Chocobo Dye Experimentation

[cearka]

I'll update my progress here.
My friend believes that each fruit adds or subtracts a value of 16 to the given color (with a little bit of variation to it) and rounding it up/down when it hits a predefined color. We've decided to test this by calculating the difference between the current and the target color and divide it by 16 to get an estimate amount of items you would need for that certain color.

Hello, I'm that friend. thoughd I'd add a few more details:



Apologies if this has been posted already. This is only a THEORY, so it might be wrong, but it's been working for me so far.

Theory 1:
I "think" that each fruit basically alters +/- 1 of the first digit of the RGB hex value, with a random chance of +/-
a few of the 2nd digit. This means, each fruit basically adds or subtracts around 16 points.

Theory 2:
If you go over 255 or under 0, it loops around. This part is definately THEORY, I haven't 100% comfirmed this, but
it seems legit so far. If someone wants to confirm this, they could test it by giving a Snow White +1RGB to get Soot Black, or -1RGB for the opposite.

So does this mean?
First you have to understand how RGB hex works. There's a table floating out there with the RGB values in integer format (which may or may not be accurate)
meaning theres a number from 0 to 255 representing Red, Green, and Blue. Each of these numbers can actually be represented
with a 2 digit Hex value, so then it would actually range from 00 to FF.

Colors are represented as six digits then, the first 2 being the red channel, 2nd being green channel, and the 3rd being the blue
channel. HEX is used in computer graphics because its far more efficient to a computer in terms of memory usage and something
to do with they way 32 bit processors work (someone feel free to correct me on the this, I'm an artist not a programmer
so I don't know the exact details for this)

So, 000000 is Black, and FFFFFF is pure white.

16 in red, green, and blue would 101010. 32 would be 202020, and so on and so forth.

So lets put this into practice. You start with Desert Yellow (216, 175, 80). Using this calculator,
http://www.javascripter.net/faq/rgbtohex.htm

it converts to D8AF50. For the purpose of this, we only care about the odd number digits, so we'll do some rounding. So it goes to
D0B050.

lets say we want to get Ochu Green (57,92,51) or in hex (395C33) and rounded Hex (406030)

so we put these to numbers next to each other, Source over Target

D0B050
406030

And we calculate the difference

-9 Red
-5 Green
-2 Blue

So, in theory, 9 doman plums, 5 valfruit, and 2 cieldales pineapples would get you to the target color, or very close that
the 2nd feeding would be much easier to figure out.

But wait, you should never have to use more then 8 of one fruit. So, if you look at red, you realize you can go the OTHER direction.
If you go the other direction its +8 Red. So now it becomes 8 apples, 5 valfruit, and 2 pineapples.


This all speculation, but again, this has been working well for me. Feel free to play around with this idea and let me know.
My target was Dalamud Red. From the default yellow, I went off the "instructions side" of that spreadsheet going on and it
was totally off, but using that result and a few other results, I was able to extrapolate the above math and my 2nd attempt using this
math got me to rolanberry red (which is super close to dalamud). From rolanberry red, it was only 2 more fruit to get dalamud
using the same math. SO, either my math just happens to be a big coincidence, or it actually works. I hope some people can test this
cause I got the color I want so I'm not gonna mess around anymore >.> But to be clear, if this does work, it means
no target should require more then 16 fruit total, and this is to go from one color to a complement.
So the maximum fruits are required from going from
Red <==> Green
Blue <==> Orange
Yellow <==> Violet

The range would be 1 Fruit < Mono < Analogous < Triatic < Complementary < 16 Fruit

There is one exception to the Max 16 rule under the above math, and that's the Monochromatic grayscale. A jump of 50% in on the monochromatic gray scale would require 24 fruit.
(ie, White or black to 50% gray, 25% to 75% gray, etc.)


However, if Theory 2 proves to be wrong, that means it would take up to a maximum 16 fruit to make a change, in which case then the max numbers turn to 32 and 48 respectively.
If theory 1 is wrong but 2 is right, then it means it really is just a case of plus or minusing in the direction you want to go, but the amount of plussing or minusing that happens from fruit has far more variables then being around 16 (such as diminishing returns, different seed numbers, higher rng range, or just being flat out random).
If theory 1 AND 2 is also wrong then this whole venture of math just gets thrown out the window. LOLOLOLOLOL.

Lastly I should mention my stable was in good condition for this.

*edit* was reading Renik's post about how the fruit actually do multiple things. If this is true (and it probably is), it seriously complicates the above theory for big jumps. I think for small jumps (under 4 fruit, with no more then 2 of any one fruit), you can ignore the secondary effects of a fruit, but for larger jumps the math becomes seriously complicated. Le sigh.

SOI:SOI:SOI:SOI:SOI:SOI:SOI
..........___|___.........
...L..../.......[`\........
..LOL===........[__|.......
...L....\..........|.......
.........\_________/.......
...........|...|...........
..........————————/........