A disclaimer of course: I COULD BE TOTALLY WRONG HERE! This isn't a video claiming "here is how it is done." Rather, I'm just explaining my research and observations, and attempting to form a hypothesis on how it was done. It could very well be that they are all hand-designed entirely, and all these "rules" are merely coincidence.... But that doesn't mean that has to be the end of the discussion. Let this be a starting point for those more knowledgeable on the matter. I haven't seen any other videos on this topic specifically, and I want more solid answers!
@@TheAetherleaf Dugtrio is an example of the 3D model being better than the original. The whole animation of the three of them all bouncing around is so adorable!
"This generation only had 4 colors in their pallets. Two of which were black and white, so each Pokemon only had 2 colors to work with." - Paraphrasing another of his quotes that disprove Diglett as evidence for Algorithm.
I'm honestly in the ring with you of there being both a general algorithm and hand picking. Only an algorithm would CHOOSE to make shiny jigglypuff/wigglytuff nasty hue shifts instead of white rabbits with red eyes.
Well then that's an argument against it being an algorithm as shiny jigglypuff is very similar to the jigglypuff character in the anime and the anime came out before gen 2 did in japan.
@@Merahki3863 Yeah I remember Playing Pokemon Yellow on my Gameboy colour and ALSO All the Original Games I think had Wierd shifts on GBC too if I remember right.... There was weirdness there that Isnt Often Revisited these days....
his point was that iconic pokemon like pikachu cant have too drastic of a change imagine a red or blue pikachu. but cool and more popular pokemon like charizard have cool shiny colors of course this doesnt apply to all of them (mewtwo and the legendary birds)
I can't really explain Garchomp, but Gengar used to look very different because it was hard to translate from 2D to 3D because its sprites used to look very different from other pokemon in the first gens. If you see it, it makes sense, however I agree that it's awful.
The bulbasaur line shinies are the most disappointing to me...bc hes my favorite and it's just not great imo. Sad. But of course shiny CHARIZARD looks incredible. Yes I'm salty.
To be fair, Gengar used to be way more pinkish, so the dark purple shiny color was way more exciting. They should've tweaked it, I blame HGSS for ruining Gengar
Yeah, that’s so [SLANG NOT FOUND]! I mean, I also love walking around on my leg and I breathe every day! Seriously, just [SCAN INCOMPLETE] with this [SLANG NOT FOUND]. Blink blink absorb water.
How much you wanna bet it was a glitch that occurred when they were testing the time capsule that allowed you to trade between generations and a couple gen 1 pokemon came out not looking right and they just rolled with it cause it looked cool
Shinies were intended. They're easter eggs to show off the game boy color's new capabilities. Since shinies were determined by DVs (IVs in later gens) in Gen 2 as opposed to the trainer ID, secret ID, and personality value, Gen 1 Pokemon could still be shiny if traded to gen 2 since they also had DVs. The ability for Gen 1 pokemon is an oversight, not shinies in general.
Their algorithm could have done all of these functions seperately, and then presented all the results to the developers and they selected from that list
But what if chrome was chrome but with chrome being chrome if chrome came and sat down ate some english coffee with some water and dogs on the moon and than I realised that earth is like more than 4 billion years old because of google chrome so I ate water and drank some moon rocks on mercury but that's not possible
Its probably something like: "Go run this algorithm, These pokemon however, Apply these fixes if it comes out looking off or if its one of these XYZ pokemon" Then the guy went through and the head dev went "Ohh i like those change these to these and add this line to the code to make it more variable"
It could be that the Artists had an algorithm to pick the colour but they hand coloured the sprites, this might explain why there are exceptions and NEAR misses So yeah, you've created a flow chart of awesome I know it might be wrong, but that doesn't matter
Shiny pokes use the same sprite as their standard version, only with an altered palette, so they can't be "hand coloured" per-se. The easiest way to "color" them would be to pop in a hex value, which is likely what an algorithm would produce. They could still be manually tweaked, or shifted a little as part of the algorithm, though.
In older games with sprites, color swaps are used to give more visual variety without having an entirely new image in the code. Especially for a GBC game, *space* is extremely limited.
@@PissedGrunty With that in mind, it's surprising they didn't do more color variations, or made shinies less rare. Shinies are a visual variety, but one that barely ever appears.
@@voliol8070 Yeah. Do what they do for the Florges line and give each mon a variety of different colors. Heck, make a 2nd type of shiny where the colors are entirely based on the secret codes and totally random. You'd get some damn weird results, but it would be interesting.
If this is true as well as having the random shifts and changes, I think overall shiny quality would have been greatly improved by having 3 results generated for each line and having someone make a 10 second decision on which to go with.
@@wesnohathas1993 They wouldn't put in that kind of time *now.* GameFreak weren't that bad back in Gens 1-5. In the first two generations they had a lot to lose. They didn't want to make a shitty sequel to such a beloved and well received game. They cared about the games for a lot of generations, even past the point where they didn't need to worry about the game flunking. Then they realized that and stopped caring because the games would make money anyway. Too much money, Sword and Shield have no right being the second best selling Pokemon games with their shitty quality.
Lockstin! Make a series where you redesign shinies from Gens 1-5 to fit some sorta reference or real world reasoning! I, aswell as others, would be very interested to see what you come up with
YES YES YES. Lockstin doing this would be perfect! I think most people would simply just choose the colors that look the best but I know for sure Lockstin would pick colors that reference something which would be interesting!
About "diglett's dirt": I think diglett was never supposed to have different colour dirt, it's simply that due to colour pallette limitations diglett's dirt was the same colour as the nose, so if you change the nose the dirt also changes. In 3rd gen thanks to better hardware the dirt and nose use different colours so by changing one the other doesn't automatically
I wanna see what all the newer shiny Pokémon would look like with the algorithm you’ve found, like designing Pokémon’s shinies based on your algorithm.
