun read to boot. Maybe not the most important aspect of computer history, but every bit of information we can preserve is worth it.
I suppose most people are more familiar with film restoration and art restoration. I’ve watched a few youtubes on this and they are fascinating. Computers are the antiques of tomorrow so I suppose this exercise is valid. As for whether in time computers will reach the revered status of items from antiquity I have no idea. Maybe time to begin laying down a few boxed Oric 1’s for your great great great grandchildren.
Good article, makes me want to research again what colors can be expected to last longer, but I guess that’s out of scope for OSnews :-).
Pantone books have a shelf life, they can’t be trusted beyond a certain age as colors will change randomly as certain pigments fade or change hue.
I have a Gretag–Macbeth colourchecker long past its use-by date. They wear out too. While IHVs may not calibrate every camera or display or printer out there thankfully now manufacture tolerances are now so reliably good and some reviews include colour profile data it can be a moot point for most people.
I don’t know if it’s possible but if someone out there took a properly calibrated picture with some handwaving would it be possible to get an accurate fix on the colours of things in the picture? You would need to know if they used a colour checker and what film they used and what filters to adjust white balance and stuff but would it be possible?
HollyB
Not trivial questions, of course this is the difference between a technical/scientific image and a retail/public image! I have a number of spectrophotometers of various ages, all of these age as well, even the ceramic tiles supplied with those devices to calibrate white balance surprisingly discolour over time.
There are some techniques in astronomy that can be used to calibrate and identify wavelengths, but I’m not sure if they can be applied retrospectively, and often they are only used relative to some surface or reference in the same image. Which is the technique they use to get around absolute calibration, they define certain stable object as the reference. Which is the same technique used in many critical medical images, a standard object/surface is within the field of view.
Yes. A lot of work for somebody! I expect cosmologists can add input. Some of the concepts and maths they play with has application in other fields. I’m sure they have the ability along with material scientists and museum curators and others.
It’s quite interesting that the formula for the bakelite used to cover the handle of the Welrod pistol is still classified. Apparently, the materials of the equipment used during the manufacture of bakelite effected the final result so even if we had the formula with the original equipment now being long since scrap it may be impossible to make as it was. It’s a bit like trying to build the Saturn V today. The ecosystem isn’t there.
Personally, I love plastic. It truly is a wonder substance.
cpcf,
That’s what I was thinking too. You may be able to re-calibrate based on other known objects (and even people) in the shot. If there’s a light bulb, that would have had specific wavelengths, etc. I don’t think anybody’s going to seriously pay to pursue this, but I think there are avenues that could potentially work.
I think if you had an old color magazine, there would probably be enough color photos of known colors (like grass and even specific people) to compensate for the fade. Even something like a coca-cola ad with a precise color of red could be used for re-calibration. This way you could reproduce the magazine colors as it was. However this assumes that the magazines where accurate when they were originally printed, which may not have been true.
If you had a picture of a solid object through a prism, it ought be possible to measure the original wavelengths in a much more direct way. The color shifting after the fact wouldn’t matter nearly as much because you’d know from the position in the spectrum exactly what color it was. And by looking up the frequency response(s) for a camera (either b&w or color) you could ultimately reverse engineer the original color. Of course the odds of anyone having taken this picture through a prism are slim, but a side effect of camera lenses is that they have this unintentional distortion in the form of chromatic aberration…
https://www.adobe.com/creativecloud/photography/discover/chromatic-aberration.html
https://www.cambridgeincolour.com/tutorials/lens-corrections.htm
Newer cameras have gotten better at minimizing chromatic aberration, digitally or otherwise, but theoretically an old photo with sufficient resolution and chromatic aberration could be used in place of a prism. This was particularly noticeable in harsh lighting. It would be much more challenging than using a prism directly. So I wonder if A) if such a photo exists, and B) whether it would be possible to run a digital transform across it to precisely derive the original colors without regards to the present state of pigmentation.
One last idea, what about VHS? Since the colors are represented by a frequency, colors don’t fade on this medium in the same way as they do with ink/photographs. I know these are noisy as hell and the color accuracy was quite bad to begin with, but on the other hand since we are trying to derive a single color with potentially millions of good samples, that might get us far better precision then a single pixel could. It may be possible to once again calibrate against known colors in the scene to derive target color.
For such a useless task, it’s fun coming up with ideas!
Aberration, metamerism, etc, etc., are all things science imaging has to deal with for every image not just daily or weekly calibrations. Even at the very very high end of CCD imaging, devices worth 5, 6 or 7 figure sums, devices still need to have calibration images taken before and after an imaging run.
You cannot just know the content, you have to know the lighting and environmental conditions as well as the sensors own response curve(Called a fingerprint because very sensor is different) with which an image was taken.
Fade in printed references like pantone books is inherently chaotic, because the colour changes are often dependant on trace element analysis of the pigments, dyes, vehicles and substrate. High end pantone or other colour books print formulas and RGB or CMYK reference values under each colour patch.
Clemson University was sponsored by industry to do a lot of work on this, they have some great resources not sure if they are still publicly available.
cpcf,
Sure, but there may be more novel ways to calibrate instruments.
