Seals of Fate is a heroic, cooperative ARPG with a roguelike emphasis on mastery. Fight together with up to 2 friends as you play the descendants of a pantheon long departed, as an ancient evil breaks free from the prison once forged by your ancestors.
Travel the seven realms, master your abilities and choose how you develop your skills as your band of heroes fights, fails, and tries again. Each character has a unique set of abilities and talents, and each player has a role to play, giving real strategic depth to your fights against the minions of darkness.
Coordinate your skills to overcome the odds and save the universe!
​
Date: March 2022
Team Size: 4+
Role: Product Owner
Engine: Unreal 4
Seals of Fate was designed with a desire to create a cooperative focused hardcore experience. I was inspired by the Diablo franchise, Risk of Rain 2, and Path of Exile. In Diablo and PoE, players can group together, but the optimal playstyle is mostly solo play. There aren't a lot of direct support options in those games, and as a support player, I wanted to ensure that was present in Seals.
​
A few features include:
-
Networked multiplayer
-
Unorthodox character designs and talent trees
-
Dynamic monster spawning system
-
Monster abilities that require teamwork
-
Three phase boss fight
Character Designs
Kamilla
The Gravityblade
Low complexity
Low range, single target damage dealer
or
Medium range area controller
Kamilla's Bladework tree focuses on mobility and high damage abilities like Lunge and Execute combined with the ability to critically strike. Lunge is a dash that stabs an enemy and Execute attempts to finish off wounded enemies, increasing it's critical chance when enemies are low health. Upgrading Lunge with the Bladesurge talent resets it's cooldown on a kill, so Kamilla can use it to execute many low health enemies in a row. She moves around the edges of the battlefield, finding isolated targets to strike and then retreat to safety before she gains enemy attention.
​
Kamilla's Gravitation tree forgoes damage for crowd control. She can reduce the number of active enemies with Lockdown, applying a long stun on an enemy while her allies take care of the others. She can create a Gravity Well that vacuums enemies in and follow it up with Black Hole to stun them all. When one of her allies knocks an enemy into the air, she can use Untether to release the target from the bonds of gravity, causing it to float away indefinitely.
Postmortem
​
Goals:
-
Create a game that heavily encourages and rewards teamwork
-
Create a highly replayable experience
-
Create memorable boss encounter
​
Observations:
-
We needed more time to work in our ally assist mechanic to monsters. There are encounters that are made easier with teamwork but some don't need it at all. It was a good base though.
-
Not all talent trees were created equal. As we progressed on characters, we became much better at generating interesting talents. I would've liked to revisit Kamilla's tree.
-
The boss encounter was pretty difficult compared to the non-boss fights. Players were unprepared the first few times they faced it. Since the game has permadeath, I'm not sure if that's necessarily a bad thing. Players are meant to figure things out over time and not beat the boss instantly. As playtesters continued, they learned not to go in guns blazing and be more methodical. They still died a lot.
​
What went wrong:
-
I overscoped. Again.
-
I like big games and I cannot lie. I struggled to find other like minded individuals to collaborate on the project, so I ended up doing a vast majority of it myself. ​When I was finally able to find help, I had to spend a lot of time creating documentation and then handling integration of their work. Blueprints and git do not work well together.
-
-
Unforeseen deadline caused a lot of content to be cut​ and others rushed
-
I wasn't able to spend enough time polishing what we had. SFX, VFX, and level design suffered. We also had to cut the 4th playable character as well as a number of other features.​
-
-
Multiplayer is hard
-
You have to do so much more testing and need regular playtesters to be available at the same times to be able to ensure the game is quality. My next project is going to be single player, I swear. ​
-
​
What went right:
-
Designed interesting characters.
-
I really liked the talent trees that we ended up with. Because I had built off of a base, we were limited with how wide and long each tree could be, which made us take only the very best ideas for a character. We probably created over 50 talents for each individual tree and pruned them down before we settled on the final 20.
-
-
The team we had was really great
-
Our concept artist provided a multitude of sketches that really helped us come up with interesting designs for playable characters, NPC's, and our boss. Our designers created a massive amount of talents and balanced player and monster stats, and our programmers did a fantastic job implementing enemy and player abilities. ​​
-
-
Monster spawning and AI
-
I was able to create a really interesting spawning system that would allow a designer to choose a pool of monsters and assign a point value to the spawner. At runtime, the spawner would take into account the number of players and their level, randomly spawning monsters from the pool in a level range around the players.
-
-
Replayability
-
The level had 3 different starting points, and each spawner on the path to the boss had different monsters in its pool, so each encounter was a little different. Two large enemies is a significantly different encounter compared to 5 small enemies. ​You never know exactly what's going to be around the next corner.
-