?

Log in

No account? Create an account
Zorba the Hutt's Journal

> recent entries
> calendar
> friends
> mandible games development journal
> profile

Thursday, November 5th, 2009
2:51p - Make The Number Bigger Postmortem

Okay, let’s polish off the Number+ postmortem real fast.

Number+ spawned out of my recent thoughts about different kind of games. Number+ is a pure skill game. The challenge is whether you can react fast enough and accurately enough to all the shit that comes flying in. If you do, you win. If you don’t, the game ends after a bit, theoretically to keep you from being annoyed at being stuck in a holding pattern. You can absolutely get better at the game with practice, and that’s rewarded by higher scores and new notes about what your number is as big as. There’s some high variance later on, but the game’s really built heavily around slow-and-steady-wins-the-race, and playing conservatively is how you beat the game (in fact, it’s nearly mandatory near the end to play super-conservatively when adding a new digit.)

Two things that worked well: sound and graphics. I pulled (almost) all my sound off Freesound and, for the first time, ended up with a game that doesn’t sound tinny and painful. That was pretty awesome. I will be doing more of it in the future. The graphical style was a total accident, I was just using white silhouettes on black for debug art for a while. Then I decided they looked good and I stuck with it. I’m not entirely satisfied with how I did them – I have three separate styles of silhouette going on – but they’re easily good enough and I think they result in a rather neat feel to the game.

Two things that didn’t work well: rhythm and balance.

One of the ideas I had for this game originally was that there would be a lot of stuff going on, all rhythmic and on-beat, and you might end up with kind of evolutionary music coming out of it. That just did not work. There’s a lot going on, and most of it does in fact happen on beats, it’s just that it’s impossible for anyone to notice. Better programmatic control over sound would probably have helped – if you’ll believe it, this is the first game I’ve written where I have the capability to *stop* sounds, and I certainly don’t have anything fancy like pitch control – but overall, I think if I want a melodic game, I’m going to have to focus very closely on that aspect.

Game balance turned out to be a major problem, and I’m not sure it’s solvable within the constraints I originally wanted.

The problem comes down to a diminishing click budget. Let’s say we trigger a challenge for the player every four seconds. This is not hard at all. Nobody will find this difficult. Let’s say we add another challenge, again every four seconds. It’s slightly difficult, but any serious gamer isn’t going to have an issue. One more thing every four seconds, suddenly it’s hard. One more, suddenly it’s near-impossible.

If you want to introduce about eight different Things over the course of this game the user will have a shocking amount of trouble just clicking on them all. Worse, if you want them all to “trigger” at roughly the same frequency, the first five will be boring and the eighth will be impossible.

I tried to solve this by reducing the frequency of the later items, but that introduces an all-new set of problems – namely, that it’s hard to figure out how to interact with the gizmo in time. There’s always a limited amount of time to react, and by the time you’re near the end, any slipup is essentially fatal.

So the first half of the game is boring, and the next third is challenging, and the last sixth is near-impossible, despite the fact that the hazards actually get [i]easier[/i] as the game goes on. And in order to beat it, you have to sacrifice six playthroughs to the God Of Learning Game Mechanics.

This is bad.

If I were re-implementing it, I’d change it from a single large number into a series of smaller numbers, each one of which teaches you about some new gizmo, then challenges you with a set of things you’ve seen before in a new combination.

A few things got lost in design. I was originally going to have some kind of “store” where you spent your number in order to get gizmos that let you pass challenges. I couldn’t come up with a way to balance it that I liked, however – the game is kind of intrinsically exponential, so I’d have maybe thirty seconds of gameplay in which the gizmo was an interesting decision, and after that it’d just be another-thing-to-click.

This is the first game I’ve been tempted to release commercially, since it would be dang easy to rip out a few dozen levels and gizmos and put it on the iTunes store or something. (Multitouch would let me do some fancy stuff as well.) There’s little enough art and sound that I could probably hire someone for relatively cheap to do things better than I did. I don’t think I have time, but I kind of like the idea.

Learning continues.

Next month’s official theme is Art Game, which I have absolutely no interest in. I’ve been thinking about games that fit the theme and the fact is that all my ideas come down to “I am shoehorning this into the theme just so I can say it fits the theme.” So I’m going to be ignoring the official theme and doing something else.


<< previous day [calendar] next day >>
> top of page
LiveJournal.com