@@cyberwill2364 Well yeah but Scyther is, if we accept the algorythm hypothesis as true, then it's not unrealistic to think that it would follow the same rules as the first stage for consistency sake.
I could have seen them way back when just running the sprites through a couple different color changing algorithms and picking and choosing what they liked, with a couple exceptions if none of the color variations looked very good.
Just dropping a comment to say I appreciate the amount of work that goes into these videos. Even if it IS just a theory it is fascinating to look into and learn. Over heard broadening my horizons in quarantine.
I might as well reach out for an answer: The Spaceworld 1997 sprites have completely different regular and shiny palletes that could possibly follow one of the patterns *exactly*. So... if you wanna look at those and make a followup or just reply, there's an idea.
I was thinking of that too, e.g. all pink Pokémon in Spaceworld have blue shinies, all green mons have desaturated yellow shinies, and all yellow tan/orange ones.
I just got back into the Pokemon games recently since the first gen came out when I was kid (currently playing Pokemon Black) and happened to discover your channel. I really appreciate the effort and research that you put into your videos as well as your presentational style. This one really blew me away though. Thanks for being a great youtuber. I just wanted to show my appreciation. You're awesome!
Gyarados’s red shiny is canonically because team rocket forced the magikarp in the lake of rage to evolve prematurely, causing the gyarados to retain its magikarp colors. That’s why you find a shiny one there
Auburn Autumn well the shiny was most likely intentionally designed to fit the lore. Kind of like how Charizard’s shiny was intentionally changed to make it more appealing.
I was thinking about a simplified chart. An algorythm could generate multiple choices for the developers to pick the one they prefer (for example "do you prefer left shift or right shift?") although the few disgusting exceptions are a counter example to this solution. It would be quite manageable with 251 for the 2nd gen.
Who else thinks that they should bring back the shiny ditto transforming into a shiny version of the pokemon its fighting? Edit: the ditto thing was from gen 4, and after that it just poofed.
That was in Gen 3, not Gen 4. Also, if Ditto doesn’t keep it’s shininess while turning into a Pokémon, it’s because it copies the opponent perfectly. If the opponent is shiny, the transformed ditto will be too
I've been watching this video since it has come out and I just noticed a flaw, you're using the Crystal sprites here. Spinarak and Sneasel's original colors changed in Crystal but their shinies stayed exactly the same(some other pokemon changed colors but only with minor differences.). So instead of Sneasel being a purply-blue that turns pink, it's an orangish-brown that turns pink. I'm not sure if this matches up with the triads or anything since I haven't tested it out yet but I'd still like to point it out.
So in most scenarios (besides dev tempering) Even tho pokemon have changed colors in future games from the original versions From GS to Crystal but also from Crystal to Gen 3. In the majority of the cases their shinies keep the same colorpalet Thats why its good that he tested it on gen 2
Tommy Cetee Not learn the move fly but actually fly, like, flap their wings and hover and even fly around. Many Pokémon that can fly don’t learn fly smh
The sprites popping in and out, the explation of technical attributes, and the music! Man, this video was so well made. All the way down to the impromptu phone call.
"Well, actually, Lockstin, you're referring to Misdreavus when you say Mismagius..." Damn, I love this character. I can point all the errors in a video and the snobby know-it-all takes the blame... Oh wait...
I’m late to the party here but this is an extremely cool reconstruction of an algorithm from its inputs and outputs, and you did an excellent job taking us through your thought process. I do hope we get some official answer on how early shinies were made. If your flowchart is right, I’d love to hear how the devs came up with things like the type rules and the pink rule
I thought this was just going to be a surface-level overview or something, didn't realize that it was over half an hour long. It's definitely really compelling, and even if this wasn't how it was done or there wasn't an algorithm the sprites probably were still following a ruleset somewhere along these lines - there's too many coincidences and weird parts otherwise.
My favorite line of the whole video by far was "Look at Phanpy! The blue is less saturated, and the red is more! But neither by much. This is the worst shiny."
My dude Lockstin putting in that effort. My guy, I would have done the same just to see if it made sense. Thanks for using your Gnoggin so I can use mine somewhere else.
I love these videos, Lockstin, all the explanation and almost Color Theory kind of info, so fun to watch! Also why I love your language and sound videos so much!
I just noticed that the color schemes of the shinies for the legendary birds are nearly identical to that of their Galarian forms. It could have been an inspiration for the galar forms.
Just for the Devil's advocate sake: If given a Pokémon, and following your chart, how many outcomes can this Pokémon reach, out of the total color swap outcomes possible (given gameboy's limited pallet)? (aside from developers tweaks) Which is to say, if you derive a rule that end up being capable of reaching every color that there is available, this has a very low chance of correlation. In other words, what is your "p value"?
Yeah I agree, if I understand you correctly, it's true that there are so many variables and his colors aren't perfect, I think there's an algorithm and that he's close but I'll love to have a funtional program or an interview to know exactly what's going on
I don’t think the flow chart he made ends up giving every Pokémon a chance of being any possible color. I think it only allows for a few outcomes depending on RNG
@@psychoticcoconut4047 well I'm just thinking about H2O. two hydrogen (dihydrogon) and one oxygen (monoxide), though I could be entirely wrong. I haven't done chemistry since 9th grade haha
@@angusperson4222 in science class the teacher said when naming a compound made up of a non metal and another non metal then only the last element is named using the mono-, di-, tri-, or tetra-
15:45 JUST realized the gen 2 shiny colors of the legendary birds are pretty close to the Galarian form's standard colors 🤯 Well, Zapdos and Moltres at least. Articuno gets a little purple in it's shiny, but not as much as Galarian Articuno.
Yeah there's some consistency with the default colors of a regional form resembling the shiny colors of the original forms, and vise versa. You can see this most clearly with Alolan grimer and shiny kanto grimer, and shiny alola grimer and normal kanto grimer. Also galarian slowbro uses some of the colors of shiny kanto grimer.
