Mikhail Shumilin

Game and Level Designer



Last Wednesday marked my senior’s game first official playtesting session. Sure, people tested Tumbleton’s Fortune earlier while it was in development, but it was mostly me asking them specific questions like how movement felt, were the levels too big or small, and what did they think of game mechanic A, B, C, etc.

This time was different – I set the game up at our 2 designated stations and started inviting people to come check it out. Overall, the feedback was mostly positive which to my delight was better than I expected.

Our main game mechanic involves players picking up powerup objects, and then attaching them anywhere in the level. Getting players to understand how to use this mechanic and what they can do with it is a hugely important task. If players can’t understand it, they won’t enjoy the game, and won’t be able to get past some of the later levels, no matter how good they are at platforming sections.

For our first official playtesting session, I chose to take a more freeform approach of not explaining everything to the player. I wanted to see if the mechanic was easy enough for players to figure out by themselves. I believe that the simplicity of a gameplay mechanic is closely tied with how “sticky” it is – how fun and intuitive it is to use. If a game revolves around a mechanic, it better be sticky, otherwise players will put down your game.

Once the players pick up a large, green and spinny powerup object (it will have particle effects emanating from it eventually so it will be impossible to miss), a message pops up on the screen: “You have picked up an object! Click on it and use it to help you cross the gap”.

And that was it. It was up to the players to figure out that you can click the object icon in the player’s inventory on the bottom right, position it on the screen, and place it.

Most players quickly figured it out after trying a couple things. A joyful “Aha!” moment followed their success which made me smile. Interestingly enough, after rolling across the bridge object that they just placed, none of the playtesters attempted to pick up the bridge object and reposition it or place it somewhere else. It’s true that I never mentioned that you can do that, but in player’s minds, once they used up an object to come up with a solution, they forgot about it.

Obviously, the directions won’t be so ambiguous in the final version of the game, but it was a nice experiment into the psyche of new players.

A couple playtesters weren’t sure that the object attachment worked anywhere on the environment, some thought objects could only snap to the ground. The 2nd puzzle where you have to attach platforms to a wall to vertically scale it aims to teach them that, and putting some text explaining that you can attach objects to a wall would solidify it.

Because of this however, I’m leaning more towards locking the object down after a puzzle that requires it is complete.

The movement of the ball (Tumbleton) was hailed as just right by almost everybody, which made me breath a sigh of relief. I spent a lot of time tweaking things like movement speed, velocity, friction, stopping speed, jumping height, and the general feel of the ball. It felt great knowing it paid off.

More importantly, players said that controlling the ball just felt fun. They enjoyed rolling and jumping around, which is a testament to the controls being sticky.

Multiple playtesters complained about how jerky and fast the camera rotation was, an issue I suspected they would bring up. I relayed this information back to the programmers and we quickly smoothened out camera rotation.

We had a similar issue earlier in development when zooming in or out the camera. You don’t want the camera to zoom via clicks – fast movements in or out. Instead, you want to smoothen it out so it has more of an analogue feel, similar to how most RTS games do it. We fixed that early on, and I suggested that we also do the same thing for camera rotation while we were at it, but we put it on hold at the time. I’m glad the problem is finally fixed.

Since the game is a mix of platforming and puzzle solving, some players said that it was kind of a bummer that after having so much fun jumping around and navigating the terrain, they had to stop and figure out how to solve a puzzle. I think they said that because this was the beginning of the game, and they quite didn’t know how to play it yet. In later levels, I anticipate players going through the levels in a more quicker pace.

This reminds me of the flow in Mirror’s Edge. Subsequent playthroughs of the game were always more fun than the first because you knew where to go and where to jump, which helped you keep the flow and speed of the game going.

Some playtesters also complemented our graphics technology like the bump mapping and shadows, a nod to our graphics programmer. Even though most of the textures in our game are placeholder (they look too realistic), it’s still nice to know that the technology is there.

In the end, Tumbleton’s Fortune playtest session #1 was a huge success, and I’m looking forward to doing more of these in the future.