8 ms·
It is commonly used as a transparency placeholder for a reason - nobody wants to actually see it. My memory from my days in television was that magenta was use
by reaperducer 7y ago
It is commonly used as a transparency placeholder for a reason - nobody wants to actually see it.
My memory from my days in television was that magenta was used to represent transparency because it was the opposite of chroma green, and was easy to generate electronically. That's why it ended up being in one of the palettes for the primitive-by-today's-standards CGA adapter in the early IBM PC's.
Chroma green, magenta, and superblack were three "magic colors" that made TV production work in the pre-digital age. Sadly, there is no wikipedia entry for superblack even though everyone saw it all the time, and few understood its utility.
- ineedasername 7y agoWhat is it about magenta that makes it easy to generate electronically?
- dylan604 7y agoIn TV, the 3 primary colors are Red, Green, Blue. Magenta is a secondary color made from combining Red and Blue. That's what makes it easy. Just turn on 2 of the 3 colors you have to make it. It's not a fraction of them, it's just full on of 2 out of 3.
- ineedasername 7y agoAh, of course! One of those things that's obvious once I know the answer.
- aaronblohowiak 7y agoFun fact: there is no such thing as magenta as a frequency of light, it is a fun trick your mind does to explain both the low energy and high energy receptors activating without the middle energy receptors also activating. In terms of pigment, magenta is one of the primary colors.
- lubujackson 7y agoYes, was going to say this. Magenta isn't on a rainbow and isn't really a normal color in that sense. By the weirdness of how our eyes detect colors amd how our brain processes them, magenta is something like an error response catch-all for colors outside our eye's capabilities and from that standpoint it is quite interesting.
- mcv 7y agoOur heads are good at mixing colors, and we register those as colors too. Brown isn't on the rainbow either. The most interesting color on the rainbow is violet. It registers in our heads as a kind of purple when objectively it's super blue. But the cones that register red in our eyes are also slightly activated with this super blue light, which makes it register as purple in our brain. I'd love to see an experiment where people are shown both true violet and purple mixed from red and blue in proportions that make it register as the exact same purple as violet does. And then check how many people see a difference between the two colors. That might finally answer the question of whether red in your head looks the same as in my head.
- SamBam 7y agoI've always wondered by violet looks purple. Thanks for that. So, to be clear, it's just a hardware detail with our red cones? They fire in the presence of photons in the red band, but also happen to fire in the super-blue band? It's just an artifact of how they're constructed, either by chance or evolution? > I'd love to see an experiment where people are shown both true violet and purple mixed from red and blue in proportions that make it register as the exact same purple as violet does. Isn't this basically exactly what the computer color violet does? E.g. there are photographs of violet at https://en.wikipedia.org/wiki/Violet_(color) https://en.wikipedia.org/wiki/Violet_(color), which presumably look close to reality, but the hex color is #7F00FF, which has a lot of red in it. Is that red not there to simulate our perception of violet, if I've understood you right?
- mcv 7y ago
- gugagore 7y agoHere is some visual support for that explanation: https://en.wikipedia.org/wiki/Color_spaces_with_RGB_primaries https://en.wikipedia.org/wiki/Color_spaces_with_RGB_primarie... It's fortunate that we mostly have 3 categories of cones and also 3 spatial dimensions. Notice that magenta is a vertex (of 8 vertices) in the cube, along with Yellow and Cyan too. How the particular space (of activation to the 3 categories of cones) is carved out by that cube is something we engineered, and in fact is different across different printing and display technologies.
- jayd16 7y agoWith RGB floats magenta is #FF00FF and that's pretty easy to remember.
- jameshart 7y agoInterested in what field you get the term ‘rgb float’? Not a term I’ve heard before (and not an accurate one) so I’d be interested if it’s a term of art in some industry...
- jayd16 7y agoComputer graphics use floats for color values very often. It was indeed misleading of me to mention floats and then use a hex value to explain it, though. Just a brain fart from seeing it earlier in the thread. With floats it would simply be (1,0,1).
- hug 7y agoI'd love if you were able provide some reading on "superblack", or even just explain it yourself. I tried looking it up myself, because it sounds interesting, but of course search results are polluted with articles on vantablack. "Super black TV" and related search terms just bring up garbage about OLED TVs and such.
- p1necone 7y agoMy assumption would be that it's whatever the darkest shade of black representable is - so 0x000000 in rgb. However that can't really be read as a distinct value in an analogue signal, maybe it's some kind of "sufficiently out of range" value so as to be distinguished from legitimate colors?
- kaetemi 7y agoI remember there was some particular shade of off-black in Windows 95 that was always transparently showing any video that was playing. Don't think it's related though.
- zamalek 7y agoThat happened because 3D cards used something called "overlay". Recall that your 3D card (Voodoo or some other 3dfx) was an additional card beyond your 2D card. The card would overlay its signal on the 2D card according to 2 parameters: a rectangle and a color. I assume that videos used the same machinery, possibly in software, so that you couldn't take screenshots for copyright reasons. Or maybe it was faster for some reason, who knows (Raymond Chen might). Either way, it's probably unrelated.
- cycloptic 7y agoThe video acceleration was done the same way with overlays because of the same hardware limitations. I remember the difference was that they used chroma keying to try and fake that it knew how to work with the windowing system -- the hardware would mask the video using the chroma key color while rendering the plane. Of course if you had another window that happened to have the chroma key color for text or something, you could drag it over the video and the text would appear oddly transparent. This happened with certain drivers on Windows up to XP, and on Unix/X11 when using the old "XVideo" extension. But this is indeed tangential and not relevant to the original comment, because the chroma key color could be anything. On one particular driver I had, I remember the chroma key color was a very ugly shade of brown :)
- zamadatix 7y agoIs this not what you are talking about? https://en.wikipedia.org/wiki/Super_black https://en.wikipedia.org/wiki/Super_black
- tesseract 7y agoIn this context, it's not a physical color. It refers to an analog video signal with a lower brightness level than the level which was defined as corresponding to pure black.
- wizzard 7y agoThat article is about a “surface treatment” for solid objects, like Vantablack. OP is talking about screens — I believe they are referring to OLED true black.
- dylan604 7y agoI was only familiar with the term blacker-than-black in analog broadcast. Legal black was 7.5IRE, but any value below 7.5 was referred to as blacker-than-black. Older CRT TVs were very susceptible to bad(cool/interesting) things when the video signal was out of spec.
- kiwidrew 7y ago> That's why it ended up being in one of the palettes for the primitive-by-today's-standards CGA adapter in the early IBM PC's. Interesting... do you have a source for that? The hideous CGA palette is legendary, but I thought the reason for its ugly palette is a consequence of: (1) only having 16KB of video RAM, and (2) only having ~10 bits of mode-setting and colour control registers. The CGA card was built entirely from off-the-shelf discrete logic chips (plus a 6845 CRTC) and there simply wasn't room to add any extra registers. At 320x200 resolution the colour depth was limited to 2bpp, and these two bits are wired up directly to the red and green bits of the display output. The blue and intensity bits are (in graphics modes) wired up to the mode control register. Thus the four hideous palettes: the B and I bits have to be set to a fixed value for the entire screen, and only the R and G bits can vary on a pixel-by-pixel basis. (There's a bonus palette of red, cyan, and white that can be enabled by yet another mode control bit; and the background colour -- which is usually black -- can actually be set to any one of the 16 available colours.)
- toast0 7y agoI think you both agree. Magenta is easy to generate, and the CGA electronics were designed around easy to generate colors, not pretty or good looking colors.
- pickpuck 7y agoI was thinking because it is one of the inks in CMYK for printing.
- praptak 7y agoYou are probably thinking of rich black, which in CMYK is black plus some of the other three ink colours. Converted to RGB it would still be black, but in actual printing it has its uses.
- jameshart 7y agoI definitely recall hearing the term super black as distinct from rich black to refer to just a mixture of black and cyan, but I can’t find a cite for that. May have been local to a specific printer? This was in a context where avoiding using four inks was an important cost saving, so maybe it was just a two color equivalent of rich black. Of course, those two colors were also immortalized by John Cougar Mellencamp in his “little ditty ‘bout black and cyan...”
- bonzini 7y agoThe 2-bit palettes of CGA were simply colors 2/4/6 and 3/5/7 of the text mode palette, where bit 2 is red, bit 1 is green and bit 0 is blue. The fourth color (encoded as 00) was the background and could be chosen from any of the 16 text colors. (In addition there was an "undocumented" palette consisting of colors 3/4/7 plus the background color).