You should explain why pokemon were chosen for their gen, like copperajah for galar. There are a few pokemon I believe poses this same question: "Why is X chosen for this region, which is based on Y, when their origins seem to have nothing to do with Y?" A few I can think of is Copperajah, the genies, and Cofagrigus
Speaking of exceptions to the algorithm: Raichu's shiny and normal body colors were swapped after Gen 2, and Spinarak and Sneasel had different normal colors in Gold/SIlver before being changed in Crystal. Spinarak had extra tampering in Gen 3, where its shiny was changed to more closely resemble its normal colors from Gold/Silver.
Really enjoyed this video. As a computer science student, this seems definitely very plausible and I would not think Devs would go into the trouble of hand picking every shiny color. Subscribed!
I feel like some of the reasons for some colors looking like an algorithm did it (the diglet in gen 2), but also there is the arguable reasoning that they just used less palette limits, and they compressed that usage in the normal diglet's color, putting red as partial shade,nose, and dirt color, so if a developer would want to change the nose to blue, its gonna change the rest too, so either the dev would have to redo and see if they can even add additional palettes to diglet or just go the safer route and just change that one palette to desired color, regardless of whatever else changes color, its just dirt and didnt look bad by old standards, so thats a second possibility. They also couldve followed a standardized format for colors, and then did specific alterings on certain pokemon.
For my work I do a lot of retroactively fitting data to unknown algorithms and the process Lockstin used is actually similar to how I would go about a similar problem - very interesting
8:41. His ringtone is Gourmet race from Kirby which is my favourite franchise created by Masahiro Sakurai *HE'S MORE THAN A POKÉMON FAN!!!! HE'S A KIRBY FAN TOO!!!!!* Just like me yay!
If you look back through his channel you'll see he wasn't always a mostly Pokémon channel. He's done many theories on many franchises across all of gaming, and not even just Nintendo.
One thing to note about "same colors" is that in photoshop they might be the same but in game they might be different. For instance, if you have two pokemon that share colors, they might actually has that palette swapped but still used artistically the same; White/Purple/Yellow/Black into White/Yellow/Purple/Black. It's visually the same (the two pokemon use purple in similar places), but if the algorithm affects the second and third color differently, then when they are shifted they come out different. To explain this a different way, sprites in old games had two pieces of data; Where pixels go, and which palette to apply. The pixel data would use numbers; This pixel is the color "3", this pixel is "1", this is "0", etc. Then the palette gets loaded which says "0" is white, "1" is red, "3" is purple, etc. If you load a different palette (a literal palette swap), then instead you could get white, blue, and green (importantly, 0 is generally white/transparent). You can see this happen with glitches where a sprite or tileset gets mixed with the wrong palette that wasn't built in the same order (one palette might put light colors where another put dark) and thus looks incredibly wrong. You can even see this in one of the GBC pokemon games because there is a normally inaccessible palette chooser in the data that the Coin Case can access.
The Charizard "factoid" drives me crazy, look at the original shiny promo art from 2001 that has a shiny Charizard. It's black and red. The in-game sprite was only that color by game limitations. So many odd "color choices" in GSC are clearly influenced by all the weird pallete rules, and are not necessarily 1 to 1 representations of the Pokemon's "actual" color. Shiny Swinub doesn't _actually_ have those ugly red dots.
I'm more mad people think the purple and green was bad It wasnt It's not nearly as good but it's not bad And at least it changed the flame color Something the redesign didnt do
Here's my opinion: Shinny pokemon were only made to make a higher demand for pokemon by making them rare and have better stats. Basically making a palette swap and making an artificial demand by making them rare.
They don't even have better stats. In Gen 2, shininess was determined from IVs, which at the time ranged from 0 to 15. For a Pokemon to be shiny, Speed, Defense, and Special needed to be 10, while Attack needed to be one of several seemingly random values ranging from 2 to 15. HP was calculated from the other IVs, so that was always either 0 or 8. So while shiny Pokemon had kinda good stats, they weren't great. And then come Gen 3, shininess was completely divorced from IVs and no longer has anything to do with stats.
This is still one of my favorite videos that I love to revisit every so often because I enjoy the process behind determining how the rules could’ve worked. Like, I never even heard of color wheel triads until I watched this all those years ago. And also it’s the copium needed to understand why they did that to my poor Espeon.
The creator of shiny Espeon could easily just like the color :p There probably was some kind of formula or algorithm as a basis but as an artist myself, sometimes we pick colors just cause we like them. Not everything has a reason :p
When there are 200+ new pokemon every game, theres no way a large portion of the pallete changes arent automated. These are programmers we're talking about.
This is really cool, think its the first time some went to in depth to how shinies were made, and thank god you did it for only one week, I feel like my brain would rot from just a bit more than a week
31:29 Agreed, Espeon needs a new shiny. It's current one makes it look like it's radioactive or absorbed the Master Emerald and became some weird Chaos God Thing.
3:27 because that would take up a _lot_ more code in the game files than just putting the algorithm in, if there was one. Duplicating sprites _always_ takes up more code than a pallete swap algorithm.
This video is so interesting ♡ I would love to see your reaction when you see gen 2 shinies through pokemon Stadium 2 cause they change drastically and that "Barnie" Charizard looks pretty good xD (Also just looking at those classic animations-I promise it won't be a waste of your time ♡)
Given how bad some early shinies are I dont doubt they were made by an algorithm. Now to get that flowchart and make it into an actual algorithm to test it out
I for one believe you're correct (always have, because it's that or some dev REALLY hated poor Jolteon). But I can't help but feel like, somewhere in Japan, a dev from Gen 2 is watching this going "Actually, we just used a randomizer to pick colors and ran with that."
