BP lessons from creating EVE: Valkyrie

Andrew Willans has been working in the games industry for a long time. Before joining CCP Games to work on Eve: Valkyrie, he managed to participate in projects such as Watch Dogs and Grow Home .

At VRDC in San Francisco, he talked about the game, the launch of which on Oculus Rift, PSVR and HTC Vive taught him a lot.
')
Willans began work after creating a prototype of Valkyrie , but before the start of full-scale development, so his story may be useful to developers seeking to turn a VR prototype into a complete game.

Andrew compares the design of VR games with the design of an amusement park: in both cases, you need to create a “holistic fantasy”. As an example, he cites the attraction "The Haunted Mansion" from Disneyland, in which participants immediately immerse themselves in an alternative reality due to the architecture, environment design and costumes of the actors.

“When I started work, there were 28 people in the team, and there were never more than forty of us,” says Willans. Given the scale of the Valkyrie , it’s (relatively) small in size, but it guaranteed that the team would quickly adapt to new lessons and tasks. The most important discovery was that the size and scale of a VR game determine how well it can convey the feeling of flying on a fast space fighter.

“Fast flight weakens the sense of volume,” says Andrew, explaining why designers had to increase all objects in Valkyrie by 1.5 times the player’s scale. "We have done more and more so that the sense of volume has lasted longer."

The developers also decided to add small “speed particles”, small pieces of space debris that fly past the player but do not have a significant in-game effect - interaction with them is limited to pushing away from the windshield.

But the main theme of the report was not details of Willans. He called on fellow VR developers to think bigger, citing an example of one Valkyrie mission, which was supposed to end with the entry into battle of a large transport vessel shooting at its target and creating an impressive explosion. In the process of testing, the players constantly missed him, because no matter how great he was, participants could look the other way. No matter how large the ship is, in VR the ship is not always so big to guarantee the attention of the players.

“I learned an important thing about VR: feedback needs to be redundant,” says Willans. “After creating the scene, you should always experiment. This is the advantage of a small development team: after testing sessions, we could always organize daily meetings to discuss feedback. ”

(Later, Willans noticed that the Valkyrie development team had mandatory testing sessions at least three times a week. If there were empty chairs for testing, the developers left the office and urged people to participate - they knew that it was critically important.)

He also reminded other developers of the value of creating a diagram of what exactly the game should be. In the case of Valkyrie, the basic principle was that the player is a space pirate. “Such high-level schemes allow us to back up and test for compliance aspects of the game during the development process,” says Willans.

Of course, this does not mean that you can not deviate from the initial design plan, you just need to be sure that the development process always meets the high-level goals of the project.

For example, the Valkyrie team made great efforts to create a single-player campaign - it repeatedly repeated that the game would be multiplayer, but fans asked for the creation of a single player mode. Willans said that they were trying to do something intermediate by adapting multiplayer sessions to a single-player game, adding an additional layer of narrative and mechanic, for example, the waves of attacks of enemies controlled by AI.

They sought to create asynchronous multi-user VR-flights, but this idea was cut out as a result. The idea was to push the players with the recordings of the actions of their friends in different missions, but failed because the team did not have enough time for implementation.

“We can ever get back to it, the prototypes were very exciting,” says Willans, adding that some of the cut-out content was reused, for example, to create game tutorials.

At the beginning of the development, Valkyrie demanded much more grind: the player could have spent hours unlocking the key features of the game.

“At the alpha stage, we found that too much Grind was needed to get to the real game,” says Willans. "This is a really great lesson, I especially felt it in my work at Grow Home : if there is a lot of great content in your game, just give players the opportunity to enjoy it."

The same attention to accessibility can be attached to the UI VR games. Willans recalled that during the Valkyrie development process, part of the game was displayed as static text hanging in front of the player. The game took a big step forward when the team had the idea not just to hang objects in front of the player, but to project them, as if the player interacts with the real futuristic technology of the world.

“It helped to create the feeling that the player’s avatar is just sitting on the box, resting and waiting for the battle,” explains Andrew, talking about the interface that appears in front of the player when he is not on the ship. "He seems to be resting, looking at the futuristic analogue of a mobile phone, as we are doing now."

The team also came to the conclusion that it was necessary to take into account the visibility and distance of visibility that are comfortable for the player (see pictures below). In short, it is usually necessary that players look within 10 meters (but beyond half a meter) and about 94 degrees relative to the direction of gaze.


On the left - a comfortable visibility area, in the center - a comfortable visibility area within the head rotation, on the right - maximum visibility area within the head rotation


On the left - the minimum comfortable distance of visibility, in the center - clear stereoscopic depth perception, on the right - the limit of stereoscopic depth perception

The Valkyrie team had problems trying to squeeze even the simplest level selection screen into this comfortable field of view, because there were more levels than fit within optimal limits.

“Our UI at the moment is a compromise, a happy medium,” says Willans. “In the end, we stopped at a rotating wheel. On it, within the optimal distance of the cone of visibility, we placed the planets, the levels that can be chosen. ”

The development team had to go on a similar compromise in the interface of the selection of ships. when it was necessary to find a balance between the need to show all the ships and make the models impressive, while at the same time maintaining their distinct details.

All this leads to the fundamental truth of modern game design: most game developers present games on a 2D screen, in a non-3D space. According to Willans, it is difficult to get rid of such a habit, but it is necessary. It is necessary that the entire development team think in 3D.

“The habit of 2D thinking is hard to break,” says Andrew. “This is a regular joke in our office. We ask each other: “Did you check this function? And in the Rift checked? And in PSVR? ""

In conclusion, Willans briefly reduced the lessons learned in the development of VR to a simple aphorism: “Get the most detail, and then gradually roll back.”

“If you want to create wow moments, zoom out to increase accessibility and comfort. It is much easier to reduce, not scale. But do not forget to leave something amazing for the player at that moment when the VR wow-effect becomes dull, otherwise he will remove the helmet. ”

And, and more: “Texts look disgusting. I'm not a big fan of text in VR. ”