Era IV OverviewAnd now we’ve made it. To the final section of Pokemon Gold and Silver’s development. Pokemon Gold was originally slated to be released in 1998; however, that was not to be. On the one hand, the infamous Porygon-seizure-incident put Pokemon in the spotlight in the worst way; it seemed, for awhile, as if the outrage over that event may have led to the death of the franchise before it started. But there were other issues as well. The success of the first game in the series had led to demand for stop-gap games to keep the public interest in the franchise high. Developers were tasked to work on Pokemon Yellow, for instance, and the development team quickly repurposed most of the sprites they had made for Gold and Silver into special-edition sprites for the release of Pokemon Blue, a Japanese exclusive deluxe release of the first games. These projects likely siphoned away many of the developers from Gold and slowed progress for those games. However, the biggest reason Gold was delayed was probably due to internal issues. In interviews, the development team has said that the first approach to this game—showcased in the SW97—had major problems that the team was finding difficult to address. The main issue the designers pointed to was the map: at the time, Spaceworld ’97 was designed to encompass the entirety of Japan, with the original setting of the first games, Kanto, relegated to one small town. This map, apparently, didn’t work; it was trying to do too much. Due to its design, and possibly to cartridge limitations, there were too many cities that didn’t contribute, and the cities were so close together that there wasn’t room for interesting routes in between. While the team doesn’t mention anything about problems with the Pokedex roster, I find it likely that they were also having trouble with it. My number one and two guesses are that there were too many single stage new ‘mons, and that the babies didn’t add enough interesting variety. Whether or not I'm right, something wasn’t working. And so, at some point, Satoshi Tajiri, the creator of Pokemon and the director of Generation I, stepped down as director of Generation II. There was some sort of gap (probably most of 1998) in the development of Gold Silver around this time; it’s unclear if that gap was while the team decided on a new director, or if the gap was motivated by letting the team work on Yellow, Blue, and a few other projects. But sometimes in late 1998 or early 1999, =Junichi Masuda took over as director. Masuda's a big name in Pokemon: he has since worked on many of the later games in this role. Masuda seems to have immediately made massive changes to the games, to the point that Era IV—which starts in 1999 and covers the entirety of Masuda’s tenure as director—is almost a completely new game that happened to draw from some existing assets from Tajiri’s time as director. First, Masuda completely redesigned the map. The exact changes to the map are beyond the scope of the Cryptodex, but suffice to say, the plotline and locations of the game after the reboot are completely new. The Pokedex roster, on the other, is harder to sum up in terms of how it evolved, and one of the ongoing questions we’ll be looking at in Era IV is what Masuda’s overarching philosophy in regards to the roster was. At this time, I can’t see very many clear patterns. I think Masuda was less interested in featuring Pokemon that linked to the original generations that Tajiri was, explaining the culling of the babies and of the numerous third-stage Pokemon that were dropped after Spaceworld ’97. I think, generally, Masuda was interested in creating cute Pokemon, and I think, in general, his goal was to make the games more friendly to international audiences by toning down some of the more oblique references to Japanese culture in the Spaceworld ’97 roster. I do think that the massive changes to the roster, at a point in development where Masuda had less than a year to deliver a completed product, were probably an overambitious mistake. We’ve already seen how so many of the newly introduced Gold and Silver Pokemon are underused or almost completely absent from the final games, and that becomes even more pronounced in Era IV. It took Generation I six years to build up its roster and properly balance all of them against each other, and even then there are a lot of Gen I Pokemon that are bad or absent from those games. And unfortunately, a lot of the elements added or significantly changed during Masuda's time in charge are the most rushed parts of the final games. If Masuda decided to throw out a good portion of the roster that had already been worked on, then that meant he was tasked with creating around forty new Pokemon in less than a year, in addition to balancing the old ones that he’d inherited. My speculation is that this is why so many Johto Pokemon feel badly balanced—better to keep Yanma and Ledian weak and unusable rather than accidentally break the game by making their stats too high. It's probably also why so many of the new additions to the roster (especially things like Slugma, Larvitar, and Misdreavus that were only created after the hiatus) are so hard to find or locked resigned to the post-game sections of the story. The less these new Pokemon impacted the difficulty curve of the main game, the better, because the team could tweak and test them all the way to the very last moment. If Morty never bothered with a Misdreavus, there's no chance that giving it 20 more Speed might accidentally make the fourth gym unbeatable. There’s also an important technological change in this era which affects our ability to analyze it. The Korean Index was last updated on May 6th, 1998. We know, however, that the roster wasn’t worked on much after Spaceworld ’97, because almost everything in the Korean Index matches the SW’97 demo. There were probably a few changes to the Korean Index between November 1997 (the SW’97 demo) and May 6th: Bombushikaa’s and Ho-oh’s sprite were updated, Stantler’s was probably updated or overwrote something, Elekid’s was at the very least touched up, and Cyndaquil could have overwrote something already there. But that’s it. So why did the designers stop using the Korean Index? It appears that the team found a way to more directly program in the sprites into the game. As far as we can tell, up until May 1998, the team’s process seems to have been to draw the sprites on a scratchpad, decide on the design they liked, and then manually update an Index number in the Korean Index with the sprite