Because their is no public algorithm, at least not a perfect one, if you consider the Flowshard as a algorithm. Pay attention to the Video you watched, please. And also Gen 5 was still made with the algorithm. Gen 6 started to take effort in shiny designs
Aaaaay the return of the Dragon shirt!!! Would love a rerun on the Dragon and Fairy type shirts, kicking myself for not picking them up when they were available
*when lockstin starts talking about complemetary colour and complementary triads*oh i know about that*starts talking about hsb* this know feels like a class
A disclaimer of course: I COULD BE TOTALLY WRONG HERE! This isn't a video claiming "here is how it is done." Rather, I'm just explaining my research and observations, and attempting to form a hypothesis on how it was done. It could very well be that they are all hand-designed entirely, and all these "rules" are merely coincidence.... But that doesn't mean that has to be the end of the discussion. Let this be a starting point for those more knowledgeable on the matter. I haven't seen any other videos on this topic specifically, and I want more solid answers!
uwu
Ello Mate
Haztac WaterCultPopcornCult no
Nice
Im here
New Good Quote: "An algorithm wouldn't be able to tell the difference between what is dirt and what is Diglett."
Neither can I, honestly.
...Diglett isn't anyone's favourite, right?
@@Halophage I mean dugtrio is my 20th favorite pokemon
@@TheAetherleaf Dugtrio is an example of the 3D model being better than the original. The whole animation of the three of them all bouncing around is so adorable!
@@N_IRL
Ikr they're Adorable
The noses booping in and out is also adorable
"This generation only had 4 colors in their pallets. Two of which were black and white, so each Pokemon only had 2 colors to work with."
- Paraphrasing another of his quotes that disprove Diglett as evidence for Algorithm.
“We humans, which I am”
New Pokemon Theory: Lockstin is a Zororark.
Maybe a ditto
@@moon9771 Nah, his eyes aren't small enough.
Latias/latios?
Maybe a mew?
@@antonpershin998 no, ash would have kissed him. But I like the theory.
"We humans, which I am."
🤔 Sounds like something a non-human would say
We have known he is not of us for a long time I believe.
@Fredd's Dead 1011010010
That's a running joke on the channel
He’s being controlled by whatever is inside that bouncy hair mound
🎂
I'm honestly in the ring with you of there being both a general algorithm and hand picking. Only an algorithm would CHOOSE to make shiny jigglypuff/wigglytuff nasty hue shifts instead of white rabbits with red eyes.
Well then that's an argument against it being an algorithm as shiny jigglypuff is very similar to the jigglypuff character in the anime and the anime came out before gen 2 did in japan.
@@Merahki3863 Yeah I remember Playing Pokemon Yellow on my Gameboy colour and ALSO All the Original Games I think had Wierd shifts on GBC too if I remember right.... There was weirdness there that Isnt Often Revisited these days....
Yeah no
Lockstin heres another series idea aswell- use your flow chart to calculate what shinies couldve looked like in gens 6-8 using these methods.
I really want to see that. I wonder which Pokemon get the Garchomp treatment or Torchic treatment.
This would be cool!
Ooooo that would be so cool!
Good idea
This please
"Or making sure important Pokémon have decent shinies"
Makes sense
"Like Pikachu"
You lost me
Pikachu is the mascot...
@Wolf World Blackyy oooh i get it now i thought the original was about how pikachu isnt an important pokemon lol
his point was that iconic pokemon like pikachu cant have too drastic of a change
imagine a red or blue pikachu. but cool and more popular pokemon like charizard have cool shiny colors
of course this doesnt apply to all of them (mewtwo and the legendary birds)
"Like charizard" would make more sense, as its shiny was altered after its creation to appear cooler
It is litterally decent. Like, it's not puke green like some shinies, so at least it doesn't make you sick but it isn't groundbreaking either.
Game freak: shinys will be a fun thing for people to collect, but it shouldn’t be an extreme passion
Shiny hunters: that’s where you are wrong
I think the new Anime finally introduced it's first Shiny Hunter.
@@Nosretep that certainly took a while if true, given that the first shiny in the anime was (I think) ash's noctowl back in johto
To be fair, Game Freak doesn't know anything about their fans
To be fair, Gamefreak didn't think Pokemon would even last past the 2nd gen...
@@zachrodan7543 You'd think they have introduced something like that during Jhoto...
no human with a heart could ever create the crimes against humanity that are gengars and garchomps' shinies
I can't really explain Garchomp, but Gengar used to look very different because it was hard to translate from 2D to 3D because its sprites used to look very different from other pokemon in the first gens.
If you see it, it makes sense, however I agree that it's awful.
The bulbasaur line shinies are the most disappointing to me...bc hes my favorite and it's just not great imo. Sad. But of course shiny CHARIZARD looks incredible. Yes I'm salty.
Leafeon disappointed me, also IMO they should’ve booted up the algorithm again for the Eeveelution’s shinies, especially leafeon and glaceon
To be fair, Gengar used to be way more pinkish, so the dark purple shiny color was way more exciting. They should've tweaked it, I blame HGSS for ruining Gengar
Pore gengar and garchamp,at least Gengar got AMAZING mega and gigantamax shinies
Lockstin: we humans - which I am.
Me: hmmm how much longer can he keep us from the truth I wonder?
What truth? He is clearly human, there is nothing to hide here. We are all just earthlings- I mean humans- here! :D
@@catattack9575 ah, yes... I mean, of course! Just like the talking mammals from the planet Earth.
Yeah, that’s so [SLANG NOT FOUND]! I mean, I also love walking around on my leg and I breathe every day!
Seriously, just [SCAN INCOMPLETE] with this [SLANG NOT FOUND]. Blink blink absorb water.
Given how broken the first two generations are, I wouldn’t be surprised if they were just a bug they turned into a feature
If that’s true then why is their a sound and effect
@@thischannelisretiring1648 hence "turned into a feature" they could have liked what the bug did and did some working on it to make it a feature.
