Post

Game designers! Tell me about a feature / mechanic / thing you added to your game... but then you later removed it and that made the game better!

25 reposts150 likes

In my anomaly finder game I had the ability to inspect objects (pickup and rotate). I thought I could add more subtle changes that way. Wasn’t a good idea. It was turning into “inspect every item to find the one I care about.” Was boring so I cut it. store.steampowered.com/app/4405240

store.steampowered.comThe Door at the End of the Hall on SteamTrapped in a looping hallway, you must study subtle changes to choose the right doors. Each passage resets the hall, but slightly wrong. Observe carefully, make smart choices, and uncover the path tha...

In TANUKI: Pon’s Summer we’ve had lots! I think the biggest one though was we used to make it so your Delivery Packages fell out of your basket if you crashed, then had to pick them up. Well, in a Tony Hawk style experience it’s safe to say you crash a lot. Fun idea, no good in context of our game

Exactly haha. It was in for so long as well, felt sacrilegious when we took it out. We also had a time system similar to the Bomber’s Notebook that was in for ages. The reality is no idea is sacred and if it’s not really serving the core always feel safe to experiment and change.

For Planted! one aspect of growing produce was that it could 'go bad' if left in the ground for too long (arcade speed). However new strategies around protecting/keeping some back for later challenges meant it didn't make sense to punish the player. Also didn't look aesthetically pleasing either!!

For my medieval library management game, I had a complex system for the state of the book. Spine, cover and pages each tracked how broken and how dirty they were. It was kinda cool but impossible to communicate and not actually necessary for the restoration gameplay (which is still WIP)

Monuments to Guilt had a little museum radio doo dad for additional context to the exhibit but as I landed on the final tone of the thing it just didn't fit to have my goofy voice bothering you with inane insight you could gleam just from looking around.

In my puzzle game, I added some rays of light and a splash of particles every time you placed a light bubble. Now the rays only appear when you place a light bubble that satisfies adjacent clues (feels more rewarding). The particles have been removed completely.

I intended for Canto I (short platformer) to have the wall jump. It had two problems: first, I couldn't find an implementation that actually satisfied me, and second, it kinda closed off level design instead of broadening it? My favorite movement options for the game were double jump and dash (cont)

and the existence of wall jumping made them situational. To limit the usefulness of wall jumping I had to force levels to not have many surfaces and that made levels look ugly and empty. I could've solved many of these problems by adding *more* stuff (non-wall-jumpable-walls for example), (cont)

but in the end, for a game that ended up having four levels total, I figured that removing the mechanic would be a better solution rather than overcomplicating the game, the design and my life lmao. It was supposed to be a quick learning experience after all (it took me 3 years, dangit Uni)

In Russian Subway Dogs when you set fire to a bear, we experimented with the flames making the floor under it dangerous, including a new floor fire graphical effect. But it didn't create a fun obstacle, and it wasn't super clear why you lost health when you ran past. So we removed that feature.

We also experimented with snapping the bounce direction when you bark deflect vodka and other objects to just 8 directions. Thinking it might be easier for people to aim where deflected objects would land, but that ended up actually making it MUCH harder and less predictable.

Lorelei and the laser eyes: Shooting Dashing Dual Character mechanics/multiplayer Novel camera based mechanics Etc

The radio in OXENFREE was once a tape player. Record any sound and play it back somewhere else to tap into the supernatural. Radio ended up being streamlined and way more on theme!

In my survival horror TTRPG, The Decay, I have a pacing mechanic call the Decay Level. When dice rolls go bad the DL rises, an event happens, and the higher the level the worse the events get. I had a mechanic where players could Act Brashly to get advantage on a die roll in exchange for

raising the Decay Level equal to the die roll. It ended up short circuiting the death spiral mechanic that the game operates on and basically made it so players could succeed almost all the time. Cutting the mechanic has made the pacing of the game smoother and aligned with how I want it to run.

I made a game about drawing paths in a grid to score points. Added a bunch of ‘special line’ types that would give curses or extra score when a line was used. It meant extra complexity, but players didn’t feel like it added any depth or fun to the game. Ended up just keeping one of the types

in Into the Slimy Mines, power was originally range-based, so you had to create a network of relays to make sure stuff deeper in the mine had juice. But in practice this was very easy to set up and your network was rarely at risk, so it didn't add anything at all to your planning.

I've removed the sprint button from our next game! We've had it in our previous two games, but this time I wanted to instead orchestrate when the player character walks/jogs depending on the context of the scene.