they wanted to use. When the games built from the source code, the instructions were to pull these sprites directly from the Korean Index into the data of the game. After May 1998, however, they seemed to have found a more efficient way to update the sprites. It looks like, instead of manually dragging their sprites to the Korean Index, they told the game to build directly from the scratchpads that they were using to work on the sprites: the game’s programming told it to take whatever was in the top left of the scratchpad, allowing them to leave drafts and experiments in the space to the right and below the sprite. As far as I can tell, this would have been much easier for the team, because every time they updated a sprite on the scratchpads, it would automatically update itself in the game, so long at it was built again from the source code. Which means that the team no longer needed the intermediate step of the Korean Index. This leaves us with several disadvantages. First, we no longer have a hint at the creation order of any of the Pokemon in Era IV. We can reconstruct a vague order in which these Pokemon were created by checking when they first got stats, when they first got sprites, etc, but it’s a bit of a tragedy, because this only gets an approximate ordering. That means that any patterns in the different Periods of development in this era are obscured, because we don’t know which Pokemon were created in batches. I’ll do my best to explain why Pokemon are where they are in my listing, but know that in Era IV, the ordering inside each Period is more of an educated guess. The other downside to losing the Korean Index is that we no longer have any sprites for Pokemon which were created but then abandoned before June 1999. The Korean Index was full of oddball sprites that were never present in any of the builds of the games we have, but we just don't have the same insight into Era IV. We know from the stats of certain Pokemon and by traces left in the data of a couple of Pokemon that were considered but ultimately rejected; however, because we don’t have the Korean Index, we usually don’t have sprites for these guys (with a few exceptions). So we’ll be talking about a few Pokemon in this era that are more or less mysteries, without even a fun visual to get our imagination going. One other technical change is that, after Masuda took over, the development shifted to making Gold and Silver not for the Super Gameboy, but for the new Gameboy Color. As a result, the sprites designed for this era of development are designed with a richer palette in mind and completely different shading. On the upside, these sprites are a lot more expressive, and give the games a more distinctive look! They also show us the exact moment when a Spaceworld ’97 Pokemon was updated (because the new sprites would now be shaded for the Gameboy Color, not the Super Gameboy), and for how long it just stuck in the roster as dead weight before the team found a replacement for it. On the downside, however, these new sprites are much less distinctive in how they’re shaded, and because of that, it’s much harder to determine which sprite artist drew them in the first place. We have a lot of other ways to determine the author of a ‘mon, but guessing at the authorship of a sprite is much less likely to work in this stage of development. Alright, that was a lot. With that in mind, let me outline the ways Era IV will work a little differently from the previous three eras on the Cryptodex:
Period 4a: Development Restarted Like I said above, Period 4a is going to cover the first year of Masuda’s time as development lead, from approximately May 1998 to just before the collection we have dating to June 1999. That isn’t to say that development was active for this entire year; I expect that the team was still on hiatus for most of 1998. It also seems like there was a period of development post-Korean Index (so after May 1998) where the team was still iterating the SW'97 designs: the scratchpad files we have have examples of Kotora, Hoppip, Akueria, and Borubea that all seem updated from their SW97 designs. So there might have been a chunk of 1998 that was still under Tajiri, before Masuda took over as director. At some point though, there seems to be a sharp break. My guess is that Masuda became director late 1998 or January 1999 and started to drastically reshape the game. Whenever that hand off actually was, by the time of our first snapshot of development—April 3rd, 1998, which gives us a look at the stats of each Pokemon—the game had already been substantially changed. By April 1999, there have been a bunch of changes to the line up, mostly replacing the baby Pokemon from Tajiri’s time in the lead. In total, 16 Pokemon had already been completely deleted from the roster by April 1999: Almost all the babies had been deleted. Mikon, Monja, Gyopin (though Furret shares its palette after June 1999, suggesting it may have still existed until recently), Para, Koonya, Puchikoon, Betobebii, Pudi, and Baririina were all gone. In addition, Taaban, Bombushikaa, Shibirefugu, Jaranra, Riifi, Animon, Manbo1, and Ikari left no traces in the updated roster. In addition to these 16, there were eight more Pokemon from earlier stages of development whose's names, data, and types had been deleted, but whose sprites still remained in some form. It had probably been decided to deleted these Pokemon by April 3rd, but that decision may have been made later than the Pokemon above. Alternatively, some of these may have been deleted at the same time as those above, but their sprites survived purely because the designers hadn't yet figured out what they wanted to overwrite these sprites with. For instance, Gurotesu exists as a placeholder sprite, and its shading was touched up, suggesting that it survived at least for a small amount of time after the reboot. Likewise, Berunrun's sprite are used in as placeholders in a slot which eventually became Dunsparce's. Rinrin's backsprite was used as a placeholder for Larvitar's backsprite, and Nameru's back was used as a placeholder backsprite for Tyranitar. Likewise, Wolfman's sprite was being used as a placeholder for what became Larvitar, and its evolution, Warwolf, seems to have been used as a placeholder for a third Piloswine evolution that was later cut. Hinaazu had already lost its name, but existed at least as a palette and backsprite for Azumarill. There’s not a lot of clear patterns to the Pokemon added during this period from what I can see initially (and I’m very interested