Frayed Knights: Why Is This Game Taking So Long?
And here are a few words of update on the making of Frayed Knights, the comedic indie RPG in development by Rampant Games.

If there's a single common thread linking several of these updates together, it is, "Things are taking longer than I expected." I've yet to hear of a game getting done significantly faster than expected. Making games is hard. You'd think that after fifteen years of doing it, I'd know this by now.

You know, when I was working on the pilot episode of Frayed Knights, I had a couple of ideas. Really silly ideas, in retrospect. I should have known better. While I've never made a full-fledged RPG before, I've made plenty of games. So I don't have an excuse. Maybe it's some kind of suppression of bad memories taking place that made me believe these things.

But here were my beliefs: That when I finished the pilot, I'd have "most" (75%+) of the code for the game finished, and that my job from there on out would just be cranking out content, which would be easier and faster than code.

Wrong on BOTH counts. But I'm just gonna talk about the code part today.

My existing code, in many cases hastily thrown together just to get the pilot out the door, was woefully inadequate for the true rigors that would be demanded of it. Or the demands of players. In some cases (like the inventory UI), I've completely replaced the earlier code. In other cases, I've had to augment the code to something like quadruple its original size.

And then all the little "bits and pieces" that were ignored / shoved under the rug? That was a lot more than 25%.

Then we have something that programmers jokingly call "bit-rot." It's how we describe our code seemingly "wearing out" and developing bugs as it gets older and remains unmaintained. Naturally, code doesn't rot or change at all with age. But it doesn't exist in a vacuum. As its environment and dependencies change, it may fail. Or we may expose bugs that were always there but never visible. Or we call functionality that was never 100% done in the first place. Or... something.

So what happens is that code that was "working perfectly" (hah!) before suddenly, inexplicably, begins failing. New bugs crop up in old code that hasn't been touched in MONTHS. In several places. And as you fix the code in one place, something else breaks elsewhere.

"Bit-rot."

So I've been spending a lot of time stitching things back together again. And integrating "proof of concept" code into actual working code that functions as it should in a real game.

Of course, there are other explanations, like my playing the Three Musketeers RPG or Knights of the Chalice instead of coding. I guess that's how those guys take care of their competition.

But hey, enough grousing. Here's what's new:

I've taken the visible "wandering monster" thing and made them actually part of the game instead of a tech-demo. So now you can dodge enemy patrols that you can actually see. But their respawn rates are dependent upon how well you've cleared the local area. There are still a bunch of issues to be ironed out there, but it's functioning.

I keep enhancing my dialog scripts to provide greater functionality within conversations. Now talking with people can give you XP or drama stars, give or take away items, set or modify game flags, or call an internal game function.

The area surrounding the Caverns of Anarchy is --- well, not done, but the principle geography is laid out. With a stagnant lake and everything. It's kind of a dismal place. But then goblins and lizard-men were never known for being brilliant landscapers. We've got two of the caverns "done," but they don't look much like caverns. That's kind of an issue when using the kind of CSG geometry we're using for interiors - organic-looking spaces are VERY VERY HARD to do. It's much better at representing regular, man-made locations.

Our explanation is that these caverns began as smaller, natural locations, and were later further excavated and built upon by their residents - who ended up living there for generations.

In spite of bug-fixes, all quests involving the Tower of Almost Certain Death should be completed this weekend (they should have been completed LAST weekend, but I kept getting distracted by other tasks), and I'll be working on some of the other quests and locations in the eastern wilderness and the "follow-up" scripting back in Ardin. The party gets (temporary) access to a major artifact and clues to a major mystery in the tower, and thus sets off the main storyline (although it was hinted at in the earlier parts of the game). This opens up the Caverns of Anarchy area, and Fishy Stuff Starts Happening.

Labels: Frayed Knights, programming

- posted by The Rampant Coyote @ 5:59 AM (Permalink)