I don't mind you asking at all! Yup, the game is narrative based, and I've found on the other games that people tend to either run everywhere, so they have to constantly hold down the sprint button, or some people miss the sprint button tutorial and they walk everywhere.

In Match Tree I added the ability to choose tools before you start a map so the replayability is higher. And I removed it and enabled all unlocked tools which made playing the maps much more interesting. Now I just realized I didn’t remove the screenshot… store.steampowered.com/app/3010870/...

store.steampowered.comMatch Tree on SteamMatch Tree is a breathtaking puzzle experience (...it's actually a match 3 game). Embark on an epic journey with pixie, gnome and witch full of adventure, mystery (huh?) and sheep. Become the master o...

Originally in GRIDHACK, as it's a picross battler, I gave the player an "Attack" button. But having to press attack constantly totally wrecked the flow - once I got rid of it and made attacking automatic on line solve, the game clicked.

Key Fairy used to have a "Calm" system, where monsters would reset when you died, and you would have to find, and unlock, an Ancient Node to permanently calm the area. We ended up benching the feature because (among other things) it made the creatures feel like "Enemies", rather than people.

Elder Scrolls-style skill leveling. Yes, it makes a lot of sense to get better at something by doing it. But many skills are more situational, so they won't naturally keep up with your other skills. I ended up with a global XP pool. Every time you level up, you pick which skill to increase.

I'm currently running an ongoing TTRPG campaign wherein each individual adventure has gameplay that fits their respective story and themes, so they each have a distinct feel to them (if I'm doing my job right), and mechanic that support this. The current adventure among other things discusses--

--themes around parasocial/exploitative relationships vs genuine and mutual care, and so I initially added a sort of friendship mechanic inspired by Nectar from the Hades games. During the main mission, players could find and gather a resource called God Favor that they could either use to heal--

I wrote code to write a puzzle's entire possibility space to file so that a hint system could do a simple lookup to get the next best move the player can take. It's a really neat idea, but...

My computer would run out of memory before it could finish simulating the most complex puzzles in the game. I ditched the system in favor of hand-authored hints, which was more work but results in a more designer-controlled experience, which I prefer.

I didn't remove it, but I regret adding a copy/past feature to my hacking game. It makes the game too easy/short compared to what it was initially when releasing

(And to be fair, less interesing/immersive than what I'd like. But well, that's how it is now 😄)

Where to start? 👀 The biochips in Rogue Trooper on PS2 originally had a minimum of 4 abilities each, which led to nested menus to select the one you want. Even as devs we constantly selected the wrong top level menu, so streamlined it down to 4 total abilities to remove friction.

I created a digital version of a board game prototype I'm working on (for play test / prez). Bees 🐝 collecting pollen on a tile-made board. Originally, there was a "bridge" mechanic : limited number, use the action of your turn to move one bridge and use it to go to other tile. Not natural/easy. 🚮😁👍🏻

New rule : players can only make a 1-tile move per turn Why it works : they must rely on the tiles effects to move further... And tiles effects is the core mechanic of the game. So...👌🏻 😏 😁

Originally my stock trading game had a main mechanic where you’d use a newspaper to make investments ahead of time. It ended up making the game extremely predictable, so I ended up going the almost complete opposite direction of almost never predictable and it’s been much more fun since!

For my anti-GenAI roguelike Art Attack: the combat has you fighting by spreading paint on the ground. I wanted the paints to mix on the ground, but the shaders were too complex, so we had the player mix paints in the inventory instead. This way, each color could do something specific.

Tried to make my deckbuilder a roguelike: added the shop, amulets, that kind of stuff... And it just wasn't fun! And core mechanic felt mudded after all additions. Switched to a regular deckbuilder, made it a narrative one. Now the gameplay is much better: crystal clear, straightforward, fun.

People kept asking for undo on death in Desktop Dungeons, but it would stretch out failure from a sharp learning moment to a boring grey trudge towards an underpowered quit after undoing constantly. Took a decade to deliver a version of undo that didn't make the game worse.

Usually I kill unviable ideas before implementing them, but the biggest one I can think of is taking cards out of the game since the input system made them redundant

TTRPG, but: GUTGUN had an aiming mechanic in early drafts, with a dice roll to hit your shots. You'd gain an increasingly large modifier to hit rolls as you progressed, until hits were guaranteed. This was to give the feeling of you growing in skill the way you would in a boomer shooter...but...

it slowed the game down tremendously. The most important feeling to translate for GUTGUN was the SPEED of a boomer shooter, and hit rolls just bogged it down too much-- and playtesters felt like their choices didn't matter when they missed. The game is more snappy, tactical, and fun without it!