[felipc]

Has Nintendo ever talked about how they do software development? Can we all drop the thousands of books that have been written about software engineering in general and just figure out what they do?

Game aside, the reviews have been pointing out how the game performs well (after day 1 patch) and is not pestered with bugs, which is an impressive feat for such an open world game where most things are able to interact with everything else.

[Server6]

It 100% has to do with retention of key talent and knowledge transfer. It seems the model for most western studios is to make one or two big successful games, then layoff all the staff and/or be acquired by EA/Activision/Microsoft. Then their next games flounder as they're milked dry. Western companies are only worried about the next quarter and treat talent as a bottom line expense.

[Matthias247]

I guess it's not only about what the studios do. Japanese also have a different kind of loyalty from employees which rarely ever change change. It's probably 20 years of average tenure, compared to 2 years in the US. And that's not because these companies pay so much more: They probably pay less for their most experienced stuff than what an employee with 2 years of experience gets in the US. It's just a different culture.

[PurpleRamen]

There is no big secret to this. They just don't go all out. They don't take big risks. They just see what works elsewhere and polish it until it shines brighter than everything else. And they only add (or take away?) until they have the necessary minimum of gameplay. For example, Zelda BotW is by far not the best survival or crafting-game which was around at release, but it was the most pleasant experience for casual gamers and Zelda-fans, because it left out all the unimportant grind which is not relevant for a Zelda-Game.

Notable in that regard: Apple did the same under Steve Jobs. Focus on the important part, and don't play around.

[flatiron]

Personally I think BOTW wasn’t “seeing what works everywhere else and polishing it” it really felt like a new game nobody had ever done before.

[X6S1x6Okd1st]

I don't think it had any new mechanics or concepts, but it def felt new. Do you feel like there were new mechanics or concepts?

[d23]

I would invert that question and ask what, to you, would qualify as being sufficiently new that it doesn’t have? To me, there are tons of things. The death / checkpoint system, weapon durability, the massive non-linear open world, the recipe system, the puzzle dungeons, the fact that, if you want to, you can essentially go challenge the final boss immediately. If these things aren’t new enough, I have to wonder what is.

[filleduchaos]

Most of those things have been fairly common elements of games for decades? For example "puzzle dungeons" is just a staple of Zelda as a franchise, there's a whole genre of games designed around being able to rush the final boss as quickly as possible even though there's plenty of other content to explore (Metroidvanias), and I can't even think of anything particularly remarkable about BotW's death/checkpoint system other than that it's fairly generous with the autosaves.

[d23]

My point is these are new for a Zelda game, and they've put them together in a complete package in a way that's not been done before. If the bar is "something no game has ever done before," it's rare that anything in life is ever going to reach that bar.

Even if you take a completely different profession, like music, revolutionary artists still have their influences and build on instruments and techniques that are 99.9% the same. It's not like they are suddenly playing flutes made out of loaves of bread. And even if they were, most of the time those sorts of things just come across as gimmicks to me.

[djur]

I'm surprised to see "Metroidvania" described as a genre where you can rush the boss quickly. Neither of the two defining games that name the genre, (Super) Metroid or Castlevania: Symphony of the Night, allow you to reach the final boss without having explored a substantial amount of the world map and collected the majority of available upgrades. Both games do have low% and any% speedruns that skip a lot of stuff, but those require the use of glitches. Do you have an example of a game you're thinking of?

[X6S1x6Okd1st]

> what, to you, would qualify as being sufficiently new that it doesn’t have?

I don't understand the question, like what's a game play mechanic that it doesn't have?

[neogodless]

Are you talking about different things?

Those all sound like things I've seen in lots of games, but not necessarily in Zelda games.

[urthor]

Most good things in life are like this.

Execution. Not originality.

[tjpnz]

I would put a lot of it down to Japanese craftsmanship coupled with relatively experienced engineers (average age in their Kyoto office is 40 IIRC). Their selection process is notoriously rigorous too and goes far beyond the usual LeetCode questions you'd get at a FAANG company.

[SkeuomorphicBee]

> coupled with relatively experienced engineers

I disagree. Nintendo has good engineers but so does many of the other studios. For me what sets Nintendo apart is not their code or technology, but their game design and game direction. The way they seem to craft their game-play and game mechanics to have everything it needs but nothing more, and then couple it with the perfect match for game aesthetics with unmatched consistency.

[ska]

Do Nintendo keep their teams together longer?

You usually can’t get a great engineering team by throwing a bunch of good engineers together, it takes time and work.

I don’t know much about the game industry but from what little I have heard it seems there is a lot of churn.

