Sakurai on Creating Games: Game Concepts (12 Titles)

Masahiro Sakurai, the creator of the Kirby and Super Smash Bros. series, shares a wealth of game-development know-how on his YouTube channel “Masahiro Sakurai on Creating Games”. This article summarizes and restructures the content of that channel by topic.

That said, a summary is only an entry point. So much of the value — Sakurai’s own words, his real-world examples, his pacing, and the footage itself — can only be gotten from the videos. So please don’t stop at reading the article; I strongly encourage you to also watch the original videos embedded under each topic.

This category covers the concepts behind the games Sakurai himself directed, told in production order. Here we bring together the key points of all 3 videos in the “Game Concepts” category (12 individual topics), structured in 3 parts following the 3 source videos. Each part opens with its explanatory video, so please watch along.

Part 1: Dream Land, Adventure, Super Star, Smash 64, Melee (#01–#05)

This category has Sakurai narrating, in production order, the concepts behind the games he directed himself. This time it’s five titles, from his debut work through Smash Bros. Melee.

https://www.youtube.com/watch?v=_gbJw7orSI8

1. Kirby’s Dream Land

The first game Sakurai directed — his debut work. He wrote the first proposal at age 19. He wasn’t yet consciously thinking in terms of risk and reward, but he had a vague sense that “using enemies is fun” — because risk and reward are two sides of the same coin.

Looking for an even easier way to make use of the enemies right in front of you, he first considered a mechanic of snaring them with the tongue, then ultimately settled on inhale and spit out. The other key feature, flight, came from feeling that an instant miss on any fall was too harsh in a health-based game. Touching an enemy and falling off a cliff aren’t that bad, yet they meant instant death, which felt heavy — so he kept the one-miss rule but let you bypass falls by flying.

Both inhaling and flying push the game toward being easier and lower-risk. There was a danger of making it less fun, but the original Kirby held firm to tuning aimed at gaming beginners. The Famicom of that era, limited by its small capacity, was full of games that raised difficulty; with unreasonable difficulty, beginners can’t get in. Kirby’s role was to guide gaming beginners into the world of games. Even though it was easy, it felt good and people replayed it many times. From the start it showed a fixation on music — the stage opening synced perfectly to the intro, the clear-dance, and so on. As a result, the first Kirby became the best-selling title in the Kirby series. Building it as a game for beginners, he recalls, was an excellent decision.

2. Kirby’s Adventure

The second game in the series, and the first Kirby with Copy Abilities. Around the time the first game was finished, the company filed for composition (on the brink of bankruptcy), and he even experienced PCs being tagged for seizure. The instruction from the company was “make a Kirby on the Famicom”. The Super Famicom had already been out for a while, yet it was the Famicom — probably because they wanted to make and sell it quickly.

But Sakurai felt the “for beginners” concept wouldn’t work on a fully matured Famicom market. So the challenge he set was “how to satisfy beginners and experts at the same time.” The answer was Copy Abilities.

The unpracticed can clear with just inhale and spit out at minimum, while the experienced master Copy Abilities for a far wider range of play — a separation of audiences. And Copy Abilities explosively raised Kirby’s character appeal, making encounters with new enemies something to look forward to (being able to give each enemy individual meaning was huge). The problems of a wider screen and stronger flight were tuned with larger info-boards, mid-boss scroll locks, and a slower flight speed. Three one-button sub-games were also a service to beginners. Giving Kirby Copy Abilities made an immense contribution to the series’ later popularity. He says he was also glad to have made a Famicom game.

3. Kirby Super Star

Released in 1996 on the Super Famicom. It had two big concepts: two-player simultaneous play and an omnibus structure.

The simultaneous two-player play started when Miyamoto asked, “Can’t you do two-player co-op in Kirby?” Mario is hard because of its speed, but Kirby is slow, so maybe it could work. But Kirby already had Copy Abilities, whose speed is nothing to scoff at. So the idea was the Helper system = separating the lead and the supporting role.

The camera keeps following the lead, and the support, even if it goes off-screen, returns via a Space Jump. By turning an enemy character itself into a playable character, both fun and personality grew. This is also where the concept of copy prevention was born (which would later cause enormous trouble in making Smash Bros.). Copy Abilities gained varied moves via inputs, and even minor enemies got a concept of durability.

The other concept, the omnibus, was a structure of choosing among several scenarios pinned to a corkboard. Thinking that games that take a long time to reach a conclusion are tough, he gathered different, quickly-resolving styles of play so that, even at high volume, you could play without getting bored. While he regrets that CG backgrounds were a poor fit and didn’t work out, the result was a definitive package combining two-player play, omnibus, and Copy Abilities. The only side-scrolling Kirbys Sakurai directly directed, he says, are the original, Adventure, and this one.

