TBF #14 – Turbo Turn Part 1
This blog post is a two-part series about the design and development of what we call Turbo Turn.
Let Me Play Faster!
Hello everyone. I hope you are doing well this week!
One of the biggest pieces of feedback we've received during the prototype build is the necessity of being able to play faster. You could see multiple people actively spamming the click button to start the next interaction immediately after the first interaction was complete.
We knew, from day one, that this was something that could've happened. But there are multiple reasons why we hadn't tackled it yet, and why we're tackling it now. In this blog post, I want to explain the problem, the possible solutions, and what we picked and why.
This post will be more oriented towards the design, its constraints and necessities, while the next blog post will be all about how we implemented this into the game.
Without further ado, let's get started!
The Issue
Ideally, the game would allow the player to click and instantly reveal a card. This would give immediate feedback. You click: you see the card turning. The first-ever version that we implemented in March was exactly like that. Why is it, then, that we didn't ship this version?
The main culprit is animation. We have a lot of animations, from small to big. They can indicate an intent, give some feedback, or carry out an action. These are put in there to give the player the maximum amount of information in the most digestible way. That's not all: these animations sometimes carry out an action that's detrimental to the game state.
For example, The Shark moves by one after you flip two cards. If you were able to click on the card as it moves, it would do something extremely weird, where the card would jump and fly mid-air. This isn't too bad, as the card would still be visible when it lands, even if it creates some clipping in the cards.
But you can have way worse. For example, what happens if you click a card that's going to be revealed by the combination of The Treasure and The Boat? Under normal game rules, this would result in either a panic or a failure state that would propagate, failing your combo or allowing you to have three cards open at once.
Some combos are even worse, as they can cause a chain reaction. The Dynamite can create such a case. The Shark can move The Dynamite: this could potentially kill The Barrel, and that would kill a neighbouring card and its pair.
If the player could click and reveal a card on the spot, without waiting for animations to complete, such a sequence could be broken at multiple places or moments. The solution we took for the prototype was to let the animation finish, and then allow the player to select a new card to reveal.
The reasoning was simple: we would wait and see if playtesters liked the game enough. Only then would we spend the time and effort to make such interactions better.
Since the prototype went well, we decided to fully commit and fix this issue first!
The Solutions
The problem is: "I want to play faster, because I know what the set of possibilities is and/or I've resolved it mentally."
There are a couple of solutions possible:
- We allow player clicks to go through, and the card would reveal itself, possibly undoing or modifying the game state in a manner that would still resolve correctly.
- We allow player clicks, but that cancels the animation.
- We accelerate all animations to be faster when we detect repeated clicks.
- We guess what the player's intent is, and derive from that what the next possible game state should be.
We initially thought that solution #1 was interesting. The biggest issue with it was that it's a complete nightmare to resolve each possible case, one by one. This becomes exponentially difficult with each new card introduced. So, after a while, we discarded it.
Solution #2 was proposed to me by some fellow game developers during our trip with La Caravane. It seems promising! Players are spamming clicks to get to the next resolution. Simply cancel the animation and move on. Sadly, it doesn't work that way. The game requires your full attention and understanding of what's happening at all times. Cancelling an animation could potentially throw a wrench in your understanding, leaving you guessing what happened. That could lead to a disastrous end on the player's side without them noticing, and just thinking that they're bad at the game. So, we discarded it.
Solution #3 seems good. That is, until you realize that it just delays the problem. If you have to spam-click to have faster animations… just make faster animations. It doesn't actually solve the issue we're facing. And so, we are left with Solution #4, which is what we picked.
We call it internally Turbo Turn, taking the name inspiration from Vampire Crawler!
What is Turbo Turn?
The idea is simple: what we need to do is let the player know that their intent has been registered, and then resolve the next possible state as soon as possible.
In practice, it means acknowledging a player's click and buffering it. When the animations are done playing, we can then automatically reveal the set of cards the player already picked. This way, players keep full knowledge of the game state, understand what's happening, and can play in advance if they know what to pick. Well, problem solved, right?
That's not so easy. Issues can still happen when a chained combination arrives and reveals that one of the cards you picked isn't what you wanted. In that case, you should be able to undo the selection you've made, otherwise it would build frustration, as you'd be unable to act on newly acquired information. At the same time, it's possible that you already picked more than one card, but we, the game developers, should only derive from your intent that you plan on undoing a subset of the combination.
Another case: a chained combination could also undo a subset or the totality of what you've picked in advance (a selected card gets killed). That should be communicated clearly to the player.
That's Turbo Turn: it allows you to play faster, and should correctly derive your intention from the game state. That will be coupled with better and faster animations as well! When making a turn-based game, the animations are always too slow at first. As we have more playtest rounds and see the reactions, we're better equipped to make animations that are faster and more enjoyable.
Next week, we'll discuss how we did the first implementation of Turbo Turn, which will need playtesting! Follow along to get notified and maybe try the next playtest with us!
