Brown: color is weird [video]
youtube.com
youtube.com
I especially enjoyed his "rants" about the design of a old toaster, and traffic lights, of all things.
The toaster one is alright, it kinda bothered me when he said there was "literally NO reason" to design something the way it was (I think it had to do with the position of the coils) - obviously there is, it's probably just not a reason he'd like (cost, ease of manufacturing, etc). It makes me wonder if there was other info he skipped on because it "wasn't important" or "stupid". That's a pretty minor point though, overall his videos are very good.
Funny thing: some people have joked that he's my alter ego (I guess we look slightly similar?), but I could never deliver stuff like he does.
I wish he did have a more technical discussion on CRT vs raster displays. There is a real meat there that is typically unexplored. What is the spatial frequency response of CRT vs raster? What resolution raster do you need to be “good enough” and what kind of interpolation/LPF should be used?
You know, "Building a CRT from Scratch" would be a great standalone educational series. It could hit on everything from making glass, to how to pump air out of something, to coil winding/transformers, to phosphor chemistry, as well as the RF electronics stuff.
https://www.youtube.com/watch?v=IYO3tOqDISE
(not the same guy at all, but your comment reminded me of this. It's really about specialization and the interdependency of everything in the economy)
[1] https://www.youtube.com/playlist?list=PLv0jwu7G_DFUoByWSHHoS...
[2] https://www.youtube.com/playlist?list=PLv0jwu7G_DFVP0SGNlBiB...
I’ll have to listen to his story on them. Probably a very different view from me as a kid just using them to watch movies pre-VCR.
My parents recently moved into a smaller home, and ... deacquired a bunch of their furniture by way of my house. In the shuffle I ended up with a bunch of my parents' old records. Being susceptible to impulse buys, I bought a record player. I actually really enjoy the experience of playing records. There's something about the physicality of it. I can't just press a button on my phone and ignore it. I have to choose what to listen to next. I have to physically flip the record twenty minutes in. It's not purely a background thing, it's a physical process and experience. Music has gone from something I ignore to something I participate in.
I'm not actually sure that's in any way relevant to the conversation at hand, but when I started ranting, I felt it was.
About the perceptual uniform, colorblind friendly colormaps in matplotlib since 1.5. Now also adapted by most other plotting tools.
In short: the RBG values in an image are not actually the intensity of light, but some nonlinear function of intensity. To do a weighted average of two colours you need to first undo that nonlinearity, then do the averaging, then reapply the function. Most software doesn't do that so the colours end up wrong. This also applies to resizing images.
Ironically, Photoshop does this wrong by default, but the open source GIMP does it correctly (in current versions).
"Professional graphics software" does what it's supposed to be and sRGB is an inferior standard in most use-cases of that software. Its gamut is simply not large enough, but it's a nice smallest-possible commond ground.
"Professional graphics software" (i.e. Photoshop) absolutely does the wrong thing for color blending with the default settings. It's not a problem of being limited to the sRGB color space, it's about the math in use being outright wrong for use on that color space. You need to go into the settings and tick some checkbox to have some hope of it doing the right thing ("Blend RGB colors using gamma"), and even then I bet it only applies to some operations like blending itself, not scaling and other filters.
This is because all non-linear image encoding modes, like sRGB or straight gamma, should be viewed as compression. You don't perform mathematical operations on compressed data. It's like trying to mix together two MP3 files by literally taking the bytes and averaging them out. That's madness and does not work. You need to go into a linear format, like raw PCM, to be able to do that (with the right word size). Same with images: you need to convert from sRGB to linear light to be able to do math on them. But 95% of software, including "pro" software, doesn't do this, it just blindly does math on compressed (gamma encoded) data, and the result is ugly (we're just used to it).
Of couse, with Photoshop you can set your image mode to linear light and then things work fine... but that's not because linear light is required to make this stuff work, it's because Photoshop is broken when not using linear light and doesn't compensate for the encoding.
