1/23/2021 0 Comments IntroductionIn 2018, there were two momentous events in the Pokemon world. A hacker by the name of Wack0 stole information from Nintendo's servers and anonymously leaked data from early Pokemon prototypes they had found. They first leaked the build of Pokémon Gold and Silver that was showcased in 1997 at that year’s Nintendo Spaceworld, a convention where Nintendo showed off upcoming games. Afterwards, they leaked some early information from the original Red and Green versions of Pokemon, giving us an unprecedented look at the early production of those games. The Spaceworld 1997 demo showed us a version of Gold and Silver very different from the released version, with over 40 completely new or heavily altered Pokemon from the final while the leak of the internal files of Red and Green gave us early data, maps and sprites, and most interestingly, back sprites for 39 rejected and unseen Pokemon. Pokemon researchers calling themselves Team Spaceworld and Helix Chamber did an amazing job analyzing these early leaks, as well as doing some deep analysis of the internal index order of Red and Green. They came to the conclusion that the internal ordering of the Pokemon in those games was, more or less, the order in which those Pokemon were created and implemented into the game. By ordering them like this and adding in what we know about some of the missing and rejected Pokemon, the internal order became an archaeological dig that showed us how concepts like evolution or types came into being. It also showed the games as they gradually developed from a Kaiju-centered creatures to something much broader and more interesting, integrating elements of folklore and legends, or even just real life animals with an elemental twist. Helix Chamber’s work on this is brilliant, and honestly the best thing I’ve ever read on video games in general. I would highly suggest reading all their articles, but especially checking out all of their articles on the internal index order, especially if you are interested in Pokemon history. Which, of course, you would be if you’re reading this. Ever since I read their analysis, I’ve wanted to do the same sort of archaeological dig on Pokemon Gold and Silver, especially since we now have so many enticing details on its development from the Spaceworld '97 demo. However, Gold and Silver have a problem. While the original games had an internal index order different from the Pokedex order, Gold and Silver’s was rearranged to match the Pokedex order, completing obscuring the one good tool we had to determine the creation order of the Pokemon. While Red and Green had a list in the code telling us the order of the creation of the Pokemon--and thus giving us a tool to determine how the design changed over time--Gold and Silver had had that tool scrubbed from the internal data. As a result, it was impossible to do the same deep dig through Gold and Silver without a huge amount of speculation and guesswork. Fortunately, in April 2020, more Pokemon leaks occurred, giving us even more snapshots into the creation of Gold and Silver. There, we got four main new pieces of data that can help us reconstruct the development and creation of Generation II Pokemon. Indeed, we got a copy of the Spaceworld 1999 demo, which is dated to three months prior to the original Japanese release, and features a Pokémon roster much closer to the final one, with some notable exceptions. We also sprite archives from June of 1999, which showed that while a lot of the ’97 Pokemon still existed at that point, many were in the process of being overwritten with newer designs. We also got copies of the designer’s scratchpads, which were spaces where the designers drew the final Pokemon sprites. What is interesting about this is that, for reference, these often included earlier designs and sometimes alternate sprites, even sometimes featuring seemingly unrelated sprites of Spaceworld ‘97 Pokemon. Maybe most interestingly, we also received a large archive of Pokémon sprites, which was found in the Korean section of the leaked Pokémon data. This back-up archive is dated to May 6th, 1998. These sprites are as bizarre as they are fascinating. While the original 151 Kanto Pokémon are mostly the same as in the ‘97 demo, aside from a handful of updated sprites, the index also had a completely independent section of sprites that followed these. These new sprites are extraordinary. There are 148 front and back sprites depicting not only Spaceworld ‘97 Pokémon, but previously completely unknown Pokémon designs. In fact, there are a ton of new designs found here, many of which are strange and barely look like Pokémon. To my mind, this Korean ’98 list is the most important document we have in reconstructing the development of Gold and Silver, because here we have the lost internal index order for Gold and Silver. Using this and comparing Pokemon that appear in similar places on the list, we should be able to reconstruct an idea of what they were trying to do every step of the development process in Gold and Silver. Why is the Korean Index List So Important? Despite seemingly being organized in what appears to be random order, this sprite archive gives us a precious glimpse into the creation of a whole new generation of Pokémon. Indeed, while information such as names, typing, or evolution data was not included in the archive, the mere presence of