In order to be properly 8-bit it’d also need to round the colours to some quantised palette. The physics of that would be much, much more interesting.
I’m guessing that Atari 8-bit computers would be the easiest, followed by pure 8 and 16-colour RGB and RGBi palettes. To do the Commodore 64 palette would be a very interesting materials science project.
And then do that with variable ones, like the Commodore 16, where you have an arbitrary subset of a quantised color space.
It just so happens that one of my colleagues just finished a PhD creating materials which pretty much do exactly this; converting a relatively broad spectrum of light into a much narrower band of light. I’ve seen them in the lab where it’s colourless and clear to start with, and then it will convert any incident light in the blue range into a much narrower band of a specific blue colour. He has recipes for just about any colour, even into UV and IR bands.
Not sure what the real world applications are though, maybe something to do with coatings for photovoltaic cells to increase efficiency
There was an article about a flat nano-tech lens, but it only works for a single wavelength of light. Combining the two could result in the ultimate "pancake" black & white camera.
I'm not sure that would work well - if you're only recording a single wavelength, then the resulting black and white image wouldn't resemble a normal one, where all wavelengths are added to obtain a pixel intensity.
I can immediately see applications in IR and UV for hyperspectral imaging using cheap sensors. Your idea of PV cells is also excellent, provided the wavelength compression is efficient.
What’s needed is not to compress colors into one narrower band, it’s to quantize them into multiple combinations of bands (e.g. combinations of three RGB bands).
I was thinking something similar, then I realize that some of the stones appeared to be 8x6 pixels. Perhaps 8x8 if you counted the highly distorted top and bottom edges. In other words, it the palette would include fewer than 256 colours. That said, I'm pretty sure user defined palettes with 256 colours would be beyond the capabilities of any (common) 8-bit machine.
It wouldn't surprise me if there were 8-bit computers that could. On the other hand, I would find it surprising if there were 8-bit computers that could. At least for choosing 256 colours out of a larger palette.
Keep in mind that 8-bit computers were limited to 64 kB of memory unless they used tricks like banking. It looks like the CPC used about 16 kB for video memory. Bumping it up to 8-bit colour at the lowest resolution would require 32 kB for video memory. (Adding a palette to that would fit into the rounding error.) Even if that memory wasn't directly addressable by the processor, the cost of RAM was another reason why those 8-bit machines were memory constrained.
> Keep in mind that 8-bit computers were limited to 64 kB of memory
Video memory is not always mapped to the CPU's memory space. A lot of 8-bit computers had dedicated VPDs with their own memory (the TI-99 is a pathological case of that - where the CPU had almost no memory and BASIC programs ran from the VDP's memory). MSX2 computers can, AFAIK, display 256 colours out of a 512 colour space (and, unlike their 1.x predecessors, VRAM can be banked into the CPU memory space.
When you say "8-bit look" it's mostly referring to 8-bit computers (even though it's most often low-resolution, pixelated, sprite-like animation with great color depth that is labeled that way)
> In order to be properly 8-bit it’d also need to round the colours to some quantised palette. The physics of that would be much, much more interesting.
Especially since determining "the color"--even in a huge palette--ia a biologically-bound process, linked to the reaction of cells in primate retinas, as opposed to being a fundamental optical or mathematical operation.
Multispectral wavelengths are to colors as chemical-stew is to smells.
I can see how it works in one of the links you can see that there's 2 faceted almost half cylinders attached together with a 90' twist so you get quantized (in space, not color value) sampling on each x/y axis.
I was thinking just that - I’d imagine it’d need to filter light to excite something that’d emit light at the wavelengths you want. It’s, as you noticed, a much more interesting problem.
Or emit white light when excited by a filtered wavelength which you’d filter again to get the color you want. With this second one you could do arbitrary palettes such as the Commodore 64 one.
Not exactly a crystal, but a quick google tells me trichroic prisms exist (https://en.wikipedia.org/wiki/Dichroic_prism). Once you've separated out RGB, could you quantize them and recombine them?
Well, “color” is fundamentally an electromagnetic spectrum. You can’t describe it by 3 values, as it might have an arbitrary shape. What it does is separate out the whole spectrum/graph into 3 graphs that overlap, and summing them up would roughly be the original. But neither of them is quantized this way, the R is not a single value, but still is a spectrum.
It’s a human sensory “feature” that certain graphs can be substituted by more standard ones that can be represented by a single number, e.g. a red LED’s spectrum between a certain range might combine well with a green and blue’s given spectrum, so that the combined graph looks similar to the original to us. Nonetheless, we have “compressed” the original and lost data.
I have to assume we're aiming for art, not science, here. If someone figures out how to get step changes to a light source with passive components, I'm gonna be impressed.
Splitting the light three ways first would just be the icing on the cake.
There were articles every few years about some density of transistors being a hard limit. The we crushed it. Then we packed more layers. Then we crushed the new limit again. The resolution was not mentioned, but effectively it would be limited too if the frequencies couldn't go up.
Making pixel art is not about low resolution. AI is finally getting close to being able to do it. If you could do it with a crystal it wouldn't have taken 30 years of trying to automate pixel art creation.
Base SDXL can't do it but with some Loras you can get something that after putting through Photoshop to regularise pixels and palette would pass the pixel art Touring test.
I’m guessing that Atari 8-bit computers would be the easiest, followed by pure 8 and 16-colour RGB and RGBi palettes. To do the Commodore 64 palette would be a very interesting materials science project.
And then do that with variable ones, like the Commodore 16, where you have an arbitrary subset of a quantised color space.
Try that without a power supply.