4. Super Smash Bros.

After Super Star, from 1996 the NINTENDO64 era arrived. Sakurai studied 3DCG on his own and drafted two proposals for 64 software. One was a four-player battle-royale fighting game = the prototype of Smash Bros., the other a radio-controlled robot adventure. The prototype (planning, models, and motion by Sakurai; programming by Iwata), nicknamed “Kakuto-Geemu Ryuoh” (Dragon King: The Fighting Game), had no special moves or items, but its rules were Smash Bros. through and through.

Smash Bros. is often called anti-fighting-game, but it doesn’t reject fighting games. He simply felt combos aren’t a mind game (they only increase the time you spend just getting beaten), and so created an accumulated-damage system where reactions differ every time, and you lose if you’re knocked out of bounds. Against the trend of increasingly complex command inputs, he aimed for intuitive control with direction + button, and the flick input (smash) was born too. Nintendo characters appearing wasn’t in the proposal; that was negotiated later. He tackled head-on the difficulty that a crowd of unknown protagonists suddenly lined up is hard to sell, and chose to borrow popular characters. Before release there was resistance — “characters beating each other up?” — but after release word of mouth spread it gradually. It was also his first time casting voice actors, and Kirby’s official voice was decided here. The policy of letting you play just the parts you like, the way you like, he says, was established to a degree in this first game.

5. Super Smash Bros. Melee

Sakurai’s proposals are often logical (problem → solution), but Melee’s concept was extremely simple. Bluntly, “dramatically power up Smash Bros.” Compared to the original, the fighters, stages, graphics, effects, motion, sound, volume, play feel, items, and extras were all powered up.

The 64 ran at 300 polygons per fighter (about 200 in battle), the lowest tier for a fighting game — Samus’s leg even looked triangular. Melee, on the GameCube, had a big jump in power and was easier to build. As his first optical-disc development, streaming playback (live audio) became possible. He adopted several orchestral pieces, which led to the decade-long “PRESS START” and to later collaborations with musicians. CG movies were also a first (he gave up his New Year’s break to write the plot). The extra trophies were introduced to bring the joy of collecting and guiding through Nintendo’s history, given that the roster of usable characters is limited. Many works need to exist on the basis of being “found nowhere else” — so he put his energy into making the most of that single greatest feature. Melee became the best-selling title on the GameCube.

Part 2: Kirby Air Ride, Meteos, Mushiking, Brawl (#06–#09)

This time, four varied titles he worked on around the time he went freelance. A racing game, a falling-block puzzle, a bug-raising LSI game, and on to Smash Bros. Brawl — the core of each concept is laid out.

https://www.youtube.com/watch?v=OC3bQC1lkQ4

6. Kirby Air Ride

In 2001 Kirby got an anime, and Sakurai became the supervising director of the Kirby series. The anime needed an accompanying lineup of games, and he judged that a racing game, Kirby Air Ride, would take less time than a proper mainline action game.

For the gameplay, he first assumed drifting is fun. Grip driving should be faster, but drifting has a unique thrill (risk and reward). You charge while braking, and release it for even more fun. And feeling that a game about aiming for the top doesn’t suit this, he designed not toward shaving off tenths of a second but toward broadening. Roam the City Trial sandbox freely and enjoy not knowing how things will turn out — that’s the big concept. With the Clear Checker (predating the achievement systems of the PS3/Xbox), he gave a sense of accomplishment even to casual play.

But the project stalled: more than a year after he’d handed it off, not even a single Warp Star moved properly. Sakurai returned as director, rebuilt it entirely, recut the spec, and finished it in three and a half months (thanks to the courses and assets being done, and the staff really engaging). The machines in the final product feel distinctive, but internally they’re just parameter differences (all parameters written by Sakurai). It’s packed with gameplay you won’t see elsewhere and remains highly popular today. After finishing Air Ride, Sakurai went freelance.

7. Meteos

Now freelance, Sakurai was thinking about how to contribute to the game industry. Rather than spending years on a single title, he spent a period doing supervision and consulting (he keeps the title names private). Around then, ex-Sega’s Tetsuya Mizuguchi asked him to “make a falling-block puzzle.” Even then it felt like an old image, but he found it an interesting challenge.