That’s why I was trying to find a physical phenomenon that might possibly record information about the wavelengths besides the pigments themselves. There may be some redundancy in the data that we don’t ordinary see or think about but could nevertheless be there physically and mathematically.
For example, even a “black and white” photo that has no color pigments at all can still provide evidence for colors as they pass through a prism. The color information is still there, it’s just represented in a different way.
https://www.cloudynights.com/topic/510385-new-sx-spectrograph/
Consider that the color on a spectrograph is there for our benefit, but the information about each wavelength is represented by the total intensity at a given position. Arguably this gives us even more information about color than RGB or pigements can. Our eyes and indeed our monitors are NOT reproducing the full real color spectrum. Our natural eyes are unable to discern the difference between a light source that produces red + green wavelengths versus a light source that produces true yellow wavelengths. Conversely different pigments can render the same apparent colors to our eyes using different combinations of wavelengths and you would clearly see the difference on a spectrograph even though your eyes cannot. Likewise we think of people who are “color blind” as being unable to distinguish between different colors that we see, yet in the exact same sense we are all color blind to wavelength variations that “normal” eyes can’t see.
Anyways this was a bit of a tangent, but my point is that pigmentation is not necessarily the only way to represent colors in a photograph. I hope I’ve managed to convey that better this time.
So, while a camera lens is clearly not an ideal prism, it nevertheless does record the exact same physical phenomenon via color aberration. With a sufficiently complete mathematical and physical analysis it may be possible to derive source colors from the fridge effects. Even if it’s theoretically possible though, there may be too little resolution to actually pull it off :-/
Here is what I don’t understand: Why Beige? Its Apple, think different and all that. Steve Jobs trying to find the perfect shade of Beige is just so wrong of an image to contemplate. Why not red like an apple? Why not green? Or Black or any other color that wasn’t boring. I grew up in that Beige decade. So I associate it with dull, corporate conformity, was Beige cool in the 70’s? Or was one of the most revolutionary of computers just trying to fit in with the crowd.
I mean if beige was just the color everyone used and no one ever thought about it ok, fine that makes sense. But this article makes it very clear that it was a very deliberate choice, which is really odd.
For the early stuff, almost feels like there’s a technical or cost reason because they all were that color. And the 70s did like their muted earth tones.
From what I remember (I was a kid in the early 80s, caveat emptor), the Apple 2 e’s were a fairly normal beige color you’d think of when you think of generic 90s computer cases. My Atari 400 was a little more brown and had a great brown and yellow keyboard. The Apple 2 c and gs weren’t what I’d call beige but off-white, tinted slightly grey not brown. Same with the early Mac like the SE. By mid-decade, Ataris and TIs were often black, my 800XL was black front and keyboard with a beige strip in the back half. The Amiga 500 was pretty beige. It’s worth noting that a lot of the computers we’re talking about attached to the single TV in a home, so they had to fit into a living room or den, like a VCR. Beige and black are safe bets there, not everyone is going to accept a bright red device sitting next to the TV.
It wasn’t until the early 90s and the dominance of the PC that everything was beige. And even by the late 90s major companies were moving to black as an option. I spray-painted my Full-AT case royal blue because “everyone has a black computer, but no one has blue”. It was a while before I saw a blue case for sale.
MattPie,
I never really cared. I bought a purple computer in the 90s because it was cheaper than the others. I prefer cases without all the lights though. There are more and more of these in the enthusiast consumer market these days. I’m solidly in the function over form camp and I don’t want any lights on my fans and RAM. I’ll tolerate it if I have to, but that’s just dumb.
I’m ok with backlit keyboards though, but not because I want to show it off. Whenever I see technology used as a status symbol I can’t help but think that’s shallow. Then again, tell that to me and my purple computer, haha.
he Apple II was one of the first beige computers, the reason beige was chose was because at that time personal computers were not common place and it was the color that would allow some of the widest adoptions as it would fit in with most people’s homes and offices.
Ok fair point, this wasn’t think different Apple where PC was dominant and it wanted to stick out. It wanted to be the PC everywhere. They were going for mass adoption. A more striking design would have made selling it to consumers or schools or businesses more difficult maybe?
I still think it was lazy, but I also wasn’t talking to retailers at the time or schools that would have probably told the odd company selling the expensive strange machines I’ve never seen before to make it look like a typewritter or calculator
I guess that’s the closes comparison, but typewriters were black before any other color.
And look they were all sorts of colors:
https://mycupofretro.com/1970s-typewriters/
Acorns BBC Micro was beige. Commodore PET was white? RM 380Z was a horrendous black colour.
Yeah how could I forget this:
https://en.wikipedia.org/wiki/File:ZXSpectrum48k.jpg
Thats the computer that makes more apply sense to me. But again I’m applying their 90’s ethic to a late 70’s version of them probably not fair.
@Bill
Sinclair was into minimalism. Most of his stuff was black. ZX80 was white!
At one point Apricot had a bigger R&D department than Apple and designed their own motherboards.