How much you wanna bet it was a glitch that occurred when they were testing the time capsule that allowed you to trade between generations and a couple gen 1 pokemon came out not looking right and they just rolled with it cause it looked cool
Shinies, the creepers of Pokémon (maybe)
Shinies were intended. They're easter eggs to show off the game boy color's new capabilities. Since shinies were determined by DVs (IVs in later gens) in Gen 2 as opposed to the trainer ID, secret ID, and personality value, Gen 1 Pokemon could still be shiny if traded to gen 2 since they also had DVs. The ability for Gen 1 pokemon is an oversight, not shinies in general.
THE COLORS LOCKSTIN, WHAT DO THEY MEAN?
HEY...!? WHO ARE YOU PEOPLE!?!? *WHAT DO YOU WANT FROM ME?!!?*
The Luigi who Saw Things.
WE JUST WANT THE COLORS. THATS ALL WE’VE EVER WANTED.
@@liamfairbanks4861 _We needed their knowledge, what they'd learned._
DOUBLE RAINBOW ALL THE WAY
Their algorithm could have done all of these functions seperately, and then presented all the results to the developers and they selected from that list
It's obvious that they are from the future, and that's why they are shiny. Because everything is chrome in the future.
Google chrome is Google chrome chrome in the future
Google chrome is Google chrome chrome in the future or chrome Google chrome
EVERYTHING IS CHROME IN THE FUTURE
But what if chrome was chrome but with chrome being chrome if chrome came and sat down ate some english coffee with some water and dogs on the moon and than I realised that earth is like more than 4 billion years old because of google chrome so I ate water and drank some moon rocks on mercury but that's not possible
FUUUUTURRRRRE!!!
FUUUUTURRRRRE!!!!!
My favorite thing is when Lockstin plays the invisible piano with his hands when he really wants to make a point.
Its probably something like: "Go run this algorithm, These pokemon however, Apply these fixes if it comes out looking off or if its one of these XYZ pokemon" Then the guy went through and the head dev went "Ohh i like those change these to these and add this line to the code to make it more variable"
It could be that the Artists had an algorithm to pick the colour but they hand coloured the sprites, this might explain why there are exceptions and NEAR misses
So yeah, you've created a flow chart of awesome
I know it might be wrong, but that doesn't matter
Shiny pokes use the same sprite as their standard version, only with an altered palette, so they can't be "hand coloured" per-se. The easiest way to "color" them would be to pop in a hex value, which is likely what an algorithm would produce.
They could still be manually tweaked, or shifted a little as part of the algorithm, though.
In older games with sprites, color swaps are used to give more visual variety without having an entirely new image in the code. Especially for a GBC game, *space* is extremely limited.
@@PissedGrunty With that in mind, it's surprising they didn't do more color variations, or made shinies less rare. Shinies are a visual variety, but one that barely ever appears.
@@voliol8070 Yeah. Do what they do for the Florges line and give each mon a variety of different colors. Heck, make a 2nd type of shiny where the colors are entirely based on the secret codes and totally random. You'd get some damn weird results, but it would be interesting.
If this is true as well as having the random shifts and changes, I think overall shiny quality would have been greatly improved by having 3 results generated for each line and having someone make a 10 second decision on which to go with.
Nah, this is GameFreak we're talking about. They'd never put in that kind of time.
@@wesnohathas1993 They wouldn't put in that kind of time *now.* GameFreak weren't that bad back in Gens 1-5. In the first two generations they had a lot to lose. They didn't want to make a shitty sequel to such a beloved and well received game. They cared about the games for a lot of generations, even past the point where they didn't need to worry about the game flunking. Then they realized that and stopped caring because the games would make money anyway. Too much money, Sword and Shield have no right being the second best selling Pokemon games with their shitty quality.
Lockstin! Make a series where you redesign shinies from Gens 1-5 to fit some sorta reference or real world reasoning! I, aswell as others, would be very interested to see what you come up with
YES
Great idea 👍........YES
Good idea
YES YES YES. Lockstin doing this would be perfect! I think most people would simply just choose the colors that look the best but I know for sure Lockstin would pick colors that reference something which would be interesting!
I want to like this, but cannot as it has the sacred number (69... nice) for likes already, so I will simply say in here that I agree
About "diglett's dirt": I think diglett was never supposed to have different colour dirt, it's simply that due to colour pallette limitations diglett's dirt was the same colour as the nose, so if you change the nose the dirt also changes. In 3rd gen thanks to better hardware the dirt and nose use different colours so by changing one the other doesn't automatically
I wanna see what all the newer shiny Pokémon would look like with the algorithm you’ve found, like designing Pokémon’s shinies based on your algorithm.
ooh that sounds cool he totally should. I bet they'd mostly end up puke green and highlighter pink though lol
"Porygon 2, Cobalt Blue"
No no, its Porygon 2 the electric boogaloo
Electric Blue-galoo?
@@shytendeakatamanoir9740 nooo you beat me to it
@@absonaura Sorry. Great minds think alike
More like, epileptic boogaloo
@@OldRock84 He was sacrificed to save the mascot Pikachu #PorygonDidNothingWrong
Lockstin: Oh yes. Right. Human. The thing we are. I am human. Of course. What else could I be.
*Glances at pokemon are people vids(tho I don’t remember is Lockstin himself made one oop)*
Hmph, humans...
I know right?
Yes, I to am human and enjoy human activities like walking on my leg attached to my foot which keeps leg on body
Yes. Humans. We are Humans. Humans are fRieNDs. We are friENds
“An algorithm wouldn’t be able to tell what is dirt and what is digglit.” Thank you for that
So, who's gonna tell him that Beedrill's *NOT* a flying type?
Edit: And that Paras/Parasect is *NOT* a Poison type.
I had to tell him that he called Misdreavus the name of its evolved form.