in whether readers see one that I’m missing!). One thing we do see is that it seems that Masuda’s period of development started by first culling all the Pokemon from Spaceworld ’97 that they didn’t want, and then allocating that newly freed space in the roster to new two-stage or three stage lines. So, for instance, the three slots that became the Piloswine line (one was later cut) were probably allocated as “a three stage ice-type Pokemon” long before Piloswine had been designed; it was probably created to fill those slots, not the other way around. Likewise, most of the other new Pokemon seem to have been designed to fit gaps created by the earlier culling. Very likely, the team didn't create more Pokemon in this era and then delete Pokemon to find room for them; instead, they deleted first, and created new ones after. From the patterns we see, it seems as though this approach to creating new Pokemon during this period was quite different that the time during Tajiri’s directorship. During the earlier stages of design, the Pokemon developers seem to just brainstorm a ton of new designs and fit them into the new line-up; they would then create evolutionary relatives for the most popular of these designs. However, here it seems more like the team was filling holes: parts of the roster already seemed earmarked for a two-stage line before the first Pokemon was even created. While, obviously, I’m sure that was done to some extent in the previous parts of development we’ve discussed—especially Era III—it appears that the default beginning point in the creation of Pokemon in 1999 was to start with a hole to fill, and to design an appropriate Pokemon to fill that hole. In the case of Period 4a, it’s clear that one of Masuda’s goals was make Generation II feel more distinct by adding two and three-stage lines wholly unconnected to Generation I. Given that most of the Pokemon deleted to make room in the roster were evolutions of Gen I Pokemon, it seems like an intentional choice to replace them with Swinub, Yanma, Teddiursa, Shuckle, and the like. The team did delete existing unique lines from earlier in development—for instance, getting rid of the Manbo1/Ikari/Gorutesu line and the Wolfman/Warwolf line—those were probably already lines that had serious problems with them and needed to be deleted or reworked (The Ikari line, for instance, was stitched together from three unrelated Pokemon, and the Warwolf line was already created in competition with the Houndour line in the first place). But regardless, the intent here seems to be to create a more diverse set of stars for the new roster. It's easy to extrapolate this too far, but this could also suggest that this stage of development was very focused on creating Pokemon to fill a gameplay purpose. If Masuda started by allocating certain slots in the roster as “ice-type three stage line” or “two-stage early game Pokemon” than the designers might have been designing to a role. Of the Pokemon in Era IV, those in 4a are the ones most likely to appear commonly in the final game, which means there was not only probably time to find them a place in Johto, but also that they were probably made with a place to fit in mind. ID #X01: Heracross (Before June 1999 sprite coloration is hypothetical, and was colorized for me by OrangeFrench!) It’s very likely that Heracross was the first brand new Pokemon to be created in the new era of Pokemon Gold and Silver's development. First of all, Heracross is one of only two Pokemon from Era IV for which we have a pre-June 1999 sprite; we’ll talk about it in detail below, but the tone of the sprite is so different from the rest of Era IV I almost wonder if it was created before Masuda even became the director. In addition to this very early sprite, Heracross already had stats, a sprite, and typing by April 1999; though the sprite needed polish and its typing changed before the final, it was a Pokemon that was pretty much finished by the time of our first timestamp (April 1999) of the rebooted development. Thus, I’m pretty confident putting it at the beginning of Period 4a. Tajiri’s stated in interviews that he was fascinated by bug catching; the beetles he collected while hunting through the forests as a kid inspired the idea of hunting for Pokemon. While we don't exactly know his relationship to the practice, it's also very popular in Japan to watch beetles fight against each other, and it's likely the sport of "beetle fighting" inspired Pokemon battles. Pinsir was likely created out of the inspiration of a fighting stag beetle, while Heracross was designed after a Rhinoceros beetle, which could also be used to fight. Fighting beetles perfectly play tribute to the initial inspirations for Pokemon. But it’s a bit weird to do it twice. Why do we need both Heracross and Pinsir? Back in 2018, when the only data we had was of Spaceworld ’97, a leading theory was that Plux, the Pinsir evolution found in Spaceworld ’97, was an early design for Heracross. According to this theory, Heracross was later separated from Pinsir to become its own Pokemon and given the Fighting-type to differentiate it. And to this day, there are a lot of people who still buy into some form of that theory. I mean, look at it: both Pokemon are designed around fighting beetles: the stag beetle in Pinsir’s case and the Rhinoceros beetle in Heracross’s. Both of them have similar moveset, both learning fighting moves, and both even have a similar pose. The two fit the same niche very well, raising the question of why they both exist at all. Okay, one looks like a Jason-type serial killer and the other one's your buddy, but other than that, same niche, certainly. Unfortunately, once the Gigaleak of 2020 provided us with more data on Gold and Silver’s development, we now know that there never was an evolutionary link between Pinsir and Heracross. In fact, for at least five months, Heracross and Plux coexisted with each other in the same roster. What seems to be much more likely is that Heracross was created to be part of a bug trio alongside Plux and Scizor. Originally, none of them had a second typing—Plux and Scizor became Bug/Steel later in development, and Heracross hadn’t yet gained its fighting type—and they appeared one after another in the early Pokedex: Scizor’s 212, Plux was 213, and Heracross was 214. Together, all of them formed something of a "fearsome bug trio": Furthermore, though their stats would be tweaked--and in Scizor's case almost completely redistributed--as of April 1999, their stats were all remarkably