Playtesting Rads & Relics

This week we’ve been doing playtesting for Rads & Relics. Up until now, we’ve been building out the core game engine, the supporting UI, and other essential systems, so this was our first time really sitting down to focus on the gameplay. The game was playable, but, if we’re being honest, it wasn’t a lot of fun: the map was too big, the enemies had too much health, and basic actions required too many clicks. We set up a very bare bones process for playtesting as follows:

  1. Start a new round of the game
  2. While playing, record any thoughts/observations in a document [1].
  3. After playing we review the notes and try to synthesize any actions
  4. Implement the changes
  5. Rinse and repeat

This may seem simple, but having any kind of structure has been useful as we think about what’s important in the game. Already, we’ve uncovered crashes, bugs, UI/UX issues, balance issues, and other gameplay improvements.

A Playtesting Example

Here’s an example of one of our playtest notes and changes we’ve made

  • It’s annoying that you can’t mute/change the audio from the game screen
  • You can’t get back out to the main menu from the game screens

This one is simple: we added a pause menu with both the ability to mute the audio and get back to the main screen.

  • I wish there was more visual distinction between characters

We only have one sprite for our player characters right now, so while we’re working on more sprites we’ve added a color overlay, so you have a better sense of who is who.

  • The first turn feels lame. All you can do is run. You can’t even make any shots

Now, this is one that has several different options available. We could make the map smaller, we could make your characters faster, we could increase your range, or we could add abilities that are relevant even if you can’t get in range. We’ve decided to make the map smaller to make the combat feel tighter.

  • I wish there were keyboard shortcuts for abilities
  • Tab can switch to characters who can’t act

Two UX issues. Tab will now only switch to characters with action points left and we added keyboard shortcuts for each ability.

  • Enemies have too much health. After you’ve set up there’s really not a reason to move.

We’ve significantly tuned down both player and enemy health so they die in 1 - 2 shots.

  • I wish the characters had something that set them apart

This is another big one. In this version of our game, each character had the exact same set of abilities. To address this, we’ve switched to a mode where each character has a unique set of abilities that will have a different effect on the battle.

  • Cover feels pointless

This variant of the game had a cover mechanic. We’ve temporarily shelved this until we find a better way to integrate it.

  • CRASH: When a character died, the game crashed.

At this point, the game crashed. We were able to fix the crash afterward, but since we already had a lot of feedback to work with, we ended the play test there.

This week we’ve only been able to do a handful of playtests. This is because each play test has exposed myriad issues, so we’re spending more time fixing issues rather than playing. But we’re starting to tighten the loop as we get some of the quick wins out of the way.

State of the Game

  • Progress this week
    • UX Improvements: New pause menu, visual differentiation for characters, keybindings for abilities
    • Balance Changes: Adjusted health values and map size for quicker fights
    • Enemy AI Improvements for better targeting
    • Bug fixes
      • Fixed a crash when a character dies
      • Fixed an issue where static objects like rocks were moveable
  • Coming up
    • A theme that has come up across playtests is the monotony of turns. Once you’ve set up your position you just hit the enemy until one of you dies. You use the same abilities again and again and it’s not very fun. We’re going to look into different ways of addressing this. I am partial to the idea of abilities with cooldowns or some other randomized way of handling abilities, but we’re going to see what feels best.

[1] I had the idea to automate some of this with AI so have humorously titled these documents Game Play Tests or GPTs. I haven’t actually been able to finish the AI automation part though.