And scizor isn't a flying type
@@cyberwill2364 Well yeah but Scyther is, if we accept the algorythm hypothesis as true, then it's not unrealistic to think that it would follow the same rules as the first stage for consistency sake.
Yeah neither is scizor
the word "Type" has multiple meanings, I think when he said flying type. the "Type" meant "category" instead of the "pokemon element".
I could have seen them way back when just running the sprites through a couple different color changing algorithms and picking and choosing what they liked, with a couple exceptions if none of the color variations looked very good.
oh speaking of shinies, i'd love to see something about the kaskade shinies at some point, and ofc great vid as always :)
Just dropping a comment to say I appreciate the amount of work that goes into these videos. Even if it IS just a theory it is fascinating to look into and learn. Over heard broadening my horizons in quarantine.
I might as well reach out for an answer:
The Spaceworld 1997 sprites have completely different regular and shiny palletes that could possibly follow one of the patterns *exactly*. So... if you wanna look at those and make a followup or just reply, there's an idea.
I was thinking of that too, e.g. all pink Pokémon in Spaceworld have blue shinies, all green mons have desaturated yellow shinies, and all yellow tan/orange ones.
I just got back into the Pokemon games recently since the first gen came out when I was kid (currently playing Pokemon Black) and happened to discover your channel. I really appreciate the effort and research that you put into your videos as well as your presentational style. This one really blew me away though. Thanks for being a great youtuber. I just wanted to show my appreciation. You're awesome!
Gyarados' shiny theme makes sense, because they're already always enraged, so being even more enraged is rare, and makes them red.
Gyarados’s red shiny is canonically because team rocket forced the magikarp in the lake of rage to evolve prematurely, causing the gyarados to retain its magikarp colors. That’s why you find a shiny one there
What came first: the lore or the shiny? The world may never know.
Auburn Autumn well the shiny was most likely intentionally designed to fit the lore. Kind of like how Charizard’s shiny was intentionally changed to make it more appealing.
@@Chanos945 Canonically? What? I know it was forced to evolve in gen 2 games BUT there's no evidence it keeps red color because of it.
I was thinking about a simplified chart. An algorythm could generate multiple choices for the developers to pick the one they prefer (for example "do you prefer left shift or right shift?") although the few disgusting exceptions are a counter example to this solution.
It would be quite manageable with 251 for the 2nd gen.
Who else thinks that they should bring back the shiny ditto transforming into a shiny version of the pokemon its fighting?
Edit: the ditto thing was from gen 4, and after that it just poofed.
That was in Gen 3, not Gen 4. Also, if Ditto doesn’t keep it’s shininess while turning into a Pokémon, it’s because it copies the opponent perfectly. If the opponent is shiny, the transformed ditto will be too
@@carotteatomique no. it was gen 2. lol
@@oucyan no its gen 1 lol
@@juliandamen9804 no its gen 0 lol
@@RedHinge no it's gen - 2 lol
I've been watching this video since it has come out and I just noticed a flaw, you're using the Crystal sprites here. Spinarak and Sneasel's original colors changed in Crystal but their shinies stayed exactly the same(some other pokemon changed colors but only with minor differences.). So instead of Sneasel being a purply-blue that turns pink, it's an orangish-brown that turns pink. I'm not sure if this matches up with the triads or anything since I haven't tested it out yet but I'd still like to point it out.
So in most scenarios (besides dev tempering)
Even tho pokemon have changed colors in future games from the original versions
From GS to Crystal but also from Crystal to Gen 3.
In the majority of the cases their shinies keep the same colorpalet
Thats why its good that he tested it on gen 2
14:31 ah yes, beedrill and scizor. My favorite flying types
tbf scyther is a flying type.....beedrill though...
ah yes, paras and parasect. My favorite poisen types
Rais Aminuddin Can fly
@@f.b.iagent3971 no, only Volcarona, Genesect, and Vikavolt are the only bug types that learn fly
Tommy Cetee Not learn the move fly but actually fly, like, flap their wings and hover and even fly around. Many Pokémon that can fly don’t learn fly smh
The sprites popping in and out, the explation of technical attributes, and the music! Man, this video was so well made. All the way down to the impromptu phone call.
"Well, actually, Lockstin, you're referring to Misdreavus when you say Mismagius..."
Damn, I love this character. I can point all the errors in a video and the snobby know-it-all takes the blame... Oh wait...
I tapped out of Pokémon after Gen 3. I was thoroughly confused by that.
I’m late to the party here but this is an extremely cool reconstruction of an algorithm from its inputs and outputs, and you did an excellent job taking us through your thought process.
I do hope we get some official answer on how early shinies were made. If your flowchart is right, I’d love to hear how the devs came up with things like the type rules and the pink rule
Thank you for making videos, especially during this quarantine (:
You know it’s serious when the intro plays just over 8 minutes in.
I thought this was just going to be a surface-level overview or something, didn't realize that it was over half an hour long. It's definitely really compelling, and even if this wasn't how it was done or there wasn't an algorithm the sprites probably were still following a ruleset somewhere along these lines - there's too many coincidences and weird parts otherwise.
My favorite line of the whole video by far was "Look at Phanpy! The blue is less saturated, and the red is more! But neither by much. This is the worst shiny."
My dude Lockstin putting in that effort. My guy, I would have done the same just to see if it made sense. Thanks for using your Gnoggin so I can use mine somewhere else.
Wow, this is cool. I'd love to see a follow-up where you use this flowchart to remake some shinies, or do classic shiny coloration for later Pokémon.
Looks at shiny espeon: "Look how they massacred my girl".
One name, one series: Carbuncle, Final Fantasy. That is the reference I see with shiny Espeon.
They did my man krookodile dirty :(
Sad garchomp noises
Shiny espeon is the best shiny eeveelution
Chuchis Garza
Umbreon still exists
I love these videos, Lockstin, all the explanation and almost Color Theory kind of info, so fun to watch! Also why I love your language and sound videos so much!