similar: Let's quickly get the most obvious thing out of the way: all three are absolute beasts and could easily dominate any Pokemon of their levels. But also note the similarities: they all have the same stat total, at 540, and they each have high attack: 130, 130, and 125. Plux and Scizor have one more stat they excel in—Plux has defense, Scizor has Speed—while Heracross has two stats its also good in—HP and DEF—and trades that with numbers radically lower in the rest of its stats. Of the three, Plux and Heracross’s stats the closest, but Heracross is much more one-sided towards physical stats, while Plux was a bit more of an all-rounder. I guess, conceivably, you could argue that that Heracross’s stats were just updated versions of Plux’s, and an attempt to make a new beetle that filled its role, but I feel there’s enough variation to demonstrate that the Pokemon team were trying to make all three feel similar enough to be a trio but different enough to feel distinct. I think the team liked the idea of making the Bug type more formidable in Generation II, and the reboot had already inherited Scizor and Plux, which were already contenders. I’m not sure exactly why the idea was to make a trio of terrifying bugs when Scizor and Plux already existed; Heracross feels very unnecessary as an addition, and I’m at a loss for why the team designed it. But again, like I’ve noted before, they continued to work on Plux all the way until a few months from the final. It even got a Pokedex entry. The evidence heavily suggests that Heracross was not there to replace it in any fashion. Maybe someone on the team liked to idea of two different beetles fighting it out? Of course, eventually someone must have realized that Heracross and Plux were just too similar. There were a number of attempts to differentiate them: the team tweaked their stats to be more different, then they gave Plux the Steel type and Heracross the Fighting type. But it seemingly was not enough to justify both, and Plux was eventually discarded. Another problem undoubtedly had to do with Scizor. Scizor was originally differentiated from the other two by being much faster than them, but the addition of the Steel type put a wrench in that plan. The team had decided that Steel-type Pokemon were supposed to be slow, which means that Scizor’s stats were modified such that it was even slower than Scyther. This meant that not only were Plux and Heracross too similar, but the new rules about Steel-types made Scizor's stats even closer to the other two. By Spaceworld ’99, all three of the bugs were very, very similar, and something needed to give. Eventually, the team decided to replace Plux with Forretress. This accomplished a few goals all at once. First, it replaced a Pokemon related to a Generation I ‘mon with one related to a new Generation II ‘mon, Pineco. Not only did this make Pineco a bit more useful and interesting (though not really enough, which we’ll discuss when we get there), but it also served the goal of giving Generation II an identity less tied to the first games. Furthermore, Plux’s design is too aggressive to drastically change its stats to fit into a new defensive niche. By switching to Forretress, the team could preserve the Bug/Steel typing while also creating a more defensively minded bug. Cute! They even let Forretress keep Plux's palette! The biggest discontinuity between Heracross and the other two is that Plux and Scizor were of course second-evolutions, while Heracross didn’t evolve from anything. Was that always the case, or did the team ever envision Heracross as a dual-stage Pokemon? It’s possible, given how early Heracross was created, and how spotty our information on late 1998/early 1999 is, that the team may have briefly set aside a slot for a first stage Heracross. As we’ll see throughout the rest of Era IV, that seems to have been the default during this period: two or three slots would be set aside for a yet-created family, and then later some of those slots would go to other Pokemon if the designs didn’t need that extra member of the evolutionary family. Shuckle, Piloswine, and Yanma, for instance, were all designed with an extra slot available for an additional evolution, which was later repurposed into an unrelated Pokemon. It’s very possible Heracross was in the same boat, and its extra family member was given a slot for a month or so before being cut. If this did happen, mind you, the extra evolution probably didn’t last for very long; I doubt it even had a sprite or stats before being cut. There’s no real evidence for this except for Heracross’s relationship to Scizor and Plux. Furthermore, the Pokedex slots on either side of Heracross held Sneasel and Plux as far as our data goes back, so if there was a Heracross pre-evolution at one point or another, it would have had to fill a slot far away from Heracross, which is not something that typically happened when the team was designing Pokemon in Era IV. Still, it’s at least worth considering, even if the idea didn’t get very far. Plux, Scizor, and Heracross all first got their secondary typings on July 30th, 1999, suggesting that these secondary typings—in addition to being a way to make Plux and Scizor feel like more interesting upgrades for Pinsir and Scyther—were also done at the same time to help Heracross fit a niche unique from these other two. Heracross doesn’t actually learn many Fighting type moves, even from TM; it learns Counter, Reversal, and Detect (from TM), all situational attacks that depend upon what the state of the battlefield looks at the time. Making Heracross only able to use fighting moves to counter its opponents was probably partially a way to give Heracross a bit of identity: it feels very fitting for a Rhinoceros beetle, which can use its horn to lift its opponents in the air, would only use its fighting prowess to use its opponent's strength against them. However, the lack of Fighting moves was also probably a result of how late Heracross’s typing changed; the team likely didn’t want to mess with the moveset too much lest they unbalanced Heracross at the last second. Or they simply didn’t have enough time to mess with the moveset too much, so they just threw a couple fighting moves on him and called it a day. The most interesting move Heracross learns, though, is not a fighting move but a new Bug-type move: Megahorn. Megahorn’s a great attack: it’s a base 120 power Bug-type move that only Heracross learns. Signature moves are always cool, and I’m always glad to see them because they add a