Sakurai is hopeless at falling-block puzzles (like Puyo Puyo), but if you know the principle that generates gameplay, you can plan it. Falling-block is an easy-to-grasp genre where stacking raises risk and can be turned into reward all at once. So he questioned, from the root, “do the blocks really have to be erased,” and arrived at the idea of using matched blocks as thrust to grandly launch them. It’s ridiculous and impactful, and techniques like vertical stacking and mid-air docking emerged too.

Likening the blocks to meteors raining down on planets, and turning it into an interplanetary war of pushing them onto each other, became Meteos. The target platform happening to be the Nintendo DS was coincidence, but it paired well with touch controls. Unusual for a falling-block game, its hallmark is that on each planet the block shapes, field width, properties, and music all change, giving the puzzle real character. The finished game is engrossing, but due to load-reduction measures it didn’t reach the intended balance and shipped with many bugs still in (all planet balancing set by Sakurai). He wasn’t involved in the later variations and learned of them in magazines like an ordinary customer — a regret — but he looks back on it as valuable work that made a falling-block game shine with singular character.

8. Sodatete! Mushiking: The King of Beetles

The game made by the smallest team among Sakurai’s works is the raising-LSI game Sodatete! Kochu Oja Mushiking (Raise! Mushiking). Its basis, Mushiking: The King of Beetles, was a 2003 card-based arcade hit aimed mainly at lower-grade elementary kids: for 100 yen out came a beetle card, and you’d play a read-the-opponent battle via rock-paper-scissors (each bug has a winning hand).

It began with a request from Sega’s Takezaki (now president of TMS). The idea was to combine an LSI raising game with Mushiking. Mindful that strength differences arise during raising, so battles could break, he cared for the mood — for instance, showing feeding as slow nibbling. He wrote the proposal over Golden Week 2005, and right after, at E3, received the offer for Smash Bros. Brawl, working on both in parallel. Even then the work was telework-like (on both sides), and they gathered only a few times. The programmer was Naito, the main programmer on Dragon Quest III and IV. He couldn’t test on the handheld until completion, and there was a miscalculation — perhaps because the development hardware and processing speed differed, hunger set in faster than intended — but the game itself was fun, and he says the experience of making an LSI game was a good one.

9. Super Smash Bros. Brawl

The Smash Bros. Brawl offer came at E3 2005. Had freelancer Sakurai not taken it, it might have shipped as-is with Melee’s 26 characters untouched (per Iwata). The biggest production concepts were two: online support and a long-lasting single-player mode.

For online, while thinking Smash Bros. is by no means a good fit for it (it needs fully-synchronous communication, which is unfavorable, and you mostly connect only with nearby people), he “did it because it’s the challenge.” The moment it actually worked was deeply moving, he says. The single-player became “The Subspace Emissary,” enriched with movies, maps, sticker-based power-ups, and an ending — he wanted to firmly establish the characters. The effort rivaled a full game; jump and movement parameters were set separately, and adding fighters meant tuning double the number.

New elements included Final Smashes (signature moves that highlight each character), Assist Trophies, and third-party guests Snake and Sonic (as Sakurai puts it, “to me, even Nintendo is another company”). My Music realized collaborations with many songs and musicians. And from the lesson that Melee’s high control difficulty was harsh on beginners, he made the speed slightly slower for the Wii’s casual audience. He introduced test players (four) for the first time. Had he been unable to make Brawl, he says, the Smash Bros. series we know today might have ended.

Part 3: Kid Icarus: Uprising, Smash for 3DS/Wii U, Smash Ultimate (#10–#12)

The “Game Concepts” finale covers three titles from the 3DS onward. From the revival of Kid Icarus to the final mission Iwata entrusted to him, “Smash Bros. Ultimate,” the culmination is told.

https://www.youtube.com/watch?v=ouH6tvozsy0

10. Kid Icarus: Uprising

After Brawl, President Iwata gave two orders about the then-unannounced Nintendo 3DS. One was to propose specs, the other to make a dedicated game. For an outsider to be involved in deciding a new console’s direction was unprecedented at Nintendo.

Sakurai left the casual route to others and set the concept of making a meaty, rich title that satisfies gamers too (going the opposite). To make use of stereoscopy he chose a TPS (the approach to the battlefield is a rail shooter; after arrival it’s a TPS). To a genre whose theory was already settled, he tried to bring a new way of showing it, as Smash did to fighting games. Adopting Kid Icarus as the theme came later (it was much-requested overseas, and he’d strengthened Pit in Brawl; the idea that Palutena casts a “Miracle of Light” letting you fly for five minutes seemed fun).