I just noticed that the color schemes of the shinies for the legendary birds are nearly identical to that of their Galarian forms.
It could have been an inspiration for the galar forms.
No they aren't,!? It's only a tiny difference and the Galar forms are totally different
If I had mad respect for him before, then I have INSANE respect for him now.
Is this going to be the story of the Pidgeys and the Beedrills?
At Least It Isn't About The Pidgeys And The Aerodactyls.
Shiny Gible : love it
Shiny Garchomp : wait is that even a shiny?
Just for the Devil's advocate sake: If given a Pokémon, and following your chart, how many outcomes can this Pokémon reach, out of the total color swap outcomes possible (given gameboy's limited pallet)? (aside from developers tweaks)
Which is to say, if you derive a rule that end up being capable of reaching every color that there is available, this has a very low chance of correlation. In other words, what is your "p value"?
Yeah I agree, if I understand you correctly, it's true that there are so many variables and his colors aren't perfect, I think there's an algorithm and that he's close but I'll love to have a funtional program or an interview to know exactly what's going on
I don’t think the flow chart he made ends up giving every Pokémon a chance of being any possible color. I think it only allows for a few outcomes depending on RNG
Me before watching these vids: 😀I like drinking water
Me after: I prefer to consume hydrogen monoxide in liquid state obviously 🧐
wouldn't it be dihydrogen monoxide?
@@angusperson4222 well hydrogen is already diatomic on its original form, maybe your right. I'm just in 9th grade
¯¯\(°v°)/¯¯
@@psychoticcoconut4047 well I'm just thinking about H2O. two hydrogen (dihydrogon) and one oxygen (monoxide), though I could be entirely wrong. I haven't done chemistry since 9th grade haha
agnus person, no, you're right, its dihydrogen monoxide
@@angusperson4222 in science class the teacher said when naming a compound made up of a non metal and another non metal then only the last element is named using the mono-, di-, tri-, or tetra-
15:45 JUST realized the gen 2 shiny colors of the legendary birds are pretty close to the Galarian form's standard colors 🤯 Well, Zapdos and Moltres at least. Articuno gets a little purple in it's shiny, but not as much as Galarian Articuno.
Yeah there's some consistency with the default colors of a regional form resembling the shiny colors of the original forms, and vise versa. You can see this most clearly with Alolan grimer and shiny kanto grimer, and shiny alola grimer and normal kanto grimer. Also galarian slowbro uses some of the colors of shiny kanto grimer.
"Differences between what is dirt and what is Diglett." I don't know why but I love that.
It would be interesting to watch a video of Lockstin try to design new colours for previously existing shinys
You should explain why pokemon were chosen for their gen, like copperajah for galar. There are a few pokemon I believe poses this same question: "Why is X chosen for this region, which is based on Y, when their origins seem to have nothing to do with Y?" A few I can think of is Copperajah, the genies, and Cofagrigus
you spend so much time on us man. you don't need to spend a whole week doing these things lol
Speaking of exceptions to the algorithm: Raichu's shiny and normal body colors were swapped after Gen 2, and Spinarak and Sneasel had different normal colors in Gold/SIlver before being changed in Crystal. Spinarak had extra tampering in Gen 3, where its shiny was changed to more closely resemble its normal colors from Gold/Silver.
if you had to ask me bellosum shiny references vileplume because it's one of the first evolution to have a different evolution if that makes sense
Really enjoyed this video. As a computer science student, this seems definitely very plausible and I would not think Devs would go into the trouble of hand picking every shiny color. Subscribed!
thank you for posting i was running out of stuff to watch
14:59 Dragonite's gen 2 shiny is so much cooler than the one we have today
>the devs making sure important pokemon have decent shinies
Guess the Eeveelutions don't count as important... Friggin' green Espeon! Blegh!
Green espeon is beautiful
I feel like some of the reasons for some colors looking like an algorithm did it (the diglet in gen 2), but also there is the arguable reasoning that they just used less palette limits, and they compressed that usage in the normal diglet's color, putting red as partial shade,nose, and dirt color, so if a developer would want to change the nose to blue, its gonna change the rest too, so either the dev would have to redo and see if they can even add additional palettes to diglet or just go the safer route and just change that one palette to desired color, regardless of whatever else changes color, its just dirt and didnt look bad by old standards, so thats a second possibility. They also couldve followed a standardized format for colors, and then did specific alterings on certain pokemon.
Love that the intro played 8 minutes in XD
For my work I do a lot of retroactively fitting data to unknown algorithms and the process Lockstin used is actually similar to how I would go about a similar problem - very interesting
8:41. His ringtone is Gourmet race from Kirby which is my favourite franchise created by Masahiro Sakurai
*HE'S MORE THAN A POKÉMON FAN!!!! HE'S A KIRBY FAN TOO!!!!!*
Just like me yay!
He needs more kirby theories then
If you look back through his channel you'll see he wasn't always a mostly Pokémon channel. He's done many theories on many franchises across all of gaming, and not even just Nintendo.
One thing to note about "same colors" is that in photoshop they might be the same but in game they might be different. For instance, if you have two pokemon that share colors, they might actually has that palette swapped but still used artistically the same; White/Purple/Yellow/Black into White/Yellow/Purple/Black. It's visually the same (the two pokemon use purple in similar places), but if the algorithm affects the second and third color differently, then when they are shifted they come out different.
To explain this a different way, sprites in old games had two pieces of data; Where pixels go, and which palette to apply. The pixel data would use numbers; This pixel is the color "3", this pixel is "1", this is "0", etc. Then the palette gets loaded which says "0" is white, "1" is red, "3" is purple, etc. If you load a different palette (a literal palette swap), then instead you could get white, blue, and green (importantly, 0 is generally white/transparent). You can see this happen with glitches where a sprite or tileset gets mixed with the wrong palette that wasn't built in the same order (one palette might put light colors where another put dark) and thus looks incredibly wrong. You can even see this in one of the GBC pokemon games because there is a normally inaccessible palette chooser in the data that the Coin Case can access.