lot of flavor and identity. Even though only Heracross learns it, it’s also sorely needed in Generation II, especially because of how little diversity there was in Bug-type moves in Generation I. What’s interesting about the move is that it didn’t exist in the Spaceworld ’97 iteration of the game. Instead, Megahorn seems to have overwritten Megaphone, a signature move for Hoothoot that lowered Special Attack. Megaphone wasn’t nearly as flashy as Megahorn, and it isn’t a huge loss. However, my favorite thing is that Megahorn was probably created because of a misunderstanding about Megaphone. Both of them are spelled the exact same way in Japanese: there's not a sound for "fo" and so the the sound is often spelled as "ho" in Japanese words. Thus, Megahorn and Megaphone were spelled the same way! As a result, it’s likely that playtesters misread Megaphone as Megahorn while they were testing, mistakedly assuming that somehow Noctowl had a huge horn on its head! The move was probably changed to fix that funny misunderstanding. Who knows, maybe Heracross was created entirely so that the team would have a horned Pokemon that could make use of the new move! The last interesting thing about Heracross is its early sprite. This sprite dates from before June 1999, making it one of the few sprites we have from between May 1998 and June 1999, a period that is otherwise quite a mystery. We know that it’s an earlier sprite, by the way, and not a later experiment, because it appears that the backsprite for the June 1999 Heracross was based upon this early design: check out the arms on the back sprite and how they match the earlier pose: If we look at the shading of the sprite, it’s clear it was made for the GameBoy Color, dating it to after the rebooted development of Gold and Silver. Given that, there’s a lot that’s strange about this sprite. Final Heracross’s face is fun-loving, kind of goofy, but most importantly, anthropomorphic. Pretty much all of the Generation II Pokemon that made it to the final have expressive faces with big eyes or humanoid facial features (even the exceptions, like Piloswine, still look like they could use their body to be expressive); in general, I think that Masuda preferred that Pokemon had a way to emote or that looked cute. This could be another reason Plux didn’t make it to the final: Forretress’s eyes are big and expressive, while Plux has a creepy looking blank stare. But early Heracross, on the other hand, has a strange, bug-like face, much more akin to Pinsir’s face from Generation I. It’s a lot less cute and appealing than the final Heracross, and it’s pretty obvious why the face was amended. But it’s also a sign of the big differences between the design aesthetics of early Gold and Silver, and the aesthetics of the final game. In the Korean Index, there are a lot of discarded designs that have strange, unexpressive faces and would look right at home next to this early Heracross. For instance, the early Scorpion sprite, the two unused flies, the strange seed plant, and even Hanamogura all have strange faces that show a lot more diversity than the final game. The differences between early and late Heracross shows just how much design sensibilities had changed: we’ve gone from Pokemon being able to embody any type of strange monster, to the idea that Pokemon have to be, to some extent, relatable and able to express emotions with their faces. Heracross was a great addition! It has personality, it has appeal, and it shows that bugs are no pushovers this time around! I’m not sure I like trading it for Plux; I liked the idea of Pinsir evolving into something greater. Admittedly, though, Heracross fills a better niche. I’m glad we got something like it in the final, even if Pinsir was left high and dry. ID X02: Totodile (Colorization of early Totodile sprite is also done by @OrangeFrench!) Moving on from Heracross, it's time for Totodile! Probably one of the most beloved starters in any of the Pokemon games, Totodile has a ton of fans, in part because of how good it is. However, as we know, Totodile (probably a portmanteau of "Tot" and Crocodile) wasn't in the first iteration of Pokemon Gold. Even so, all of our evidence suggests it was one of Masuda's main priorities. Even if there's no sign of the Totodile line before the hiatus, by April 1st, 1999, all three of the Totodile family had stats, though these stats were barely more than placeholders, stolen directly from the Squirtle line. They also had name data, and by June 1999, each of them had both a front and backsprite; these front and backsprites were well fleshed out and used all the way to the final, suggesting that a significant amount of work had already been put into them. Totodile, like Heracross, even has a very early sprite that probably pre-dates the June 1999 sprite bank we have. This sprite (seen above) notably has a lot less personality, a brown nose, and has a different design for its stomach, more akin to the design of Feraligatr. While this sprite could have just been an experiment while the team was concepting what Feraligatr would look like (it was found on Feraligatr's Scratchpads), I think it's more likely that it was a first draft of Totodile, right at the beginning of the rebooted development. From all of this, it’s very likely that one of the first things that changed after the reboot was a redesign of the original water starter line, Kurusu, Akua, and Akueria. Though their stats (and presumably movesets, though I don’t have access to that data) would need a bit of tweaking, Totodile and its family were already substantially finished by June 1999. I’m on the record as being a huge Kurusu fan. It’s super cute, I love the Pleisosaur design that it has, and I think the evolutions do a great job building on the design that’s already there. Saying that, I get why Masuda might have first turned towards the Kurusu line as something that needed fixing. The line is very lacking in design hooks. If you look at Bulbasaur, the bulb on its back not only is a cool and memorable part of its design, but suggests how the Pokemon will change as it evolves. Charmander, likewise, has the flame coming from its tail, a distinctive trait. On top of that, Charmander becomes an iconic dragon; Squirtle is clearly a turtle, and Bulbasaur, while a little more vague, looks enough like a four legged dinosaur that it’s easy to grok what its supposed to act like. The Kurusu line doesn’t really have any of those elements in their designs. Kurusu doesn’t really hint at what its evolutions are going