the designs can tell a lot about the changes that the roster underwent. Some of these drafted Pokémon are potentially inspired by old designs, were jokes, or simple experiments, and by looking at this archive we can discern a few things. First of all, it gives us a sense of where their minds were as they were developing these games. Some of the early Pokémon seem much more like Generation I designs than anything in the final list, and we can see the ways these early designs gradually evolved into the final ones we ended up receiving. Here in the Cryptodex, I make a further hypothesis. I believe that the Korean Index gives us insight into the order Pokemon were designed and input into Gold and Silver. I believe the Korean Index is a brainstorming sheet that they added Pokemon too as they designed them or decided to use them in Gold and Silver. I don’t think it’s a perfect guide to development order: many of the Pokemon designs found in it probably overwrote earlier designs for the same Pokemon, so even if the initial Pokemon was designed in this order, the designs are probably haphazard and show multiple eras of development. In addition, since it was a brainstorming sheet, Pokemon were probably added and subtracted from it off and on, so a lot of the Pokemon designs could be out of order and unchronological. Given those caveats, there are a lot of clues that the early Pokemon on the list were the first developed, and the last in the list were the final Pokemon added before Spaceworld ’97. They were probably added, deleted, and shifted into the game as they decided on the 101 extra Pokemon that would be in the Generation II roster, so some of these, though they do not appear in Spaceworld ’97, might have still be considered for the final list shortly before or after. But I hope to show if we keep in mind that this is loosely chronological, we can further understand how it was developed. What I’m aiming for with this Project In some ways, this is a continuation of Helixchamber’s analysis of the R/G Internal Index List. I’ve made a poster that summarizes their findings, which you can find here: However, the development of Gold and Silver didn’t work in quite the same way, so the project is going to be slightly different. I’m going to go through the 1998 list, one by one, and analyze each of the Pokemon found therein. Once I’m through those, I’ll analyze the Pokemon added in the ’99 builds and then those only found in the final. I’ll try to guess and which designer made it, why it may have been deleted, what we know about it, and then chart how it developed over time, whether it eventually became a final design or was deleted along the way. In doing so, I’ll be dividing the Pokemon into development eras as well and try to group Pokemon around similar design philosophies, to break down the list into coherent pieces as much as possible.
Though we have a number of snapshots of the development of Gold and Silver, there are substantial gaps in the builds we have. As a result, there are some stages in its development we have to reconstruct. Obviously, there aren’t clear cut-offs like I’m designating, and each era blended slowly into the others. However, we can still try to delineate these eras to show when ideas about the games changed, and when the rosters were substantially redone. In general, here is a list of the “design eras” of the game, as far as I see them: 1) Holdovers from Red and Green: There are multiple Pokémon on the 1998 back-up list that were clearly designed during the Red/Green era and reconsidered for Gold and Silver. We know, for instance, that a few of the baby Pokémon and Kotora/Raitora were originally from Red and Green, and I have suspicions about some others. Jagg and Ikari, when we get to them, might be the best example of this. Furthermore, many of the old Red and Green designs were probably redesigned by the time we see them in the 1998 list; either they had gone through more iterations since they were put into the Gold and Silver master list, or they were redrawn when they were put into Gold and Silver. So while I’ll point out when it’s likely these were based on Gen I Pokémon ideas, actually pinning it all down might be frustrating. 2) Early Newcomers: There are a few Pokémon which predate the Spaceworld 1997 build, and by using some promotional material and some clues here and there it’s possible to determine what that handful of newly introduced Pokémon was, be it an actual design or just an idea which was later reworked into a later design. There was probably an entire build of "Pocket Monsters 2" around 1996 with these designs present in it, but any idea of what that looked like will have to be reconstructed from what we know from the early parts of the Korean Index and early development materials. It is probably the case that all the Pokemon in the 1998 list before at least Ho-Oh were from this period, but for reasons mentioned above they could have been redesigned or have been input out of order, so this is far from definitive. Furthermore, the move list in Spaceworld 97 has some odd moves that fit none of the Pokemon that exist in that build, suggesting that earlier Pokemon who used these moves were cut prior to it. I’ll try and point to designs that seem to be from this era as we get to them. 3) Spaceworld 1997 (September 15 1997): The earliest build we have a copy of. This build looks substantially