Chocobo Dye - The Missing Manual

[Rex]

Thanks so much guys! Using your advice and that calculator (which is a beautiful thing) Lyric is finally Dalamud Red.

Thank you so much!

[Syrehn]

IMHO, which RGB value we use, doesn't really matter too much, since all the methods uses the same +/- 5 factor. So there shouldn't really be too much difference other then 1 or 2 fruits.

But what is more interesting is this "range" business. As I said in micofeeding, everyone has been so busy and focused on getting to the color they want, there has not been any tracking/data logging of the range a given color has based on the fruits people have fed their birds. Since I stop paying attention to the the R&D thread after post 700th or so, there must be some kind of pattern if one exists.

I believe from the microfeeding method, we finally are starting to see that single feeding of the right fruits in the right pattern can nail the color on the spot starting from Desert Yellow, more importantly, allows the feeding to be duplicated. If we find this is true as more people applies the method, we can then move on to this "range" business, and start to include that into coloring factors when we jump from one color to the next. But I think this range is going to be a long project, especially if SE doesn't have a fixed range for all the colors (more RNG? LMAO).

You're right about the different RGB scales, the key is that we're using -/+ 5, and the difference in fruit is very minor for the majority. As mentioned it's really going to come down to personal preference as both scales have shown that course correction can be required.

Myself, Ravenguard, and a few others who were assisting player's early on in the Chocobo Dye thread noticed the " dye range"; I'm 98.5% sure it exists. It's evident in the fact that multiple players can start at the same colour (assuming all are attempting to get the same desired colour), feed the same fruits (in the same quantities/order etc.) and end up at different colours. It's also apparent when you see multiple players feed multiples of the same type of fruit but get their "feather's growing" message on different fruits (i.e. player 1 gets "feather's" message after x1 berry, player 2 gets "feather's" message after x3 berry, etc.) I'd also say this is evident in watching the players in the dark colour range feed the same fruits and dance around Soot Black. It makes me wonder if the darker dye ranges have a tighter dye range window.

Without knowing exact RGB values I doubt we'll ever truly be able to pinpoint down each dye range. Another consideration is that we can't even be sure if it takes all 3 of the RGB values to trigger a dye range or just 2 which is why we usually recommend a combo of fruits to adjust all 3 RGB values. Yay for more RNG in the math! >_<

Due to this I very much doubt that any Feed Suggestion will be able to be duplicated consistently across all players. There will be players who will try a suggestion that worked for someone else (assuming start/desired colour are the same) and it won't work for them.

Regarding "Micro Feeding", as I showed in my manual math in a previous post, whether you combine the fruits or not shouldn't really be an issue, the end RGB from all the fruits should equal the same number regardless of how it might be skipping around the dye ranges behind the scenes. With individuals stuck in colour loops, which again I think could potentially be just crappy dye range RNG, Micro Feeding might assist in breaking that loop. We were always recommending users who were stuck in the loop, or close to their desired colour range and trying to prevent a loop, to try small quantities of a single fruit type to try to avoid overshooting.

[Chopski]

I asked this question in the R&D thread, so i'll ask it here too just in case;

I'm curious, does anyone know where the RGB values for the FFXIV Chocobo Colors spreadsheet have come from?

I've just done a few tests using a colour dropper on some screen shots of the dye pallete and I'm getting different values.

eg; Woad Blue is stated as 41,50,74, but im getting it as 47,56,81. Any idea why this is?

[Giantbane]

especially if SE doesn't have a fixed range for all the colors (more RNG? LMAO).

I highly doubt the range is random. Possible, yes, but it's sort of a silly way to build it so I'm going to go out on a limb and say this isn't what they're doing unless we find some strong evidence to the contrary.

There's probably a single algorithm that converts from RGB > dye. The ranges would be dictated by that algorithm (whatever it may be).

[Chopski]

Yes, I need to figure that one out. Problem is, the calculation starts off by rounding to the nearest 10, purely because 2 fruits increase the increment by 10. eg; 1 plum + 1 valfruit = Blue +10.

Think I need to hold on to what it takes off, round that to the nearest 5 and see if collectively 1 more or less fruit would get closer to the final RGB values.

I have done this now, it was a very rare case during testing. But if the rounded remainder matches a fruit it will either add or subtract it from the recipe. I could possibly take it further but it requires a bit more thought. I have just done the "safe fine tuning" for now, which will take some recipes closer to their final values.

