Date: January 2000 Source: NextGen, Issue 61, p. 84.
This is a vulgar story; I beg your indulgence in advance. We were in the closing hours of OddWorld: Abe’s Oddysee. The game was complete and on the disc, which was getting a final run-through by designers Jeff Brown and Paul O’Connor (we didn’t have a test staff on that job, so the designers did the In-house testing). Jeff discovered a crash bug in one of the later venues; “it might have been the Forest Temple ender,” so of course everything came to a halt while we tried to track it down.
Eric Yiskis fired up the debugger and Paul played through the game, trying to reproduce the crash. Things were painfully slow. Eric was watching the code one line at a time, translating to a slow-motion, fractured play experience for Paul, meaning his timing was all off and he kept making dumb mistakes. Add that to the accumulated fatigue of several weeks of crunch time, and we weren’t at our most efficient. Still, Paul played through the sequence multiple times and we couldn’t find anything wrong with the code or reproduce the crash.
Meanwhile, on the other side of the room, Jeff continued to reproduce the bug by playing off the disc. So, we have a real mystery on our hands: a game that crashes off the disc but runs just fine on the development station, and code that looks correct despite a line-by-line examination by our programming staff.
Impasse. Deadline. Creeping panic.
And then Craig Ewert, another of our programmers, pops the disc out of the machine, turns it over, and sees there’s a bit of crud (well, to be fair, it was… a booger) on the disc. He wipes it off, pops it in the machine, and the level plays just fine. Bug solved. Instead of running it through the de-bugger, we should have run it through the de-boogerer.
Lorne Lanning, President/Creative Director, OddWorld Inhabitants