The core of the concept was, first, time-limited aerial battles (rail shooter) — since a TPS from the outset is hard for beginners, an attempt to fit play you can do without much thought to musical development. Next, ground battles (TPS) — since aiming at a small area on a handheld is unsuited, he made enemies many times bigger (even minor foes over four times Pit’s size). The slide-pad flick input (Smash know-how) was used for dashing and emergency dodges, and aiming used the touch panel like a trackball. Almost the entire scenario unfolds during play, and since the original had a cheeky tone, he made it comical (all written by Sakurai). Weapons were treated like characters across 10 categories, and that no “must-use” standard emerged without updates was a good thing. He also introduced AR Cards and the combined Three Sacred Treasures (which also appear in Smash). Development was rough due to the 3DS’s unstable performance and a cobbled-together team that didn’t mesh, but he says he thinks it’s a masterpiece — a game he’d love to play on a smooth, large screen.

11. Super Smash Bros. for 3DS / Wii U

Told “the red handheld or the blue console — which will you pick?”, Sakurai took both. At E3 2011, while swamped with the new Kid Icarus, it was announced that Smash Bros. would come to both the 3DS and the Wii U.

The first handheld Smash would show off the 3DS’s charm, while the Wii U, an HD machine, would show off graphics — but both struggle to escape being extensions of the past. So he focused on linking “the individual (carrying it on a handheld) and the gathering (meeting around a console).” Even across two systems, sharing systems and motion meant it wouldn’t be twice the work (what’s possible is pulled down to the 3DS’s power). The 3DS got Smash Run (adapting Air Ride’s City Trial style, using the trait that each player has their own screen), and the Wii U got Smash Tour (a party game). Custom fighters let you customize special moves, and there were Mii Fighters and amiibo support (realizable precisely because there’s strength variation; Smash’s physical figures were groundbreaking).

On the technical side, keeping four-player battles, stereoscopy, and 60 frames per second on the 3DS is truly amazing (the fruit of BNS’s tech). Smash’s first online updates became possible, enabling fighters to be delivered via DLC. As a groundbreaking internal advance, at last people besides Sakurai began building parameters (from then on, several planners tuned, and test players increased). From E3 2013, reveal trailers took off in earnest, beloved through to Ultimate as an entertaining form of announcement.

12. Super Smash Bros. Ultimate

Smash Bros. on the Nintendo Switch was requested back when the Switch was still unannounced. This was the last mission Iwata gave him. The concept was clear, bluntly “Everyone Is Here.”

Past Smash games had wildly different team makeups per title, so assets couldn’t be reused — but Smash 4 and Ultimate were built at the same Bandai Namco Studios. Staff, environment, and assets carried over, and the move from the Wii U was relatively easy. Past fighters not in Smash 4 were surprisingly only 7 (Pichu, Snake, Young Link, Ice Climbers, Wolf, Squirtle, Ivysaur). He forced through “if there’s any chance to have every past fighter, it’s now or never.” The proposal was over 200 pages (it even raised the possibility of VR support). New fighters were considered with input from a survey (about 1.8 million valid votes). He listed 70–100 tuning items per fighter.

The play feel was paced about between Melee and Smash 4 (exciting for veterans, but right on the line where the unpracticed can still keep up). Stages were merged to over 100 (300 with Omega forms, etc.). The crisis of half the capacity, due to the card medium (16GB), was solved by compressing the music to 40% of the Smash 4 era with a mysterious technique. For single-player, with limited room amid the explosive volume, he planned Spirits (in place of figures; evoking the originals through pseudo-recreations via combinations) and World of Light.

First revealed in March 2018, with “Everyone Is Here” announced in September and release in December. But it didn’t end there — on to DLC production (Joker’s entry announced the day before release). During development the COVID era began, and they quickly shifted to telework. Communication and test play became hard, but as discussed in advance, all DLC was completed and delivered within 2021. It took three years (a span rivaling main-game development), though. The future of Smash is unknown even to Sakurai, but a Smash expanded to the limit can’t keep going down the same path of expansion — it’s reached a very difficult place. And what he planned with his free time after developing Ultimate was this very program — with the words “the next project is this program,” the Game Concepts come to a close.

Summary

We’ve now gone through all 3 videos and 12 topics of the “Game Concepts” category. Finally, let’s recap the key points of each part (= each source summary video) by individual topic.

Part 1: Dream Land, Adventure, Super Star, Smash 64, Melee (#01–#05)

Part 2: Kirby Air Ride, Meteos, Mushiking, Brawl (#06–#09)

Part 3: Kid Icarus: Uprising, Smash for 3DS/Wii U, Smash Ultimate (#10–#12)

There’s a lot here, but I hope you’ll revisit it starting from whichever part caught your interest. Please also check out the related categories.