The amazing artist Matt Nava, who helped to build The Journey, talked about the development of his outstanding underwater adventure.
We’ve talked with Matt Nava from Giant Squid about the production of ABZÛ. We’ve discussed the choice of art style, the peculiar challenges of underwater gameplay and work with environment production.
Intro
Giant Squid is a small game development studio led by Matt Nava. Nava was the art director of the games Flower and Journey when he worked at thatgamecompany previously. The team is comprised of developers from across the industry, with experience working at Epic Games, Big Red Button, and Sony Santa Monica among others.
The team grew to 10 devs during the development of ABZÛ, and is comprised mainly of programmers who specialize in different aspects of the game creation process, from graphics to gameplay. We also have an environment artist and a technical artist who helps us bring animate characters. Matt serves as our creative director and art director.
The score ABZÛ, Giant Squid’s first project, was composed by Austin Wintory, who also wrote the music for Journey and The Banner Saga.
The Waters of ABZÛ
After developing a sandy desert for three years during the development of Journey, Matt was eager to create a world that was vibrant and full of life. The ocean setting in ABZÛ was perfect for that. Matt was inspired by his experiences scuba diving off the California coast, where he had a very memorable and friendly encounter with a wild sea lion. We also saw an opportunity, as there’s a big lack of games tackling the ocean with an artistic approach.
Art
The light, color, and sound of an environment to deliver an impactful narrative to the player. The ocean turned out to be the perfect setting for this kind of storytelling, as even simply adjusting the distance you can see into the murk can have a huge effect on the player’s emotional state. Through carefully crafting all the art and environments in ABZÛ, we created an emotional curve, with lows and highs, that delivers a story without the use of any dialogue.
The goal early on was not to simulate scuba diving, but to communicate the majestic “dream” of the ocean. That philosophy carried into the art style, which we tried to use to create the most iconic, memorable representation of fish, corals and other elements of the sea.
Visual noise versus clarity was a relationship that we had to balance carefully. In many areas of the game, there are an overwhelming amount of things – fish, kelp leaves, ambient particles, seagrasses, corals. Making sure that the overall image was still readable was an important factor even when designing the smallest elements of the environment. The stylized look was both aesthetic and also very functional in keeping the image parsable.
The hundreds of sea species in ABZÛ are all based on real creatures from oceans around the world. However, when we studied them, we realized that each real-life fish is infinitely complex in the way it moves, acts, and looks. We very intentionally chose what we felt were the most iconic elements of each species and focused our design on those, the outcome being a stylized representation of the actual thing. This approach worked very well in that it help us convey the essence of the creature, while simplifying less important details and reduced the noise.
Controls
One of our biggest challenges designing ABZÛ was creating a control scheme that captured the fluid and weightless sensation of swimming in the ocean. We again took inspiration from scuba diving, where neutral buoyancy is key to navigation – when at rest, the diver neither rises nor falls. This is actually quite different than many game swimming controls. We also knew that we wanted to create the sense of underwater ballet, and let the player do complete flips. Many games stop the player from pitching when they are vertical, so the the camera doesn’t turn upside down. We managed to solve these camera problems and ensure that the player has an unprecedented level of swimming fluidity while never letting the camera flip. Its a difficult and rarely approached problem, and I am proud to say that we were able to tackle it.
Choosing Game Engine
At the time that we started development on ABZÛ, the team was very small, it was just Matt. We need to hit the ground running with tech and tools that could support his vision for a dense, vivid world. Unreal Engine 4 was brand new at the time and after vetting other software we decided that we would go with it. Unreal has many powerful tools that we were able to take advantage of, like advanced material creation and rendering capability. However, the best thing was that we were able to extend the engine to create bespoke features specifically suited for rendering multitudes of fish, volumetric undersea murk and lightrays, procedural kelp simulations, among other things.
Animation
Our team had no dedicated animator, so we needed to be very smart about how we would bring the world of ABZÛ to life. We took several different approaches, ranging from traditional skeletal animation running through a complex state machine for animating the diver, to completely procedural motion based on rope simulations and tidal forces for kelp forests and foliage. Most fish in the game were animated using mathematical motion in their materials alongside morph target posing, which freed them from needing expensive skeletal rigs. This was key to letting us render schools of tens of thousands of fish dynamically.
Environment Production
Creating the environments of ABZÛ was one of the most challenging aspects of the game. For a long time we struggled with our pipeline; it was very difficult to make changes to areas once they had been blocked in. We found that making edits to the terrain and rock structures of the world was so important that we dedicated a lot of our dev cycles to creating custom tools for composing terrain out of easily malleable smaller patches. Once we had a pipeline that allowed us to iterate quickly, we were able to focus on making sure that the environments guided the player through composition, color, and detail placement. We tested the environments with players to make sure that they felt like they were exploring but never felt too lost. We relied heavily on lighting, atmospheric density, and color to convey the mood and emotion of the narrative through the environment.
Difficulties
Giant Squid faced several formidable challenges during the development of ABZÛ. It was the first game from a new studio; we had to hire a new team while establishing a production pipeline and tools. At the same time, we were making a very unique game that involved solving difficult and rarely approached challenges, from completely freeform underwater controls to underwater lighting mathematics and fish behavior simulation. We managed to solve all these problems and ship a game purely on the merit of the strength and grit of the team that we were able to bring together. Our devs are dedicated and passionate about making games that push boundaries and create new experiences for players. It was their determination that overcame our challenges.
When it comes to a budget you always wish you had more, however, we are quite proud of the quality bar that we were able to hit with a budget that is quite small by game industry standards.