Pixelsocks: Can you tell about your game?

Loren Schmidt

Loren: Star Guard is a simplified, low fidelity, run-and-gun sort of platformer. The core of the game is the level design—that’s certainly what I spent the most time on. My goal was to have the game be fairly scalable with a fairly low threshold for entry. Once you’re over that you can play it at whatever level you want, and it should provide an interesting challenge. It sort of has the same hook as a shmup, but without the brick wall that happens when you’re just not good enough.

Pixelsocks: It’s striking that you’ve taken the old breed of challenge-driven platforming and updated it to be fun by more modern standards. What kinds of decisions did you make to get that working?

Loren: The main thing is not penalizing death. I’m not saying that would work for any project, but it seems to work for this particular type of game.

Some people run into trouble with that, actually. For the vast majority of people it’s a forgiveness, but for some people it becomes a crutch. So what happens is they’ll run into an area over and over again, absolutely heedless of how much they’re dying, and it’s not fun. Well, it’s kind of excruciatingly painful. So it doesn’t work for everyone.

Pixelsocks: On the subject, death is an unpopular mechanic these days. Did that climate influence your decision?

Loren: I was mostly thinking about reducing frustration, because I like the hook of seeing yourself improve over time. I didn’t feel like penalizing the player really served that goal. So it wasn’t really because I don’t think death is good—I just didn’t want people throwing the controller at their computer.

Pixelsocks: It’s always nice to meet a developer who doesn’t want to hurt his players.

Loren: I made a point of making the game feel fair. If there’s a blind drop, it’s always relatively safe at the bottom. There might be one guy down at the bottom, but there aren’t horrible traps everyplace.

Pixelsocks: The game’s enemies are quite diverse. Did the simplified art style help free you to focus on varied enemy design?

Loren: The low fidelity of the game didn’t speed things up very much because I work iteratively. I’ll draw something out, put it in the game, say, “Oh god, it’s too slow. What was I thinking?” take it out of the game, change three pixels, recompile, and play it again. So most of my time was spent on that kind of iteration, and I would have spent most of my time doing that regardless of fidelity. That said, I do think that this sort of low fidelity aesthetic can be an opportunity for people who don’t like doing assets.

Pixelsocks: Star Guard has a “modern arcade” sort of feel to it. Have you considered a feature like leaderboards where you can compete with your friends?

Loren: I have, actually. I’m considering doing a portal version of this game, and if I do I will use leaderboards. However, I have a few tuning concerns. When you play at a very high level, some things are fundamentally broken. In levels eight and nine there are these turrets that have a two-out-of-three firing pattern. That is to say there are three turrets in a row, and there’s always one hole: a turret that doesn’t fire. The problem is with speed runs; the turrets are random. So you can end up, through no fault of your own, waiting for a turret for a half a second.

Pixelsocks: Still, you should never underestimate the internet. You never know how players will work over your game until you’ve released it.

Loren: Yes, and it’s amazing how diverse play styles are. Some people play very cautiously; they’ll kill every enemy and search every room. Other people kill almost nothing. They feel like it’s too much work to kill enemies and so they just jump over. It’s really crazy.

Star Guard – Level 3 from Sparky on Vimeo.

Pixelsocks: Is this the first time you’ve been present for so many people playing your game?

Loren: Physically yes. I got a lot of feedback over the internet from all the testing I did.

Pixelsocks: Have you been surprised by any reactions?

Loren: One thing that’s been interesting is how people react to the boss. Not everyone has made it to the end, but there’s been an interesting split between people who thought the boss is way too hard and gave up, and the people who liked it. More specifically, nothing in the game resets when you die except for the first phase of the boss. I’ve considered changing it so it’s consistent throughout the game.

Pixelsocks: The game certainly develops an expectation, and then suddenly shifts.

Loren: Yes, and your playstyle has evolved to whatever idiosyncrasies at that point.

Pixelsocks: In contrast, however, it’s also the end of the game, so some change can be justified.

Loren: Right, so some people said it reinforced the climax and they wanted this huge hiccup. So I don’t know. From a design point of view, it might be better if it were consistent.

Pixelsocks: I suspect that Star Guard uses enough of the platforming vocabulary—things like safe spots to stand—that players will be able to grasp both approaches.

Loren: The boss’s attacks, and a lot of the hazards like laser beams and turrets, are based on shmups. So you’ll have a curtain of fire with a hole and you have to pause for half a second before you move again. That sort of pattern-based fighting is what I was going for with the boss. So you are at a disadvantage the first time you fight him, and that’s a little unfair.

Pixelsocks: Well, the line between surprising and unfair is probably measured in success.

Loren: Now, I do try to comment every action. So if someone is about to fire, there’s a little event that warns you, or if a turret is about to stop firing, it’ll fade. The idea is that when something surprising or new happens, it doesn’t entirely blindside you. I also try to stick that in the level design, like in the example where you’re jumping in a deep hole, I won’t put lava at the bottom.

Pixelsocks: Did you decide to omit music from the game because it adds loneliness to the already sad story?

Loren: I actually was collaborating with a musician during development, and it didn’t end up happening for various reasons. So I had intended for it to have music, but in some ways I think it’s better without. The soundscape is already cluttered; there are a lot of little attention-grabbing effects. So perhaps ambient music or even just some sort of background sound would have been a better fit.

No related posts.

Related posts brought to you by Yet Another Related Posts Plugin.

Tags:   excellence in design nominee · GDC · gdc 2010 · IGF · IGF 2010 · loren schmidt · star guardNo Comments

Trackback to this article.