to look like, beyond bigger sea-dragons. It doesn’t have any distinctive design hooks in its design; it’s mostly smooth and light colored. And it isn’t even that clear what animal it’s supposed to be inspired by: is Kurusu a seal? As cute as it is, it doesn’t do the job a starter is supposed to do, and I think the design team recognized that. Totodile, on the other hand, succeeds at all of these design requirements. While I personally think Totodile is much less cute that Kurusu, it immediately pops out to the player. Its face is extremely expressive, telling you that Totodile is excited and fun-loving. It’s teeth and snout immediately let people know that it’s an alligator (or crocodile) and there are all sorts of aspects to its design that could potentially grow as it evolves. Will it get a bigger, fiercer jaw? Will the spikes on its back become more pronounced? Spoilers: all of the above. Another way that Totodile’s design feels fresher than Kurusu’s is that Totodile’s sprite is filled with sharp angles: the pointiness of its snout, its teeth, and its back spikes all give Totodile a prickly, triangular look to it. This is a notable departure from most of the designs in Gold and Silver: notice how many Generation II Pokemon are circles or at the very least spheroid and squishy. Natu, Igglybuff, Octillery, Forretress—they’re all just circles. Kurusu fell into this trap; it wasn’t circular per se, but it did have a soft, huggable look to its body that didn’t help it stand out. Even if you don’t immediately notice why, Totodile looks distinctive amongst the rest of Generation II’s Pokemon, helping make it far more memorable. Totodile is also a triumph because it gave Game Freak a way to resolve a puzzle they’d been trying to solve since the initial stages of Gold and Silver’s design. The problem was that new starter Pokemon were tricky to design: two different gameplay needs pulled their designs in opposite directions. On the one hand, the designers want to make Gold and Silver new experiences for players: they don’t want the game to feel exactly like what they had done before in Pokemon Red. Thus, the designers would want to make the starters as different from Charmander, Bulbasaur, and Squirtle as possible. On the other hand, the starter Pokemon fill a very specific role in the gameplay of the Pokemon games. The starters are supposed to be flexible, all-around strong Pokemon that are good no matter what team you construct with them. They’re also supposed to have a huge movepool, so that they offer maximum flexibility to the player, and good enough stats that they can carry an inexperienced player through hard sections of the game. Thus, the designers would want to make the starters all pretty similar, because the stronger they are and the larger the movepool, the more choices a player would have. So one gameplay need wants the starters to be different from the first generation of starters; the second wants them to be more or less the same. What are the designers to do? When we discussed Chikorita and Cyndaquil, we discussed one way the designers were able to square this circle. Chikorita and Cyndaquil have very similar stats to their counterparts (in fact, Cyndaquil’s stats are identical to Charmander’s) in Generation I, and their movesets aren’t very different either. The way to make them feel unique comes from the design of the games. In Generation I, Bulbasaur was the easiest starter to use, while Charmander was the hardest. However, in Gold and Silver, Chikorita immediately faces difficulties and throughout the game has a harder time than the others, while Cyndaquil blazes through Bugsy and has a decent time against most of the other gym leaders. Thus, even though Chikorita’s really similar to Bulbasaur, the experience of using Chikorita is quite different. Totodile, on the other hand, tries to solve this tension by providing a drastically different experience in a Water starter, while still being a flexible, all-around good Pokemon to use. While Squirtle was a defensively oriented Special Attacker, Totodile’s stats are distributed much more towards Attack and HP, meaning that it’ll excel with physical moves, not the Water moves that Squirtle’s more or less balanced around. As a result, Totodile learns much more physical attacks than Squirtle: it learns Rage and Slash, which Squirtle does not, and it only learns one single Water move before level 52. As a result, Totodile is going to spend most of the game using its physical stats, rather than even bothering with same-type moves. Notably, Kurusu has a third distinct playstyle. While we don’t have stats for Kurusu, it learned Aurora Beam and Body Slam, both strong sweeping moves, alongside Safeguard and Mist to make sure it wasn’t debuffed. If Squirtle was a defensive special user, and Totodile a physical dps, Kurusu seems to have been designed to be a versatile offensive Pokemon. So while I love Kurusu, I can definitely see the argument for replacing it with the Totodile line. And given how popular Totodile is with the Pokemon community, that decision has been borne out by time. Totodile’s an interesting, effective design, one of the biggest successes of the rebooted 1999 development. ID X03: Croconaw Croconaw is Totodile’s grown up brother. A little bit bigger, and sporting what almost looks like a cave-man fur across its body, Croconaw is still cute, but notably looks far more grown up and a bit more dangerous than Totodile. There’s not much that’s new with this guy, but let’s jump in and see what we can find. My gut feeling is that Croconaw and Feraligatr were designed after Totodile. Croconaw, for instance, looks like the stereotypical second-stage evolution, clearly awkward looking on its own but acting as a design bridge between the aesthetics of Feraligatr and those of Totodile. Croconaw has the white, pronounced jaw of Feraligatr, but its pose and the spikes on its back look a lot more similar to Totodile. If I had to guess, however, I’d say all three were designed more or less at the same time. Their stats were all changed at the same time in the data we have, and Totodile, Croconaw, and Feraligatr are some of the few new Pokemon that have both a finalized backsprite and frontsprite by June 1999. This suggests the team thought of them as a group to be worked on together (as opposed to the Cyndaquil or Chikorita lines, which lack one or two key members of the family until much later). If I had to guess, I’d posit that Totodile was designed by a different creator than Croconaw and Feraligatr. Notably, the