Guess my RGB
March 31, 2024 12:29 PM Subscribe
This is good.
posted by Going To Maine at 1:21 PM on March 31 [2 favorites]
posted by Going To Maine at 1:21 PM on March 31 [2 favorites]
Well it might be because I have some filters on my phone to dull out the colors a bit, but it took me 73 tries :)
posted by bxvr at 2:01 PM on March 31 [2 favorites]
posted by bxvr at 2:01 PM on March 31 [2 favorites]
I don’t know hex numbers very well, but give me this test with Pantone solid coated and I will be 99.9%
posted by Devils Rancher at 3:23 PM on March 31 [2 favorites]
posted by Devils Rancher at 3:23 PM on March 31 [2 favorites]
This is really nice, and the kind of post that I've come to miss on Metafilter.
posted by JHarris at 3:58 PM on March 31 [5 favorites]
posted by JHarris at 3:58 PM on March 31 [5 favorites]
I wish there was a button labelled "Give Up"...
posted by WalkingAround at 4:03 PM on March 31 [2 favorites]
posted by WalkingAround at 4:03 PM on March 31 [2 favorites]
I wonder why this game is so very popular today. There've been a bunch of other games with this idea, here's another from a couple of years ago. Or this one from ten years ago or this one that desperately needs sliders. Susam's decision to limit it to 16 choices for each channel is smart.
I would like someone to make a game like this for a perceptual color space like OKLab or HSLuv. Those are all so non-intuitive, I could imagine a guessing game would help build intuition.
posted by Nelson at 4:08 PM on March 31 [3 favorites]
I would like someone to make a game like this for a perceptual color space like OKLab or HSLuv. Those are all so non-intuitive, I could imagine a guessing game would help build intuition.
posted by Nelson at 4:08 PM on March 31 [3 favorites]
7 tries to solve 06A.
binary search on the light channels seems to be a useful strategy if, like me, you're not super color-wise.
posted by Sauce Trough at 5:03 PM on March 31 [1 favorite]
binary search on the light channels seems to be a useful strategy if, like me, you're not super color-wise.
posted by Sauce Trough at 5:03 PM on March 31 [1 favorite]
I wish there was a button labelled "Give Up"...
right click--> inspect--> <body style="background: rgb(#,#,#);
posted by logicpunk at 5:05 PM on March 31 [5 favorites]
right click--> inspect--> <body style="background: rgb(#,#,#);
posted by logicpunk at 5:05 PM on March 31 [5 favorites]
This is the kind of thing I can obsessively hit the new game button on. Thanks, I think!
posted by mollweide at 5:18 PM on March 31 [1 favorite]
posted by mollweide at 5:18 PM on March 31 [1 favorite]
right click--> inspect
What I don't get is how come that's the "give up" option rather than the "very first move" option.
Here's a good game that anyone can play: stand up, press the palm of a hand over each eye, and walk. No peeking, because that would spoil my fun.
posted by flabdablet at 6:30 PM on March 31 [1 favorite]
What I don't get is how come that's the "give up" option rather than the "very first move" option.
Here's a good game that anyone can play: stand up, press the palm of a hand over each eye, and walk. No peeking, because that would spoil my fun.
posted by flabdablet at 6:30 PM on March 31 [1 favorite]
Pretty interesting. It took me WAY more tries than I would have thought!
posted by rnicholson2020 at 9:18 PM on March 31 [1 favorite]
posted by rnicholson2020 at 9:18 PM on March 31 [1 favorite]
This thing jolted me back a few decades, when the ability to recognize Pantone colors actually mattered. To this day I can spot Red 032 a mile away. Colors are fun.
posted by kinnakeet at 4:46 AM on April 1 [2 favorites]
posted by kinnakeet at 4:46 AM on April 1 [2 favorites]
What’s cool about this is, someone’s done this as a research publication decades ago. We take for granted physical color is a linear scale of wavelengths, perceptual color is a circle composed of these 3 RGB channels (matching the 3 “cones” (opsins ) in your eye), with a lot of redundancy: ie you can give the perception of a single color through multiple different combinations of R, G, and B. When they had people do the physical version of turning three knobs to match colors, they create 3 curves that match the response curves of the opsins in our eye. A surprising detail: despite all the computation to perceive color, the most peripheral information (the transduction curve of photons into a chemical signal) is preserved all the way through perception and behavior!
ps magenta isn’t a “real” color, but I digress…
posted by rubatan at 11:09 AM on April 1 [3 favorites]
ps magenta isn’t a “real” color, but I digress…
posted by rubatan at 11:09 AM on April 1 [3 favorites]
I understand why some people would like this. For me, it just removes the joy 8 year old es_de_bah got from using color sliders and XY for the first time. Or even the time I spent mixing paint and tweaking color matches. It's like online dating without dates, pictures, or sex. </yuckingyums
posted by es_de_bah at 3:44 PM on April 1 [1 favorite]
posted by es_de_bah at 3:44 PM on April 1 [1 favorite]
OK you nerd-sniped me, Magenta: the Color that Doesn’t Exist. The argument is that red and purple are at opposite ends of the visible spectrum, so there's no single light emission that looks magenta. Which, well, sure. But humans don't perceive color with some sort of linear wavelength detector. We see red and blue simultaneously, we perceive magenta. Color is a psychological phenomenon, not just physical. It's why we have a "color wheel" and why we talk about colors in terms of a three-dimensional vector space.
If you really want to bake your noodle, take a gander at impossible and chimerical colors. I've experienced them best using this video. They work by over-stimulating color receptors in your eye, then removing that stimulus so you briefly see the opposite of the color. There's literally no physical emission like the color that you are absolutely perceiving.
posted by Nelson at 4:10 PM on April 1 [2 favorites]
If you really want to bake your noodle, take a gander at impossible and chimerical colors. I've experienced them best using this video. They work by over-stimulating color receptors in your eye, then removing that stimulus so you briefly see the opposite of the color. There's literally no physical emission like the color that you are absolutely perceiving.
posted by Nelson at 4:10 PM on April 1 [2 favorites]
They need a hard mode where the color is placed against different backgrounds, creating color consistency illusions.
posted by betaray at 6:49 AM on April 2
posted by betaray at 6:49 AM on April 2
The linear wavelength to perceptual colorwheel problem is a very interesting one. Converting back and forth is mathematically "non-trivial."
In graduate school I took a couple webcams and started setting them up for augmented reality. It turns out the camera sensors can detect more than the visible wavelengths of light, if you remove the "hot filter" (example: IR-photography). My idea was to smash the IR-visible-UV spectrum into the visible spectrum, to have increased perception (without having to resort to injecting novel opsins into my own retina a la Neitz Lab).
The problem is, even with only the visible spectrum you get a lot of "discontinuities" where you perceive what should be nearby colors as polar opposites, and resolving all of them is really hard if not impossible. It turns out evolution and all that time spent in development does something... though (a la Neitz and other weird plasticity experiments) I'm inclined to believe you could eventually get used to it and maybe even make use of your new found beauty in the world.
Just putting this here in case someone every wants to attempt the same, in an era of better onboard tech.
posted by rubatan at 9:56 AM on April 2
In graduate school I took a couple webcams and started setting them up for augmented reality. It turns out the camera sensors can detect more than the visible wavelengths of light, if you remove the "hot filter" (example: IR-photography). My idea was to smash the IR-visible-UV spectrum into the visible spectrum, to have increased perception (without having to resort to injecting novel opsins into my own retina a la Neitz Lab).
The problem is, even with only the visible spectrum you get a lot of "discontinuities" where you perceive what should be nearby colors as polar opposites, and resolving all of them is really hard if not impossible. It turns out evolution and all that time spent in development does something... though (a la Neitz and other weird plasticity experiments) I'm inclined to believe you could eventually get used to it and maybe even make use of your new found beauty in the world.
Just putting this here in case someone every wants to attempt the same, in an era of better onboard tech.
posted by rubatan at 9:56 AM on April 2
« Older "I wasn't really happy, you weren't really happy" | Whatever else Farley’s work is, it isn’t AI—even... Newer »
This thread has been archived and is closed to new comments
* 13 tries
posted by moonmilk at 12:59 PM on March 31 [2 favorites]