Meanwhile GIMP, a few versions ago, fixed this whole mess and actually applies blending and filtering in linear light (regardless of what mode your image is using in memory), making its blending and filtering much more pleasant and correct than Photoshop's. So yeah, it may not support CMYK or other "pro" features... but for a good chunk of basic editing in RGB space, you're actually going to get more correct and visually appealing results if you use GIMP than Photoshop.
https://www.imagemagick.org/Usage/filter/nicolas/
For another clear example of the problems of resizing in linear colorspace, see "Catch #3" here:
[0] http://colorbrewer2.org/#type=sequential&scheme=BuGn&n=3
https://www.youtube.com/watch?v=MJBfn07gZ30
Here's a part of a documentary talking about the cultural aspect of how we perceive colours and the power of naming a color (with the Himba people)
Dark cyan = teal.
Dark magenta = purple.
Dark yellow = olive.
And maybe some of the other tertiary colors too, but yeah orange/brown is the more prominent one.
A: A stick!
Wasn't that about magenta/fuchsia? ( #FF00FF )
That's why this video is good educational material: it covers the contextual aspect of color that a lot of folks will have never even thought of. After all, pink is just red that you added white to, why not just add black to orange to get brown ("because that works for paint, but is literally impossible when your medium is light").
Brown is, in general, a muted orange. Take the red and yellow, add a hint of green (or purple, honestly) - and you have browns.
We've affixed a lot of specific names to colours around orange and brown, but they're basically variations of saturation and lightness on a certain range of hues (between yellow and red). The distinction between brown and orange is mostly subjective.
Which is what I like about this video. It gives a good sense of how much of the difference between two 'colours' is just naming and perception of a hue.
We call dark orange brown. Sure dark red could be called burgundy but it's not a word used as much as brown. So we use the word brown as much as the words 'dark red'.
And the difference between "pink" and "brick red" is that both are just desaturated reds at different levels of lightness.
...unless you've got a universal web dark-mode extension. Then they're orange swatches. The swatches are visually brown by default, because the page they're shown on is white by default.
The video said "brown is orange with context", but I'd say it's more like "brown is orange with contrast."
White light + black fog from a fog machine? Can you "pigment" air like that?
Here's a paper with nice graphs, drawing where different languages put the boundaries between words -- they divide up the hue-saturation space into a varying number of blobs:
https://www.pnas.org/content/115/31/7937
(The data isn't theirs, it was painstakingly colected by going and asking lots of people "what colour is this pantone?". The paper does some fancy stuff on top of that. But it has nice graphs.)
https://i.imgur.com/dGDaVLz.jpg
https://nasaviz.gsfc.nasa.gov/vis/a010000/a011900/a011924/c-...
Like a combination of OLED and E Ink.
This will also give you the possibility to turn your TV into a painting (without using power) when you are done watching.
The E Ink layer would be great for status overlays or permanent elements that rarely change, like the date and battery charge, or a list of recent notifications while the "main/active" display sleeps when the machine is idle.
Some ebook devices also have ‘backlight’―usually diffused light from the sides so it doesn't shine directly in the eyes. I think this is the crux of your idea, because currently even if you use lcd/oled with backlight, no-power eink will be severely washed out in comparison. And if eink gets good color and contrast in the future, it will then fulfill the two-mode technology.
The page on the wiki.laptop.org isn't too clear on the technology—presumably it's more complex than just LCD and eink slapped together. The Wikipedia page on the XO laptop seems to have some detail: https://en.wikipedia.org/wiki/OLPC_XO#Display
If you think about it, that's what an LCD is doing. The backlight is emitting; the liquid crystals are absorbing. The display is white (i.e. passing through the emission from the backlight) where the liquid crystals aren't active; the display is black-ish (i.e. attempting to block as much of the emission as it can) when the liquid crystals are active.
It's sort of like painting on a lightbox with black paint.
Hue also makes sense. It's the distribution of wavelengths being reflected.
