Dreamview Ahead: Purgatory Panic

Purgatory Panic is hypnotic in motion. It’s a 2D platformer that plays out along spiralling stages: by wall-jumping and dodging projectiles you progress further into its corkscrewing form, until you reach and destroy the dastardly boss at each level’s centre.

We were blown away by the demo that Dutch developer Erik Roggeveen (known in-game as surrounded_) released for this year’s DreamsCom, which you can play in Dreams right now.

But we just had to know more. In conversation with The Impsider, Roggeveen reveals how he achieved that unique art style, gives his thoughts on what makes a great platforming game, and offers advice on how to make a compelling 2D platformer in Dreams.

How long have you been interested in game development?

I remember as a kid I used to draw Super Mario levels on paper, so in a way I guess almost as long as I've been playing games. I just never had a way to turn those ideas into actual playable levels.

In the past I've messed around with things like Flash and Game Maker, but even creating simple actions was so complicated that I never got very far. Dreams is the first time I've managed to complete a game.

What is Purgatory Panic?

Purgatory Panic is a 2D platformer with short, challenging stages. There's no combat, just jumping and avoiding obstacles. I'm trying to design it so that anyone can reach the exit - but each level also has three collectable souls in hard-to-reach places. Grabbing all those is going to be much harder.

Also, for the first time, I'm collaborating with a composer instead of just grabbing existing songs from the Dreamiverse. landroid9000 is making a custom soundtrack for this game, which has been an amazing experience.

You’ve got this amazing spiralling visual effect where it looks like the player is travelling deeper into the level. How did you come up with this idea? Was there anything in particular that inspired you?

It actually started as a sequel to Cow Runner. I had this idea that instead of running the same circle over and over, I could add more variety by making the cow run on an infinite spiral with randomly generated segments.

That plan turned out to be a little too ambitious. I couldn't make it work, but I did like how it felt to run on the spiral - especially once I fixed the camera to the cow so that the world rotated around you. When I added a jump button it almost felt like a platformer, so I decided to go with that idea instead.

How did you manage to make this effect work in Dreams?

The "spiral" is actually a big hollow tower. If you stand in the middle and look up, the perspective makes it look like a spiral.

Originally my plan was to only show the shadow of the tower, so that the image would be completely flat, like a 2D game. That worked, but the shadow looked very fuzzy.

I realised I could get much sharper visuals if I pointed the camera at the tower itself. To still get the flat 2D image, I turned off all the light sources so that everything is completely black. Then for contrast I put a huge text gadget in the background so that you can still see what's going on.

The platforming is really well done - it reminds us of high-quality platformers such as Super Meat Boy, The End Is Nigh and Celeste! What do you think makes a good platforming game?

Thank you so much! I've thought about this question a lot. I used to think sharp and precise controls make the difference between a good and a bad platforming game, but now that I'm making one myself, I think it's even more important that the levels have a nice rhythm.

Just running to the right and occasionally pressing the jump button gets boring. Each level needs to have a sort of ebb and flow of different situations and obstacles, with quiet moments for players to catch their breath after a tough section, but not so quiet that it breaks up the flow of the action.

Who is the little horned character we’re playing as? What’s their story?

Basically, they work for an insurance company that offers rich clients an afterlife policy. So if someone is worried they won't go to heaven, instead of behaving better they can just buy the insurance. After they die, an employee of the company induces a near-death experience so that they can escort the client's soul to heaven.

It's a very cynical story, I know. It seemed to fit with the dark visuals.

You’ve had experience designing cool 2D games like this with Ultimate 6-in-1 Retrogame Multicart. What’s it like making 2D games in Dreams? What are the advantages, and what are the big challenges?

I'd say the biggest challenge is that you have to work in a more structured way. The tools of Dreams are great for improvising 3D spaces without having to worry too much about the gameplay. If you put a puppet in the scene, everything basically just works. (I'm exaggerating a little, of course.) For 2D games you really need to turn on the grid to make sure everything lines up properly.

The big advantage is that your game plays much more precisely. For instance, in a 3D game, it can be very frustrating to jump on a tiny platform. In a 2D platformer you can create intricate jumping challenges with lots of tiny platforms, because the player has complete and intuitive control over every jump.

How many levels long can we expect Purgatory Panic to be, and when do you hope to release the full game?

I honestly don't know yet. I haven't really planned ahead that much. I'm just making levels until the game "feels" complete to me. It'll definitely be this year though!

If you had one piece of advice for someone wanting to make their own 2D platformer in Dreams, what would it be?

Before you start building levels, take some time to fine-tune the controls. Any sluggishness is much more noticeable in a 2D game. I went a little overboard and made everything from scratch with Movers and Motorbolts, but even if you just use the Sliding Platforming Puppet, you can really change the feel of your game by going into the puppet's settings. Increase Turn Speed, give yourself maximum Air Control, experiment with acceleration and deceleration... These things really make a difference.

And of course the most important advice that I probably should have followed myself: start simple! Definitely don't centre your very first 2D platformer around a spiralling visual effect that makes everything twice as complicated ;)