Celeste is a tight, tough and tantalisingly precise platformer from the creator of the TowerFall series. Starting life as a game jam project in a 'fantasy console' called PICO-8, this new slice of indie challenge has blossomed into a beautifully crafted old school-style title. With its Nintendo Switch release date imminent, we chatted to one of its creators about the journey to Nintendo Switch, the games that shaped Celeste and what it takes to build an indie title in 2018.
Nintendo Life: Could you introduce yourself?
Noel Berry: Hey! I'm Noel Berry, and along with Matt Thorson and many other very good friends, we created Celeste.
When did the development of Celeste start?
We created the original Celeste in August of 2015, but started on the new one in January of 2016.
When was the decision made to bring it to Nintendo Switch?
There was a time when we were far more optimistic about how fast we could get Celeste done, and initially thought we could be a launch title for the Switch. Both Matt and I grew up playing Nintendo games, so being able to release on a Nintendo console was a really exciting prospect. But we missed that by ... nine months, and wrapped up the game at the end of 2017 instead. Better late than never!
We actually have experience of the game's original engine PICO-8, a 'fantasy console' and its ecosystem. In fact, We've visited the creator and his cafe in Tokyo many times.
That's awesome, I'm hoping to someday make a trip there and visit for myself!
How did you discover PICO-8? What was your experience in game development prior to using it?
I had been following Zep, the creator of PICO-8, for a little while due to his work on Voxatron, and at some point PICO-8 showed up in my Twitter feed. Once I had some free time I decided to dive into it a little bit. Before PICO-8, I think pretty much everyone on the team had been making games for several years. I started out with Game Maker as a teenager, and later worked in Flash and then C#.
How did you find working with it?
It's awesome! I highly recommend PICO-8 to... basically everyone! It's accessible and has a strong vision. It does come with some fairly big limitations, but I often feel constraints are great for experimentation.
Which engine did you choose to develop Celeste in after PICO-8?
The newer Celeste was created in C# and XNA, a now deprecated framework originally created for Xbox 360 games. We use various open source ports that are highly maintained to get the game out on other platforms.
How was the transition from moving the game over to a format that was supported by consoles?
The new Celeste took us around two years to create... so going from the PICO-8 game to the full game people will be playing on consoles was a lot of work. I did recreate the original PICO-8 game inside our new one (as an unlockable), and porting that over was only about a day of work.
Is it true the initial concept was realised within four days from a game jam?
Yes, the original PICO-8 game was created in four days.
Could you explain the game's origins?
Matt and I wanted to create a short game in PICO-8 over a weekend. We wanted our concept to be fairly minimalist (partially due to the time constraints we set on ourselves and the limitations of the tool), but that also had a lot of depth in mechanics. The idea of a character struggling to climb a mountain felt like it fit really well with this.
How did the ideas in the game grow from this initial version?
The game stayed very true to the vision of the PICO-8 version, but it was able to explore the space, story, and characters a lot more. Our plan was to take our initial jam game and see where it would go if we had more time and a larger canvas to work with, and it kept surprising us.
Was there anything you experimented with but ultimately left out?
Yes, basically every chapter in the game had ideas, mechanics, art, and levels that were eventually thrown out. I think when making a game you have to be alright with throwing things out, especially when they don't really end up matching the vision. It's not lost work or time, it's part of the process of coming to the final realisation of the game.
It's fair to say that Celeste is a departure from your previous work, such as TowerFall?
Yes, though I think a lot of the lessons from TowerFall make their way into Celeste.
Was this a conscious decision or did the game develop organically?
I think it happened fairly organically. Matt and I had approached Celeste from the beginning as a single-player adventure, and it grew from there.
What do you think about the game being mechanically compared to Super Meat Boy?
On the surface I think the games are similar - both are difficult indie platformers. However, I think at a closer look they approach that concept in very different directions, thematically and mechanically.
Was there a process behind balancing the difficulty?
Yes: Making our friends come over and play very rough segments of the game, over and over. And over. We also had a general philosophy that, though we want the game to feel hard, we always want it to feel fair. The game works in the player's favor, wherever possible, and when it felt like a level or mechanic was hard in the 'wrong' way, it was cut or modified.
What was your interest and exposure to rumors of NX?
We've all been pretty big Nintendo fans our entire lives so.... we were all pretty excited for whatever Nintendo's new console was. I don't really recall what specific rumors we had heard at the time.
What do you think of the Switch hardware?
What influenced the design of Celeste?
What Nintendo games did you play growing up?
What does the future hold for Celeste?
Hopefully, a game that people remember and come back to fondly. I don't think we have any plans to create a sequel to Celeste - it really feels like we told the story we wanted to tell.
We would like to thank Noel, Matt and the team for their time. Celeste hits the Switch eShop on 25th January for $19.99.