[tjpnz]

This is a Japanese company so most of their engineers will be hired directly from university and typically stay on until retirement. Based on what I've heard they build large groups of engineers who'll stay together more or less permanently. Then they'll rotate these groups between different projects. Sometimes they'll be on a game, other times they might be doing something with hardware. So the groups end up multidisciplinary.

[ska]

That’s basically ideal for a lot of engineering work, so long as you keep fresh ideas coming into the group somehow so it doesn’t stagnate …

[haunter]

> I would put a lot of it down to Japanese craftsmanship

This is such an orientalist and borderline racist view it’s crazy. If it were true then it would also imply the other japanese game devs also affected by it. There countless bad games from Japan, don’t even have to walk far from Nintendo just look at the Pokémon games and how GameFreak release them with 0 optimization. And then the countless misses from Square Enix, Bandai Namco etc.

[tjbiddle]

Very tired of people saying something is 'racist' when in actuality the comment is just a generalization of a certain people.

Racism is hating a cultural group or having a prejudice against them, not making a comment on it.

If anything, OPs comment is a compliment on Japanese work.

[joenot443]

No one minds attributing the wild spunk and ambition of USA startups to “American exceptionalism. I think it’s pretty reasonable to say that a certain trait is broadly associated with a culture without implying EVERYONE in that culture has to exemplify it.

Grow up man, this is some elementary stuff that you shouldn’t need explained to you.

[illwrks]

I think you're missing the point. My family in Japan, and their social/business networks, are exactly like this. Detail oriented and loyal to a fault. While they are not software engineers the expectation is to do your best. Some of it is done through company policies or implied in a social context. That is not true for all Japanese people but it is true of a large majority. I have first hand experience.

[illwrks]

I guess it depends on the company and what it produces also.

[whywhywhywhy]

It’s weirder to suggest everywhere in the world works exactly the same and culture has no impact.

I haven’t worked with Japanese companies but Chinese companies work couldn’t be more different than western.

[zpeti]

Japanese are also extremely attached to their past and culture, which shows up in the game how its almost like a modern mythical representation of both old japanese myths and an obsession with technology.

There, is that racist too?

The result is beautiful, I love the whole "ancient technology" concept in it, which is not something we have in our world (except for pyramids perhaps).

[PurpleRamen]

> If it were true then it would also imply the other japanese game devs also affected by it.

That's a pretty racist argument, to generalize this way. Not all Japanese are the same, not all share the same mindset, education and philosophies.

[mouzogu]

the director aoji aonuma has to be the oldest looking salaryman i've ever seen. i think that says a lot.

long hours, very high level of expectations, fine tuned attention to good gameplay design, and the protection of higher ups execs like miyamoto from bean counters.

[DirkH]

I don't think it is racist to highlight an aspect of a culture and how it might at a group dynamic level causally influence the outcome of something.

The claim that Japanese culture causally leads to better craftmanship might be wrong. As you've mentioned, there are plenty of counterexamples that argue against Japanese culture having the claimed causal influence. But this doesn't make the original claim racist nor even borderline racist.

[vincvinc]

In 2017, Breath of The Wild developers shared a lot of details of the process at the game dev conference CEDEC.

Nintendo made blogs take down the pictures of the slides afterwards, but here is a good summary:

https://www.youtube.com/watch?v=CZzcVs8tNfE

[JenBarb]

Sakurai (creator of the Kirby and Super Smash Bros. series) has been running a youtube channel about his game design/development philosophies.

https://www.youtube.com/@sora_sakurai_en

[luisgvv]

I second everyone to watch this channel, his insights on game design are deep and goes through detail explaining what is a good gaming experience

[nscalf]

It might be worth listening to the Acquired podcast episode on Nintendo. They are very far from perfect, and have had a good number of serious failures, along with some very strange decisions that clearly hurt them. Nintendo has a die hard stance on modding that is definitely net negative. Just a few weeks ago they went after some giant Twitch streamers for playing modded content. They also consistently ship technology that is generations behind.

One big thing they pointed out is the type of gaming they target. While the Playstation and Xbox general aim for very serious, high "skill" players, Nintendo often launches just above the seriousness and skill level of mobile gamers. It's easy for me to sit down with my extended family and play Mario Party or Mario Kart, but they'd hate me if I had them play Elden Ring. They also are strongly against much of the free to play content.

I left that episode questioning how much of Nintendo's recent success is due to them outcompeting versus the competition making a series of unforced errors.

[iamwil]

Yes, it's rare, but they have.

