This document is a braindump of everything that I believe is stopping Overcooked! 2 from being the best game ever made. Please consider all of the following if you are thinking about making a spiritual successor.
When a run goes well it goes really well, when a run goes bad, it goes really bad. This is because of a feedback loop designed into the game which amplifies both failure (and therefore success as well). These are the components of that feedback loop:
- Fire ruins food and takes burner/oven out of service and requires effort to be put out
- Missed ticket deducts 40 points from current score (why so many points? the player is already loosing by being behind on score potential)
- Missed ticket resets the tip combo multiplier, hindering future earning potential
- The missed ticket is still on the order queue and still must be [re]cooked
- When you finally serve the missed order, the game prioritizes the ticket with the least time remaining which means that the dish will almost always serve out of order, thus breaking the tip combo a second time. Since you're behind on tickets but still need to serve the same dishes, you end up in a cascade of missed orders
- When things derail like this, pots/pans have to come off burners to prevent burning. This adds extra downtime when you eventually have to set them back
It almost feels like two separate games. One where you're on the good side of the feedback cycle, and one where you're not. I believe the best implementation of the Overcooked format would include some form of scaling difficulty to push all players to their limit, not just beginners (e.g. think tetris). Without addressing the above, failure is much more binary than it has to be. The best moments in Overcooked exist in the gray area where the game is still recoverable, and by addressing the intensity of the failure feedback loop described above, you create the potential for a much wider gray area between success and failure.
- Tip are awarded in arbitrary discrete increments so if you play a level many times slightly improving each time your score goes up in "steps" rather than a line. Tips should be awarded by % of remaining ticket time to more accurately represent how "good" the player is doing rather than these arbitrary thresholds.
- On some levels the 3-4 player score thresholds start to hit a skill ceiling because of poor level design. For example a level with only 2 plates means you can only serve 2 dishes every 15 seconds regardless of how fast you're cooking. Other level design factors hard-limit the kitchens theoretical max output such as number of burners and timed level platforms/obstacles.
- In general, the plate respawn time and/or the number of plates available is too restrictive
- The star quotas are inconsistent in difficulty. It appears the staff in charge of setting them got worse at the game over time, not better.
- Some levels last way too long for how simple they are. The ideal level duration is enough that the average player "just" starts getting into a comfortable groove before it finishes (look at Story Kevin-1, it's 4 minutes...)
- The game enforces that there is always at least 2 orders on screen to prevent fast playing teams from having to pause because they don't know what to cook. In practice this needs to be at least 3 (and the ticket refill time should be faster on average). You should NEVER punish the player for playing your game too well. Having future ticket information hidden rewards strange meta-game strategies like counting recipes for RNG prediction.
- The pseudo grab-bag RNG for tickets is confusing and it's probably better to just be 100% transparent with a true grab-bag randomizer algorithm. See tetris for an example of what this is
- An option for seeded RNG for tickets could open up the potential for a new kind of competitive scene
- Horde levels are one of my favorite parts of the game but they fall off in replayability because there's nothing to grind for once you've survived with 100% health. An "endless" horde mode with infinitely scaling difficulty would be incredibly addicting
- Rebind controls to allow for actions to be split up (e.g. use different buttons for chop and throw)
- Allow players to choose where they spawn in a level before it begins
- See each level's high score separately for all possible player counts
- See each multiplayer player's high score for a level instead of just the host's
- Show the previous high score on the results screen when it is beaten
- Remember/favorite your last picked chef model when launching the game
- Remember/favorite your last DLC main menu background when launching the game
- You should walk slowly when using bellows/squirt gun instead of coming to a full stop
- Anti-cheat
- Button to drop backpack
- Fix a bunch of bugs
- Serve recipes in the order the cards appeared, not picking the one with the minimum time remaining
- Fix inconsistencies with how cook/chop/mix progress is recalculated whenever an ingredient is added
- Skip tutorial messaging
- Improve server tickrate from 10Hz to 60Hz (10 is absolutely unheard of for an action game)
take these less seriously than the above
- Online leaderboard
- Level Editor
- Speedrun mode, where instead of trying to get the highest in a set amount of time, you have to serve a set number of tickets in the shortest amount of time possible. Note that a target number of tickets is much more interesting than a target score.
- Jump button
- 8-player coop
Asymmetric multiplayer where one or more players are playing "overcooked" and the others are playing "action-arcade diner dash"
At one point I thought about taking the best of both games by:
- Adding a "day" system to overcooked
- Scale the complexity of recipes over time
- Scale the impatience of customers over time
- Making the overcooked randomizer items into "cards" and adding more