Developed/Published by: Asher Vollmer and Greg Wohlwend
Released: February 6th, 2014
Completed: 9th February, 2014
Completion: This is an interesting one. I got a high score of 9,420 and decided it made the most sense that I delete it and never play it again.
Trophies / Achievements: 240/1000, but who cares about that kind of thing on iOS?

Threes kind of completely breaks the idea of this blog, doesn’t it? You definitely can’t finish Threes. It’s not a thing you can do. There’s not even—like in many endless games—an amount of content you can see before you’re done, unless you definitely count seeing all the high numbers (who do have a bit of personality/back story?) as finishing it. That wasn’t going to happen.

However, I spent six hours—more or less exactly, I checked—playing Threes on Sunday and at the end I decided I didn’t really ever want to play it again. I think that, contrary to what people demand of reviews, the point where the person playing the game gives up is a super interesting and educational point, and not enough developers really pay attention to that. In general, I force past that point (hell, I beat Duke Nukem Forever through gritted teeth) and that sickness is why I started this—to see if finishing things does gain me more than giving up earlier, whether I like it or not.

So now here I am, trying to see if I learned something by stopping far before that point. In fact, my decision to stop with Threes was completely arbitrary. I didn’t top my high scores (I was third, that day, and I’ve slid down to fifth) I didn’t get some particular amount of achievement points; I just realised that to get a higher score was going to not simply require the expense of time, but that time would be exponentially increased by the randomness inherent in Threes.

Threes. It’s simple. You can combine 1s and 2s to make 3s. But anything above that must be equal to combine: 3+3=6, 6+6=12, and so on. The board is 4×4; you move tiles by sliding them and the entire board slides that direction (unless a row or column is full.) Every slide introduces a new tile on the board. The game is over when the board is full.

I probably didn’t need to explain it, but I did. It is—honestly—very very good. But after six hours I stopped playing. There’s a why to this, and one of the things someone might say to me is “ah but you didn’t understand…“ or “this isn’t a problem because…” no, you see… I already stopped. In six hours, the game actually un-clicked.

In Threes, when you slide tiles, a new tile slides in on one of the rows (or columns) you were sliding. So if you slide only one row, the next tile (which you see the type of in advance) comes in. Brilliant; you have a 2 at the end of the row, you know a 1 is coming next, and it slides in. Perfect.

However: lets say you’re sliding three rows. In one you have a 2, and a 1 is coming next. The other two rows? A 1 is really, really bad. You have a one in three chance of it coming in the right row. 

OK, so maybe you should move something else, right? Get yourself into the right position. No, because that 1 is coming now.

This can scupper an entire game. Is it fair? No, but when I question “should it be?” As far as I’m concerned, yeah, it should be.

Another example: the white tiles. You’re not always sliding in a 1 or a 2; sometimes you slide in a bigger number. As a player, you almost always expect it to be a 3. Except, sometimes it’s not. Is there any way of knowing? Well, the designers have already stated they’ll be noting when the white tile is bigger than a 3 with a + symbol. Maybe this will help, but when your board has 3s, 6s, 12s, 24s and more on it, you’re more likely than not to end up sliding a new 12 right next to a 96, in a game-ending fashion. Is it fair?

These two examples are good examples of things that there are probably ways to change, depending on the type of game you’re trying to make. Maybe you want that randomness! I think what I’ve noticed is that in a very mathematical game, I don’t. Like… at all. But here’s one problem that seems more difficult.

You have a crowded board; a 96 and another 96 are some distance from each other. You want to combine them. However, every move you make brings a new tile on the board. So you have to spend most of your time making sure you are clearing the new tiles that come in—at a maximum, in general, you can allow a couple of moves before you start needing to clear multiple tiles in a single combo move—and every single one of those moves can just keep the 96s floating back and forth, never quite touching.

But of course, allowing (say) more free movement? Could break it totally. Decreasing the number of tiles that comes in? Will definitely make it too easy.

Now, I didn’t master Threes. If I had, maybe I’d be saying “this isn’t a problem, because…” as I was capable of (say) keeping the board so clean that I always had full movement, and could intuit moves allowing me to get my big numbers closer together (on the rare occasion they were far apart.)

But I didn’t, and I’m not. Because after six hours of obsessive play, I stopped and, other than writing his, I haven’t looked back.

Will I ever play it again? No, but there’s probably a version of this game I would.

Final Thought: I played the entire thing with the sound down. I didn’t even know there was an (apparently good) soundtrack until yesterday.