Game maker's toolkit has one on how they solved their open world problem for BotW:

- https://www.youtube.com/watch?v=CZzcVs8tNfE

And at GDC, they talked about their Chemistry engine for BotW:

- https://www.youtube.com/watch?v=QyMsF31NdNc&t=2354s

- https://www.youtube.com/watch?v=Bk5swSyJ5zQ

Game historian has some tidbits of how they made certain design decisions for their successful games

Mario Kart:

- https://www.youtube.com/watch?v=NDHZKYETDyU

Super Mario World:

- https://www.youtube.com/watch?v=s2bTQK6vbKI

Super Mario Bro 3:

- https://www.youtube.com/watch?v=MxT6IwUtLSU

GMT, Snowman, and Extra credit does analysis of how Nintendo designs their Mario levels.

- https://www.youtube.com/watch?v=dBmIkEvEBtA

- https://www.youtube.com/watch?v=z_KVEjhT4wQ

- https://www.youtube.com/watch?v=Vwj3On5o58U

- https://www.youtube.com/watch?v=3fI9pfDf60g

- https://www.youtube.com/watch?v=ZH2wGpEZVgE

Generally, it follows the ramp up and down of the Hero's journey in storytelling.

1. Introduce a new mechanic in a safe area where it's impossible to die.

2. Increase the challenge with the mechanic by adding variations

3. Ramp up challenge even more by combining with previously learned mechanic

4. Have periods of rest inbetween the challenges.

[peanuty1]

I think they have a couple advantages: 1) Zelda on Switch has very basic graphics compared to modern AAA titles on other consoles and only a 30 fps framerate. 2) The last Zelda game came out in 2017 so they've had tons of time.

[tompagenet2]

I know people say this, and of course the art style is stylised. But playing it last night and watching the day-night cycle, the grass moving around on the first sky island, the physics simulation as I dropped my clumsily-made creations and I thought it was pretty impressive. I do think it's a wonder this runs on a low power tablet computer from 2017 but if I open up Teams on a brand new machine it can be a laggy mess. I do think we have lost track of how much computing power we have and how poorly it is used.

[jeroenhd]

> I do think we have lost track of how much computing power we have and how poorly it is used.

I think this is the real lesson to take away from the Switch. The device is woefully underpowered, but developers know there's a huge market out there so they just have to Make It Work. The end result is that many games, especially first party ones, are super well optimized for their hardware. You could see this with the 3DS too, the things those 285MHz could pull off were definitely very impressive.

On PC and other amd64 platforms there's so much raw CPU and GPU compute available that it's possible to get away with performance impacts. Doom Eternal is one of the few well-optimized big games that just seems to play well on any device with a GPU you throw at it. Compare that to some recent releases and you really wonder how bad things must've gotten.

Of course, highly optimized game development takes time, effort, and skill, and that doesn't come cheap. As long as gamers accept the inefficiencies on other platforms, games will continue to be released in a subpar state. Nintendo cares more about the quality and reputation of their brand (in some areas) than it does about making money so it goes the extra mile; I doubt EA or Bethesda care as much as long as they keep making money.

[Cthulhu_]

See, pretty visuals / moods don't need high performance hardware, it's what you do with the tools given to you. I'm sure plenty of kids have just sat in Minecraft for a while watching their world go by, and it's using 1x1m blocks and 16x16 textures.

I wish there were better crossplatform native desktop development environments. Teams made / is making the switch to React at the moment, but it's still a web application.

I've recently gotten into the early access program for Beeper, which promises to be a native cross-channel chat app connecting things like Slack, Teams, Whatsapp etc into a native app. I like the native app part, but the downside of one-app-for-all is that it's lowest common denominator in terms of features and visually it looks like none of the other apps. Still uses 130-140 MB of memory at the moment though.

[turbo_fart]

2015 actually!

[BiteCode_dev]

It would be like figuring out what Steve Job, Usain Bolt, or Killian Jornet do. It would be interesting and helpful, but you will not be able to replicate it by following a recipe.

Being exceptional is, by definition, exceptional.

[HellDunkel]

I have tried to figure this out myself and found two facts that stood out: 1. On BOTW game designers did not allow polishing within two thirds of the game dev process 2. The executives do a lot of play testing.

To implement both at the same time is quite something if you ask me.

[pjmlp]

Easy, with proper algorithms and data structures thinking about a single kind of hardware, instead of developing on a octacore with 32 GB and SSD with a RTX GPU and then expecting everyone else has the same setup.

Basically by doing development like we used to do in the 8 and 16 bit home computer days.

[gowld]

Zelda is great because it doesn't rely "software development". It uses game design and art.