Salve One Postmortem
After 3 months developing Slave One, this project has been one of the most incredible experiences we had. The project helps all the team to grow and improve their own skills and teamwork. So, now that the project is finished, let's take a look back to these months of developing and take note of the best and worst of all of it.
What Went Right:
Keep the Star Wars feel:
The main challenge of the project! Star Wars has a lot of material produced previously for the game and one of the fans of the series have a clear vision of what Star Wars universe is like. We know that achieving the quality and feel of the series was not an easy task, but at the end we somehow achieve it.
At the start of the project, and until near the end of it, that was not the case. Even though constantly iterating with the art assets and the audio, some of the limitations we had during the first part of the development doesn’t allow the team to achieve the quality we needed to give the players the Star Wars space opera feeling.
Once the code team was able to develop better rendering effects such as the normal map tiling and the shadow mapping, and the audio wwise implementation issues were fixed, the overall experience improved and the game started having the feeling the players can experience now in the game.
Making an interesting 2nd player:
The most critical point of the entire project! The second player was a mess itself. This feature was one of the requested as the core at the start of the project, but the design team had a lot of trouble with it.
Giving a second player a good experience inside the Star Wars dogfight was a challenge since the first moment. We started developing different options for the second player such as slip screen and giving the second player the weapons control, giving him a different screen with the ship stats and handling the spaceship resources, being able to move inside the spacecraft in a 3rd person view…, was a moment that we were even thinking about a pizza delivery game and the second player being the one that prepares the pizzas! Some of the ideas just weren’t doable and some just make the first player boring. Others weren’t accepted by the producers.
In the end, we went for a management role but integrated inside the same screen trying to increase the communication between both players and integrating him into the action of the battle. To achieve that, the most important elements are the repair puzzle and the shield management, that allows the second player to feel useful while playing.
The technology development:
To be able to do the final product, be needed to develop all of the tools and features the engine has now. The challenge of creating the game itself was big enough and we had also the challenge to develop the engine and the subsystems.
Even though the code team has just 6 members, they were able to create all the tools the designers and artists needed and code the game logic for the game itself. Some of the subsystems had issues during the development, but alt the end all worked properly and the result is what can be experienced in the game.
What Went Wrong
The explosions:
The explosions are an important part of the experience, even though the tie fighter explosions somehow look like explosions, the other enemies explosions still don’t have the quality the rest of the assets have.
This issue was mainly due to the lack of experience in the team developing particles and the issues the art team not being able to create the assets for the particles sometimes due to an unstable engine build, others due to their lack of involvement.
In the end, some basic explosions were created for the turrets and the generators and the tie fighters explosion ended up being quite decent.
Meeting the goals of the milestone:
During half of the project, the team was not able to meet the goals for each of the milestones. This was the main problem that made the scope of the project reduce a lot. At the start of the project, the game was intended to have 3 levels and side selection to decide if the player wanted to play as rebel or empire.
But during the development of the game, we realized that that was too much for the team. The development times were larger than the expected ones, being the main time consuming feature the enemies AI and the collisions.
The development was slow to the point that we had to reduce the alpha development time and increase the vertical slices time to be able to test the features we wanted to implement in the game.
The team mood:
After the first deliveries, the team mood towards the project changed drastically. From an outside point of view, half of the team give the feeling that they were doing the project just an obligation and doesn’t feel motivated with the game at all.
During the project, some possible solutions were given to the members to try to solve this problem like increasing the group working time together and offering help and support if they need, but at the end, they didn’t come to the work sessions or ask for help/support.
A Final Word:
After doing the project, we all realized the difficulty of organizing the team even though it isn’t a large one. The own planning of the tasks and expected time they will take was one of the more important points to take into account while planning the sprints and the goals for the milestone. Taking care of team’s mood is a crucial factor that in our case was not well managed and that drastically affected the game quality.
Finally and one of the most important points, working with a well-known IP such as Star Wars is a bigger challenge than any of us thought before starting the project. Star Wars have a lot of information and resources available, but it also has a very specific flow and feeling that makes it’s universe unique and a challenge to reproduce.