[AttacKat]

Should had read here first, oh well.

I'm curious, does anyone know where the RGB values for the spreadsheet have come from?

And IME, I can tell you any single pixel dropper will be wrong. Even with direct capture of the Dye UI, the game still does some form of color shading and this is especially noticeable with the lighter colors such as the pinks and the light blue/greens. Just zoom up to 500%+ and see for yourself.

I highly doubt the range is random. ... There's probably a single algorithm that converts from RGB > dye. The ranges would be dictated by that algorithm (whatever it may be).

The similator does mention this: http://lists.w3.org/Archives/Public/...lSep/0447.html
and I see it uses that in its formulation to determine the "current guess" color.

I have done this now, it was a very rare case during testing. But if the rounded remainder matches a fruit it will either add or subtract it from the recipe. I could possibly take it further but it requires a bit more thought. I have just done the "safe fine tuning" for now, which will take some recipes closer to their final values.

Cool ...

And I have more work for you ... LOL

Updated the RGB value to 4th rounding now, so all the values are 3, 7, 11, 15 ... I used GIMP which gave me the ability to take the average RGB of an area instead of a single point, so I changed the RGB based on a 7x7 area average. You can trash my old 8th scale from your calculator if you like.

[Chopski]

Updated the RGB value to 4th rounding now, so all the values are 3, 7, 11, 15 ... I used GIMP which gave me the ability to take the average RGB of an area instead of a single point, so I changed the RGB based on a 7x7 area average. You can trash my old 8th scale from your calculator if you like.

No problem, will sort.

And IME, I can tell you any single pixel dropper will be wrong. Even with direct capture of the Dye UI, the game still does some form of color shading and this is especially noticeable with the lighter colors such as the pinks and the light blue/greens. Just zoom up to 500%+ and see for yourself.

Are we talking about the same thing here or am I missing something? All these squares have solid colours in the center for me. Even the coloured circle on the companion window is a solid colour with no shading and has the same rgb value.

[Seiko769]

Okay need help. I'll actually be checking back this time btw XD okay after a lottttttaaaa work I finally managed to get to Regal purple. Still tryna get to Midnight Blue cause it looks great and its close to Soot black. Problem is I cant figure out how to get to it. Supposedly plums will get me to royal blue than apples over to Abyss than berries to Midnight BUT plums and apples are opposites so I just get juggled between Regal and Royal @.@ what do I gotta do here

[Syrehn]

Okay need help. I'll actually be checking back this time btw XD okay after a lottttttaaaa work I finally managed to get to Regal purple. Still tryna get to Midnight Blue cause it looks great and its close to Soot black. Problem is I cant figure out how to get to it. Supposedly plums will get me to royal blue than apples over to Abyss than berries to Midnight BUT plums and apples are opposites so I just get juggled between Regal and Royal @.@ what do I gotta do here

Starting Colour: Regal Purple
RGB Estimation: 95R, 42G, 71B

Desired Colour: Midnight Blue
RGB Estimation: 20R, 21G, 49B

Note: I used the "Original" RGB values with the +/-5 integer for my calculations.

Manual Math
Focused on bringing your RED down first and then the BLUE.

8 Pear (-40, +40, -40) 55, 82, 31
8 Berries (-40, -40, +40) 15, 42, 71
4 Apple (+20, -20, -20) 35, 22, 51
2 Pear (-10, +10, -10) 25, 32, 41
1 Berries (-5, -5, +5) 20, 27, 46 <- End RGB Value (approx.)

Combined Feed Option
You should see at least 1, if not more, "Feather's" messages by the time you're done.

x10 Pear
x9 Berries
x4 Apple

Micro Feed Option
You can break the feeds down if you want. You can do them something like the below (mimicking the manual math order) or break it down further.

x8 Pear
x8 Berries
x4 Apple
x2 Pear
x1 Berries

FFXIV Calculator Reference: The calculator generated a feed suggestion very close to the above; it just had x1 more Berry (did the manual math then cross checked the calculator).

You can choose if you want to mass feed or micro feed based on the above but either combo should result in an end RGB of around 20R 27G 46B (assuming +/-5 average integer and using the "Original" RGB values) and should push you close, if not into, the Midnight Blue dye range; your mileage may vary.

[Seiko769]

boy am I glad there are people on here smarter than me XD thanks for the help Ill be back in a few days with hopefully good results