spiky, hard angles of Totodile are gone in Croconaw’s design, replaced with more curves, most notable in Croconaw’s large snout. The eyes are slightly different as well, but most strangely, the palette for Croconaw (and Feraligatr) is a green/turquoise color, whereas Totodile is a bright blue. If you look closely, the palette is actually different for all three, with Croconaw’s being just a slightly lighter shade of green than Feraligatr. This is just another example of how Croconaw feels like it was purposefully designed as a bridge: even though the difference is barely noticeable, the designer of Croconaw wanted even its color scheme to be a transition from Totodile to Feraligatr. Obviously, this palette difference could be for a million reasons, but one possibility is just that someone else was working on Croconaw and Feraligatr and settled on a different shade of blue they liked more. Obviously it wasn’t a palette change that the team liked overall, because in Crystal version they changed all the sprites of Croconaw and Feraligatr to just share the same blue that Totodile uses. One thing that is especially notable is how the Crystal sprite brings Croconaw much more into line with Totodile's aesthetics. If you examine it closely, not only does Crystal Croconaw have the same blue palette as Totodile, but the floppy bangs it has on its head in the initial sprite have been replaced with spikes more in line with Totodile. For some reason, the back sprite has also switched which shoulder the cave man furs tie over, even though that doesn't match the front sprite. Who knows why. We know that Crystal's sprites were tweaked to better match Sugimori's official art that was released after Gold came out. Sugimori based his Croconaw art more on Totodile's coloring and design, so these tweaks to Croconaw's sprite were probably to bring it in line with those. Does that mean that Sugimori was the one who designed Totodile? Does that mean he's the one who edited the Crystal sprites? It's hard to say--the shading of GBC sprites makes it hard to say anything definitive. If Croconaw was made by a different designer than Totodile, it was probably designed by the same person who created the original three Water starters for Generation II. We’ve discussed this before, but it’s worth bringing up again here. For whatever reason, the designer decided to leave an easter egg in Croconaw’s design by leaving it with the exact same pose of its predecessor, Akua: Alright, alright, the pose isn’t completely definitive. Their heads are angled up at the exact same angle, but of course, they have completely different bodies and their tails are in different positions, so I’m willing to grant that it could be a coincidence. But man, I look at these two sprites and their similarity feels too obvious to ignore. On top of that, their faces also share a likeness: just look at the ovular shape of the mouth and snout. It seems to me pretty much beyond doubt that the designer of Croconaw put just enough of Akua into the new creation so that it kept some of the personality and was a subtle reminder of what came before. The only other notable thing about Croconaw is that its Silver Pokedex entry changed between the end of Spaceworld ’99 and the final. In the early silver entry, the Pokedex makes reference to a behavior of real-life crocodiles: It spends most of the day underwater, with only its eyes and nose sticking above the surface. I don’t think this entry really works; given the pose Croconaw has in its spritework, it doesn’t feel like the sort of creature that lurks below the waters’ surface. I mean, look at that defiant pose it’s striking! This is a dude who doesn’t hide, but faces down his enemy. Game Freak might have thought something similar, so by the final, they gave Croconaw a new entry, which was actually just reworked from an early Totodile’s entry that was substantially the same: It opens its huge mouth to attack. When it bites something, any fangs that fall out will grow back one after another. Meanwhile, Totodile got this entry, expressing a key feature of its excitable personality: It is very rowdy despite its small size. It will bite anything that moves in front of it, no matter what. Honestly, it doesn’t feel like the Pokedex team worked very hard on Croconaw. Its early entry and its final just reference fun crocodile facts: Crocodiles spend most of their time lurking beneath the water’s surface, and when they lose their teeth chomping on things, crocodiles are also known to grow them back. Which is fine and all, but I don’t really get a sense of Croconaw’s personality from its revised Pokedex entry. Honestly, it seems like the team just opened an encyclopedia to the crocodile entry and wrote down whatever seemed the most neat. Croconaw’s not a bad design, by any means, but it certainly is the least memorable of the Totodile family. Like I said about Totodile, I personally prefer the original Water starter line, especially the middle evolution, Akua, that Croconaw replaced. Still, I understand why they did it, and Croconaw retains a lot of the mischievous energy that the team crafted into Totodile. I’m certain that even Croconaw, the overshadowed middle stage evolution, has more fans than Akua ever would have. ID X04: FeraligatrAnd finally, we make it to Feraligatr. Feraligatr’s ferocious: Our boy Totodile has completely grown up, leaving this absolute monster in his place. While Meganium’s my personal favorite Generation II starter, Feraligatr is a great example of how to transition a first stage Pokemon from cute to fearsome but still keep it feeling like the same ‘mon. Feraligatr feels like a natural extension of Totodile—well, minus the strange turquoise color that they thankfully got rid of by Crystal—in that it keeps all of the traits of Totodile but just ramps them up to eleven. It is worth noting that Feraligatr, according to its name, officially changes real-life species when it evolves: Totodile and Croconaw are both crocodiles, and Feraligatr is an alligator! Obviously, the difference between these two species is small, and so this isn’t as big a deal as, say, a Pokemon transforming from a fish into an Octopus (what kind of design would that be?). What’s the difference between Alligators and Crocodiles, you might ask? Well, there are three main ones: Crocodiles are greener in color typically, Crocodiles live in freshwater while alligators live in salt-water, and Crocodiles have a pointed snout while alligators have a rounded snout. Feraligatr breaks this