The Charizard "factoid" drives me crazy, look at the original shiny promo art from 2001 that has a shiny Charizard. It's black and red. The in-game sprite was only that color by game limitations. So many odd "color choices" in GSC are clearly influenced by all the weird pallete rules, and are not necessarily 1 to 1 representations of the Pokemon's "actual" color. Shiny Swinub doesn't _actually_ have those ugly red dots.
I'm more mad people think the purple and green was bad
It wasnt
It's not nearly as good but it's not bad
And at least it changed the flame color
Something the redesign didnt do
Black and purple I get but how exactly are game limitations supposed to turn red into green
This is something I have spent WAY too much time trying to figure out myself. So glad you finally made this video
16:52 Paras and Parasect aren’t poison types but probably would be if three types were introduced
Or they would be Bug-Grass-Ghost types, cuz they're basically the zombified husks of a once-living bug, now controlled by a mushroom.
Fun fact: they were the only Pokemon 4x weak to Poison in Gen 1
Epic in-depth theory. I love that you use logic to explain things!!
I think they had an algorithm that made the shinies, and the developers would sometimes mess with the shinies
7 days on a theory about a video game? That is dedication to your craft right there!
Here's my opinion:
Shinny pokemon were only made to make a higher demand for pokemon by making them rare and have better stats.
Basically making a palette swap and making an artificial demand by making them rare.
I don't think shinies have better stats. Maybe back in Gen 2, but not now
They don't even have better stats. In Gen 2, shininess was determined from IVs, which at the time ranged from 0 to 15. For a Pokemon to be shiny, Speed, Defense, and Special needed to be 10, while Attack needed to be one of several seemingly random values ranging from 2 to 15. HP was calculated from the other IVs, so that was always either 0 or 8. So while shiny Pokemon had kinda good stats, they weren't great. And then come Gen 3, shininess was completely divorced from IVs and no longer has anything to do with stats.
@@DrPumpkinz So that's how people figured out how to hunt shinies in Gen 1
This is still one of my favorite videos that I love to revisit every so often because I enjoy the process behind determining how the rules could’ve worked.
Like, I never even heard of color wheel triads until I watched this all those years ago.
And also it’s the copium needed to understand why they did that to my poor Espeon.
The creator of shiny Espeon could easily just like the color :p
There probably was some kind of formula or algorithm as a basis but as an artist myself, sometimes we pick colors just cause we like them. Not everything has a reason :p
When there are 200+ new pokemon every game, theres no way a large portion of the pallete changes arent automated. These are programmers we're talking about.
This is really cool, think its the first time some went to in depth to how shinies were made, and thank god you did it for only one week, I feel like my brain would rot from just a bit more than a week
Lockstin: says important pok'e mon like cover legendaries always have developer tampering
Shiny Groudon: *Snot color intensifies*
Kyogre looks so nice compared to the booger Dino
Eevee's shiny is silver.
This is a possible reference to silver foxes, as eevee is heavily inspired by cats and foxes
As a programmer, i'm going to say that the shinny colors are just Unnused colors of the pokemon made by designers... but this is waaaaaay funnier.
I disagree. Many of the early shinies were super ugly. I doubt those were intentionally designed.
31:29
Agreed, Espeon needs a new shiny. It's current one makes it look like it's radioactive or absorbed the Master Emerald and became some weird Chaos God Thing.
Its just an alien cat 👽👽👽
Chaos Espeon confirmed!
8:15 wait, this was all just the intro?!?!?!
As an artist with an interest in color theory, I found this incredibly fascinating.
"Or make some pokemon have decent shinies like pikachu" haha as if pikachu's shiny is decent XD
It would be slightly better with Blue Cheeks.
3:27 because that would take up a _lot_ more code in the game files than just putting the algorithm in, if there was one. Duplicating sprites _always_ takes up more code than a pallete swap algorithm.
This video is so interesting ♡ I would love to see your reaction when you see gen 2 shinies through pokemon Stadium 2 cause they change drastically and that "Barnie" Charizard looks pretty good xD (Also just looking at those classic animations-I promise it won't be a waste of your time ♡)
To quote lockstin,
“I
Freaking
Love
This.”
Given how bad some early shinies are I dont doubt they were made by an algorithm. Now to get that flowchart and make it into an actual algorithm to test it out
I for one believe you're correct (always have, because it's that or some dev REALLY hated poor Jolteon). But I can't help but feel like, somewhere in Japan, a dev from Gen 2 is watching this going "Actually, we just used a randomizer to pick colors and ran with that."
I wonder what the Gen 5-8 Shinies would look like with the Algorithm. I'm surprised nobody has don't it yet...
Because their is no public algorithm, at least not a perfect one, if you consider the Flowshard as a algorithm. Pay attention to the Video you watched, please. And also Gen 5 was still made with the algorithm. Gen 6 started to take effort in shiny designs
@@psychopath682 Geez. >:/
Aaaaay the return of the Dragon shirt!!! Would love a rerun on the Dragon and Fairy type shirts, kicking myself for not picking them up when they were available
I think it is canonically mutations
I'd be concerned if it was the other way around
*when lockstin starts talking about complemetary colour and complementary triads*oh i know about that*starts talking about hsb* this know feels like a class
Let's take a moment to appreciate the dedication he put into this video. You just gained a new sub, my dude
9:04 the seal of orichalcos!
This is a surprise to shiny coloring and reasoning there coloring alt
This is the kind of content that makes you one of the best content creators on UA-cam! Always so in depth and interesting to watch!
Some shiny are random and other are hand made.
Found the one that didnt watch the video
This is your best yet most underrated video, I love it