Neon Journey

I’ve recently designed and programmed “Neon Journey”, a deckbuilding game. You can get the downloads here: OSX. Windows.


For those who don’t know, I think that games about building are the way forward, and in particular deckbuilding games are some of the best building games that humans have made to date. One of the strongest aspects of deckbuilding as a mechanic, in my opinion, is its flexibility: you can throw a lot of other mechanics in with it. However, I find most deckbuilders fairly stale in this regard, returning to boring “victory points” setups instead of covering new ground.

One time, me and the boys were talking about game design, and someone said that a spatial metaphors could clearly communicate complex concepts quickly to the player, while also being ripe with potential depth. To my knowledge, a deckbuilding game based around maneuovering in space didn’t exist yet, so I immediately started thinking about ideas for how to implement it. After several false starts, I was finally successful in its implementation, and so Neon Journey was born.

Overview of how to play

The main avatar of the game is called the Neon, and every time it successfully moves a space, the player gains a point. Each turn, four cards are drawn from the deck with various patterns on them. Each pattern corresponds to a different action: some patterns move the Neon, some allow it to dash, some drop a ramp that it can later use to get additional moves, and so on. Then, the player picks a card from the bank (a set of four additional cards), and replaces a card from the hand with it. The replaced card returns to the player’s deck,and the commands on the remaining hand cards are executed. Then, the cards are discarded and the next turn begins.

There’s a small tutorial screen in the game itself, where things are explained better if you actually want to play the game. For reading this article, the above should be enough.


  • Back in the day, all of the binary goal essentialists were screeming about how with the variable match lengths that a score goal brings to the table, it’s difficult or impossible to control long arcs. Ever the defender of score systems, I haughtily responded that by tying the arcs to the loss condition, we could have long arcs and continuous scoring side by side. That’s exactly what I’ve done with Neon Journey: on the title screen, you can see a number that’s the average score of all your games (adjusted down a bit if you have very few games), so there’s no binary win/loss. Each game is exactly 20 turns long, however, and cards get progressively better as the turn counter grows, adding pressure even in the later turns and allowing arcs in spite of the lack of binary win-loss.

Note: at present, I’m actually a supporter of binary win loss over score goals. Also, I don’t think the arcs in Neon Journey are that compelling. But it’s a proof of concept for the other team, at least.

  • Hopenager and I think that a game should have as few possible input actions for the player as possible. I support basing games around one core, difficult, complex decision, which I call the core decision, that’s instantiated over and over throughout the game. Neon Journey successfully shows how such a thing looks in practice. Although there are only 4 moves per turn, picking which card to take has many repercussions. It affects what actions you get to do on the immediate turn, what cards will be reshuffled into the deck in the near future, and what cards will be in your deck in the far future, as well as modulating it all by the order that the cards are in, which adds its own considerations. In short, there are a lot of things to take into account when picking between what amounts to a single decision with one possible action and a few possible choices, which perfectly demonstrates what I think game designers should aim for.

  • I’ve historically played a lot of a game called “Eight Minute Empire”. In that game, cards are layed out and periodically move to one side as space opens up when cards are drafted. I think this is completely ingenious, and way better than the “opponents buy cards and then random new ones show up with no warning” thing that deckbuilders like Ascension have going on, and also better than the “every bank is a puzzle, once you know how to play it you’re just reading the script” thing that deckbuilders like Dominion have going on. This way, each card in the bank sticks around for several turns, but also not forever if it’s not bought. I think it’s the prime way to handle info horizon in deckbuilders


In spite of my careful machinations, the game has several flaws, which I’ll address below:

  • One of the great joys of deckbuilding games is that the cards acquired early in the game can be used to ease the acquisition of additional cards in the later game. That doesn’t happen at all in Neon Journey, so there’s always a very moment-to-moment, tactical feeling about which card to get, with no larger strategy. When I conceived of the game, I expected that the ratios of different directional arrows in the deck along with the cards that develop the board, would be enough to create longer strategy arcs, but I think I was wrong. Hastily, I implemented two mechanics to try and solve this: first, there’s a green zone that appears on the board called the “trash zone”. When in the trash zone, replaced cards are “trashed” instead of put back into the deck. This was meant to give the player more control over the deck, and let him pick current cards to try and strategize to move his Neon into the trash zone (which moves each time it’s used). The other mechanic is the fact that as the game progresses, the card generator will generate better and better new cards for the bank, leading to a feeling of progression and “late game power spike” that I think people crave. This has nothing to do with the cards acquired, though, so I think that it too was an inadequate solution. Sad!
  • Four options per turn is way too few. I remember one time me and the guys at Dinofarm were trying to figure out how many choices made for the most interesting decisions, and I think now I’ve found a lower bound. This is compounded by the fact that the cards generated are very random, and often you get cards that are incredibly awful for your current situation (note that I tried to minimize the amount of cards that are terrible in general, but situational considerations mean that almost always, at least one card is clearly bad). Anyhow, I think you want to generate a lot of possible choices, and then let the player prune off the truly garbage ones, leaving a few remaining. It’s hard to build that pruning into the game.
  • LOOK AHEAD. You knew this was coming. It’s super mentally taxing and prone to mistakes to look ahead to see what each move does. Or so I thought. Once I was actually playing the game, I decided that this look ahead was 100% ok, and didn’t bother me at all. That’s not what I was expecting because I hate look ahead, but maybe that’s the upside of 4 possible moves each turn, haha. Maybe you’ll hate it, anyways.

The Future

I made this game pretty quickly, and I never plan to update it unless people are bleeding out of their eyes because they like it so much. I will fix any bugs, though. I’m pretty fickle and unmotivated when it comes to game design (and really everything) so I’m not going to falsely promise that this game will ever be any better than it is right now. Hopefully you learn something from it, I know that I did.

Written on February 3, 2018