first rule, by being greener than the other two (maybe Sugimori changed its palette to address this inconsistency!). But if you look at its sprite, it absolutely has a more rounded snout, a clear contrast to Totodile’s. Croconaw, the odd one out, unfortunately has an alligator like snout. (Alligator on left, and Crocodile on right) As for the second difference, I have no idea whether Feralgatr prefers salt water or fresh water. It’d be pretty awkward if, on evolving, a Croconaw suddenly couldn’t breath in the fresh water environment in which it lived its whole life. Everything that we said about Totodile goes for Feralgatr, so there’s unfortunately not much more to add here. Feraligatr’s stats started out as a reflection of Blastoise’s, before becoming focused towards HP and attack; likewise, Feraligatr specializes in biting and slashing moves, as opposed to the heavier dependence on water moves that Blastoise and Akueria had. Feraligatr, like its two relatives, was also substantially finished by the first sprites we have, in June 1999. In fact, the only things that got changed around before the final of Gold and Silver were the Silver sprites for each member of the Totodile family. Totodile’s early Silver sprite was the same pose and concept, just cleaned up a little for the final game; Croconaw’s early and final both show it growling at an enemy, but the pose is pretty different between the two, and substantially cleaned up by the final. Feraligatr, however, has the Silver sprite that went through the most drastic transformation. In the original Silver sprite, which we first see in Spaceworld ’99, Feraligatr is on all fours! It’s a pose that Feraligatr has never been associated with, even on the anime, so it took a lot of people by surprise. I, certainly, think it’s a really cool pose for Feraligatr, and it gives it a lot of personality. The designers may have thought it diverged from the Gold design a bit too much, however, because the final changed the Silver pose. Its head is still pointing downwards, and we’re viewing Feraligatr from an angle pointing almost directly at the camera, but now Feraligatr is standing up on its hind legs, just like the Gold sprite. It very much feels like this revision was made to bring the Silver sprite more in line with the official design, not because they disliked the previous Silver sprite. There is one thing I find fascinating about this final Feraligatr sprite. Look at its pose, and compare that pose with Akueria, the water starter that Feraligatr replaced. Don’t those poses look similar? Obviously it looks different because Akueria has such a long neck, but the angle facing the camera is almost the same, and though Feraligatr’s tail is hard to see, it appears to be in the same position as Akueria’s. Certainly the similarities in poses are not as definitive as Croconaw’s. But I also have a hard time ignoring these similarities either. That’s it for the Feraligatr line. From what we have, their design history seems pretty simple. However, that early Totodile sprite suggests that there was a lot of work done on these guys in the period of development between the Korean Index and the June 1999 sprite sheet. Unfortunately, we have so little information from that period that any interesting experiments concerning the development of the Totodile/Croconaw/Feraligatr sprites are lost.
Saying that, what we know seems pretty conclusive. It appears that Masuda wanted to replace the Akua line, and that’s exactly what happened. It seems as though the design team was very decisive about the replacement Water starters, and quickly came up with a winning concept and strong sprites. However much time the team spent perfecting the sprites of Totodile, Croconaw, and Feraligatr, they were very happy with them by June 1999. The team stuck it out with these sprites all the way until Crystal, demonstrating just how good these designs were.
11 Comments
anglure
11/5/2023 09:54:39 am
A bit of a misconception on alligators vs crocodiles: Both live in freshwater, but crocodiles are more tolerant of saltwater than alligators. The saltwater crocodile in particular is a crocodile species known for traveling miles out to sea!
Reply
PJIceskull
11/5/2023 12:26:11 pm
Interesting enough for Feraligatr, it's pokedex entry in Silver says that it falls on all fours due to it's body. I wonder if this Spaceworld ’99 sprite is meant show this or if then entry was referencing the sprite.
Reply
Asmorano
11/9/2023 02:10:44 pm
This is a great point I missed. You're right, I bet that Pokedex entry was based on the earlier sprite.
Reply
Commenter
11/6/2023 02:12:43 pm
Regarding the alligator/crocodile thing, the Japanese names for Croconaw and Feraligatr are Alligates and Oodile respectively, so their designs match their original Japanese names. (Totodile is Waninoko, with wani being the Japanese word for crocodile/alligator)
Reply
Asmorano
11/9/2023 02:10:07 pm
Oh duh I should have checked the Japanese names! This makes sense, thank you!
Reply
Trip
11/9/2023 05:03:39 am
Alright, I hate to be that guy, but I've read this line something like two dozen times and can't figure out what it means:
Reply
Asmorano
11/9/2023 02:09:36 pm
DPS (Which I should have capitalized) is an MMO term that stands for "Damage Per Second." It's a common term to describe a character in a game whose main job is just doing a lot of damage to the enemies as fast as possible. Physical DPS usually means a sword fighter or a close range combatant, as opposed to a "Magical DPS" or a "Ranged DPS." It's usually contrasted to the "Tank," which is supposed to take damage and protect the other members of your party, and a Healer, which is supposed to...heal.
Reply
Trip
11/18/2023 04:24:45 am
Ah, okay. As someone who is not a competitive gamer and doesn't play any MMOs, that was completely lost on me. I've heard the term "sweeper" used to discuss those who do a lot of damage in Pokemon, though again, I don't do competitive. And the fact that it's not in all caps definitely obscured that it was an acronym. Thanks for the clarification!
HMB
11/13/2023 11:36:14 am
I liked this article. Will other generations be worked on?
Reply
Asmorano
11/13/2023 12:02:10 pm
I'm glad you liked it! Hopefully you've checked out the rest of this very long project?
Reply
Great article as always, it seems this era will be full of surprises.
Reply
Leave a Reply. |
AuthorMy name's Aaron George, and I'm both a historian and a fan of Pokemon, especially of development. Reach me at @Asmoranomardic Archives
February 2024
Categories |