But what does saturation correspond to? When do things become saturated/desaturated in the natural world?
https://www.rapidtables.com/convert/color/rgb-to-hsv.html
I suspect a lot of natural stripe patterns require saturation to distinguish them.
This means that the signal getting to our brains is actually a report of "how much {(red minus green), (yellow minus blue)} relative intensity" there is in view.
Saturation is our name for the feeling of having strong visual qualia, brought about by the magnitude of one or both of the relative-intensity signals being large.
• This is why oversaturated bright colours can trigger migraines in people—they're the strongest encoded signals, causing the most synaptic firing.
• This is also why some oversaturated colours don't seem as "bright" as others—you can make a "Brown 2.0" that's ultra-mega-brown, but it can't be as intense in {R-G, Y-B} terms as the pinkest pink (https://culturehustle.com/products/pink-50g-powdered-paint-b...), since pink/magenta is a full relative intensity pin {1-0, 0-1}.
• This is also why staring at oversaturated colours for long periods leaves more of a visual afterimage than staring at equally-bright white does. It's the {R-G, Y-B} intensity, not the brightness, that's using up the electrochemical messenger molecules in your retina.
I would include this as part of an introduction to color theory.
If you were to make a V2, I’d suggest discussing gray, since I’m sure you’d find a way to make gray interesting (I mean it is though) :)
I had wondered if there was any example from own language of this, in order to get the benefit of that perspective. This seems to be it.
Perhaps we had a name for brown before orange. Whereas other colours tend to be named for the saturated version.
Has cultivation changed citrus colours, as with carrots?
We happened to be in the lab, so I pointed to a brown spot on the wall, made by a very powerful laser. After that, we referred to it as the brown laser.
I've seen many people try to argue that magenta doesn't exist, or isn't a color, etc., simply because it doesn't correlate to a single wavelength.
Regardless, what is he trying to say, then? That brown can't be understood because there is no such thing as a brown laser? My point remains.
On the other hand, your monitor does not absorb colors, but emits it. If you shine red and green light on a white surface (or on your eye directly), you will see green. This is called additive color. RGB (red, green, blue) are primary additive colors.
And really you could say red green and blue are subtractive primaries (as well as additive), but we use cyan ink because it subtracts red light. (etc)
The subtractive color model isn't really good for paints anyway, it is best with translucent inks/dyes, where the white from the paper comes through.
Each individual wavelength activates all three to some extent: think about this as a point in 3d space. For example, 400nm corresponds to something like (0.1, 0.05, 0.0), from that chart. 500nm might be (0.1, 0.4, 0.3).
But we experience a mix of many different wavelengths at once. So we can not only experience just these points in 3d space, but we can also experience any linear combination of any of these points. For instance, a mix of half 400nm and half 500nm light might be "sensed" by us as (0.1, 0.225, 0.15), even though maybe there's no individual wavelength that corresponds to that point. Any linear mix of any number of any wavelengths covers the entire gamut of what we can perceive.
The question then for someone picking primary colors for an additive display is: if I can only do a linear mix of three wavelengths, what wavelengths should I pick? What covers the biggest subset of the whole perceptible gamut? It just so happens that red, green, and blue do the best.
If you swapped green for yellow, there would be a section in that 3d space that you could no longer create. Specifically, the area where M cones are strongly activated compared to L and S. Unsurprisingly, this would be the greenest greens.
He also didn't mention that grey is dark white, by the same phenomenon.
And of course red is dark pink, for rooster teeth fans.
It's one way of conveying information the brain can interpret in a flexible way.
But the brain is also good at depth perception. Let's combine these :
For example in the app I released yesterday StlToRelief : https://news.ycombinator.com/item?id=22319140
I use a color gradient to increase the perception of depth. It makes a thin bas-relief standout.
With color-mixing 3d-printing this is an effect we can now exploit.
There is a closely related phenomena that we have when doing war paints for camouflage. Area (i.e. peaks like the nose) that would naturally catch the light appear brighter so you have to paint them dark while hollow area that naturally appear dark must be painted white. The resulting painting make a face look flat which stand out less.