Creator Interview: Probability 0/Downwell

In 2010, indie game developer extraordinaire droqen came up with a game called Probability 0 (also available on Steam)— a procedural platformer based on going down an endless pit, with throwing stars.

Five years later, indie game developer extraordinaire moppin came up with a game called Downwell— a procedural platformer based on going down a reasonably-endless pit, with gun boots.

Neither of them knew of the other’s game when they came up with their idea, yet the similarities are hard to ignore. A fantastic case of parallel evolution in game development.

I decided to get both together and ask some questions, as a sort of A/B testing of game development, as it were.

How did you come up with a downwards platformer?

droqen: It started with the idea of a platformer where you can only shoot down. It seemed interesting because it meant the only time you could effectively shoot is when you’re the least safe: in mid-air, between platforms, with no solid ground beneath you.

That idea naturally lent itself to enemies coming from below.

I remember I was also thinking in contrast to this game called Icy Tower, where the entire focus was on climbing platforms. I knew that if the screen was always scrolling up, the focus would end up being on climbing instead of other, more interesting things: mostly shooting, but also avoiding fall damage.

moppin: I was super into Spelunky at the time, and wanted to make something similar but playable on mobile.

I thought it might be cool to make a platformer with an emphasis on descent and not much horizontal movement, and having played Ridiculous Fishing on the iPhone and loving it, I could imagine the game I was imagining working well on a smartphone held in portrait.

How did you design the enemies and obstacles to suit the downward environment?

droqen: I just kind of messed around with whatever caught my fancy. I introduced fall damage because it felt right, and I made enemies that did things that I felt the game was lacking. So, I made a really basic enemy, then I added a ranged enemy, then I wanted more ranged-shooting enemies so I made some that shot in other directions.

moppin: I made a lot of generic enemies taking examples from various platformer games and shmups. I experimented with a lot of them and just kept the ones that seemed to work.

Initially all of the enemies could be dealt with by both stomping and shooting, but around halfway through development I realized having non-stompable and non-shootable enemies would mix up the gameplay and add depth so I added that.

droqen: Our approaches differ in funny ways :) I have one conditionally unstompable enemy and one that takes half damage from stars, but otherwise the only kind of stomps vs. stars depth I have is in my upgrade system. I don’t think I ever even thought of making a non-stompable or non-shootable enemy! Maybe I should steal that idea…

How did you come up with and develop the upgrade system?

droqen: I wanted to make a better skill tree. That’s it. I had, and still have, a problem with upgrades that don’t really feel like upgrades — ones that give you tiny bonuses (+5% this, +10% that) or that are chosen out of obligation, e.g. to satisfy requirements of later upgrades. I wrote up a thing about this back when I was running the Probability 0 devlog, but I think the skill tree I made was a real success there!

moppin: I’m a huge fan of Spelunky and Nuclear Throne, and I think anyone who has played those games can see the heavy influences in Downwell.

The upgrade system is very similar to Nuclear Throne, only differing in the lack of exp (radiation) system. In Nuclear Throne, upon collecting enough exp the player would get to choose an upgrade from a random selection of 4 during the transition of levels.

At one stage in development I had the system be exactly the same in Downwell, where the gems acted as exp and you would level up after collecting a certain amount and then get to upgrade during transition. I trashed that as it was it was just too much stealing ideas from NT, and also being a smaller game, I found having upgrades after every level gave the game a better tempo.

droqen: My level transition used to be a menu that reminded you to open it. It sucked. My friend Jim McGinley convinced me to… actually I don’t remember what he said, but eventually we figured out that a big level up laser that also told you how far you’d gotten was a much better idea than a message at the top of the screen that said “hey, press <whatever> to open the menu, do it already”.

Why did you decide on your game’s particular pacing? How did you tailor the controls to follow that pace?

moppin: I generally like fast, twitchy controls for platformers like how they are in Spelunky or Super Meat Boy, so I tried to have something similar to that.

The Gunboot gimmick itself gives the player the ability to slow the fall so I decided that the max fall speed should be pretty fast to encourage the usage of the gimmick.

droqen: I love N. Its use of momentum feels incredible and in contrast I’ve definitely always preferred careful, planned platforming to twitchy and fast. I focused on fall damage because it just worked out so well with my super-simple level generation algorithm… Even a level without enemies is interesting to me when I need to plan these huge, curved jumps just to avoid falling too far and taking a point of damage. Enemies, abilities, other obstacles, and the time pressure of the scrolling screen all just magnify that.

It’s weird how I started with the downward platformer idea because I didn’t want to put too much focus on the platforming but here I am, talking about how much focus I put on the platforming…

Let’s talk art and music. Were you after a particular style or did you just go with the flow?

moppin: Before I started working on Downwell I was going through Game-a-week, where I was making 1 game per week. I had made 12 of these 1 week prototypes prior to starting work with what would become Downwell, and often with making these prototypes I would only use black and white for making the graphical assets as there just wasn’t enough time to spend on art. Naturally I started working on the 13th prototype (what becomes Downwell) using only black and white, and I just kept that palette only adding red to the mix somewhere in development.

As for sound design, I have Eirik Surhrke and Joonas Turner to thank for their super super amazing work!!! Although the retro look of the game wasn’t something I intended for nostalgia purposes, it was quickly decided that the music and the sound effects should be lo-fi to fit the aesthetic. Eirik used Gameboy for making many of the songs in the soundtrack!

droqen: Oh, I totally just went with the flow. I don’t even remember where the idea for layering tracks & adjusting their volume came from, but JMickle did an amazing job with the music and everything. The art style is the way it is out of necessity. My pixel art skills were even worse than they are now. You never notice it in the game because you jump a lot and rarely walk along flat land for more than a fleeting moment, but the main character’s walk cycle is TERRIBLE! Most of the enemies animate by flipping back and forth, haha.

How did you feel once you learned the other game existed? Would you do things differently if you knew it was there?

droqen: I thought to myself, “Yes! This looks great! I wonder if he played Probability 0?” And then I played it, and it was totally solid. I mean, it never crossed my mind that it was a clone or anything — just that it’d be cool if I’d inspired it. If I knew about Downwell in some kind of strange time travel loop… I’d probably get art with more contrast, add more set pieces & dramatic shifts, and work on making deeper areas that feel really different. I would give it an ending, because I think the endless nature of it really turns a lot of people away.

moppin: It was Willy Chyr of Manifold Garden (at the time called Relativity) who gave me an introduction to Droqen around September 2015, and at the time I didn’t know about Probability 0 — Droqen kindly playtested my game and mentioned P0 in his feedback, then I played it and immediately saw the similarities and got super scared thinking that Droqen must have thought I had copied his game! I had played Starseed Pilgrim but never Probability 0

Upon playing for enough time to see the similarities, I quit playing and haven’t touched it since then from fear of getting too many ideas from it and constraining myself as a result.

I think, if I knew that the game existed and had played it before working on Downwell, it would have been a lot different to avoid being so similar to P0…

Thanks to droqen and moppin for participating in this experiment.

You can find more of my games non-writing work here.