different from the final games, and has a lot more resemblance to Generation I than the final games, especially in terms of art style and Pokemon designs. It was also substantially different regarding other aspects, such as with the map of the region resembling Japan as a whole, and Giovanni of Team Rocket infamy set to make a return. What’s most surprising to me is how substantially “complete” the roster is, as there are no placeholder sprites of any sort. As well, almost all the pokemon—with the exception of Girafarig and maybe a few others—have complete movesets. The stats for these Pokémon are all placeholders, as are their cries, but for the most part, it feels as though development was winding down and the game could be released by the next year, as it had been advertised. However, for whatever reason it wasn’t. The game seems to have been put on hiatus, and when work resumed on Gold and Silver by 1999, it seems to have started a new development cycle and gotten rid of a lot of the Pokémon designs that had dominated 1997. 4) May 1998 Back-up (May 6th 1998); ie "The Korean Index": A rather puzzling set of sprites. The Kanto Pokémon mostly reuse their Spaceworld 1997 sprites, though a few were updated… Only to seemingly never be seen again. For example, Weepinbell, Dodrio, Goldeen, and Seaking all got updated back sprites, though the June 1999 archive reuses the 1997 ones. As for the Johto sprites, they are comprised of a mix of scrapped design and ones which appeared in the Spaceworld 1997 demo. It’s unknown if those scrapped designs were made before or after the 1997 demo, though I believe we have reason to think that most of them were prior to Spaceworld '97. The only thing that’s for sure is that by the time this Index was made, Stantler had been added, alongside the final design of Elekid. As well, Crobat’s sprite was tweaked, and Bomushikaa received brand new sprites. Maybe the importance of this list is how much didn’t change. Since so many things are fundamentally similar, it shows that either the design team had moved to other aspects of the game (maps, stats, encounter tables, story progression, etc), that they had paused development during this period, or that they had started work in a different file system, and this was the backup of the old development that had created Spaceworld 97. Whatever the case, this archive is quite the fascinating list of what could have been. 5) June 13th, 1999 Build: There are three reasons why development may have been paused in 1997. A month after Spaceworld 1997, the infamous Porygon episode of the Pokemon anime aired in Japan, giving hundreds of kids seizures. In the wake of that incident, Pokemon’s future may have been unclear, and they may have stepped back from development to think about the future of the franchise and if they even should release a sequel. Secondly, it looks as though development was moved from the original Gameboy to the Gameboy Color sometime during this period, and they may have had to overhaul what they had and remake it from scratch. Thirdly, the Pokemon anime debuted in America during this time, and became an overnight hit. The design team may have decided that to appeal to an international market, they needed to rethink their designs to be less Japan-centric. In particular, the 1997 world map resembling the regions of Japan was scrapped during this period. All of these reasons could be at least partially true, but by the time of our next snapshot of the design, the games are now well into their final phase as Game Boy Color titles. Many of the Pokémon from Spaceworld ‘97 are gone, and the new designs from the sprite backup from 1998 are nowhere to be seen, implying that they were likely never implemented. Many of the Spaceworld sprites were still kept in the game, but were now clearly used as placeholders for new Pokémon whose sprites hadn’t been created yet. A few sprites were even tweaked to match the games’ finalized artstyle. The game seems very much in transition at this point, but in transition to a new style distinct from the previous games. 6) Spaceworld 1999 Build (August 17th 1999): Substantially more complete than the 1997 demo. All Pokémon from the final games are there, but not all of them have had their design finalized. There is also evidence that between these two versions a few very briefly considered Pokémon were designed and then discarded, such as a Girafarig pre-evolution and a Piloswine evolution. What this says to me is how quickly Pokémon were being iterated in this era and how much they were still not sure about the final lineup. 7) Final Build (November 21st, 1999): Only a few months after Spaceworld 1999, the final build was completed, and yet, it still has a substantially different lineup from the August Build. Some Pokemon seem added in at the last second, while some of the last 1997 Pokemon who made it this far are finally culled from the lineup. I’ll point out significant changes along the way as we get closer to this lineup, but these designs will mostly be for comparison. As you can see, Pokemon Gold and Silver went through a convoluted development cycle, almost as convoluted as the original games. Still, it’s a fascinating cycle, and one that can teach us a lot about the thought processes that lead to the final game. So I hope you enjoy the journey!
0 Comments
|
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |