🎨 A painting/Fediverse experimentation:
With percentages of Red, Green and Blue, you can obtain a precise RGB color definition.
So, please vote for one or more color channel and with the mix of all the votes, I'll challenge myself to paint a fantasy landscape, using the result as the dominant color.
(PS: I'm doing full time inking/penciling on episode 37 and a bit of colored art cool-down would be great! Also I need a bit of 'Fedifun' this week. Thanks for your votes and boosts! )
@davidrevoy What's interesting here is that the colorspace is limited - for example, you can't obtain black this way unless nobody votes in the poll. Unless you clamp the result such that the color that receives the most votes is equivalent to HTML ff, then you'll never obtain white either, or anything close to it, because the three channels can't add up to more than 100%.
@jordan @davidrevoy you could use several of such selection/clamping methods to get a small and practical palette to work with 🤔
@batterpunts @jordan True; it sounds like this vote is a chaotic system with an obvious "great attractor " around 35% to 55% range in each color; and a bias maybe to blue or green because I annunced in the message I'll do a landscape (or maybe because blue is the a common favorite color over green and red).
I'm curious of what could have been a better design for the rules, using the same tool.
@davidrevoy If I understand correctly, your method restricts you to colours with 33% luminosity. If you want to do this with the full RGB colourspace, you might do red/green/blue/white as options. Then you say your origin is O=(0.5, 0.5, 0.5), and each colour corresponds to a vector pointing in the direction of the associated vertex of the [0, 1]³ cube representing the RGB colourspace: R=<0.5, -0.5, -0.5>; G=<-0.5, 0.5, -0.5>; B=<-0.5, -0.5, 0.5>; W=<0.5, 0.5, 0.5>. Multiply each vector by its percentage, add it to your origin, and your total can reach any point in the cube.
If everyone votes for the same colour, you get it at full saturation and brightness. You'll still get a grey if the RGB votes are evenly split, but its lightness can vary depending on the proportion of votes cast for white.
@davidrevoy @khird
I'd propose a different method... We could take the RGB result of the original poll and convert it to HSV, then just set saturation and value to 100% and convert back to RGB. This would give a fully saturated color on the HSV color cone rim, while each vote still biases the result to itself. The chance for grey is miniscule, cause R, G, B would need to be equal for that.
You could do separate polls for saturation/value, but I'd think you'd need at least one degree of freedom.
@khird
I don't think the luminosity is locked to 33%: audience can vote for more than a single color channel(so result of additing RGB is not 100%); and lightness perception of Red, Green and Blue is very different. (Check HueChromaLightness aka HSY' colors in Krita for Lightness perception).
> You'll still get a grey if the RGB votes are evenly split
Yes, that's an effect I haven't predicted with percentage and votes; but it was probably obvious for someone in stats.
@jordan @batterpunts