The QA tester's mentality
Interview with Joe Cain, game producer and 1990s Sega QA team alumnus
A little while ago I was analyzing the Bug! ROM, as you do. One thing led to another, and I started talking to one of the game’s testers, Joe Cain.
I dreamed of having the job that Joe did in the 90s: game tester at Sega. Check out his credits: he did QA for Ecco the Dolphin, Sonic Spinball, and Sonic & Knuckles, among many other games that I couldn’t get enough of as a kid.
In the interview below, which is condensed and edited from our correspondence, Joe describes his time at Sega, his stints with Atari and 3DO, and how working in QA informs his outlook today.
Q. In your time at Sega, are there any notable projects that you worked on that were ultimately canceled? Did you ever get to see any of the infamous Sonic X-treme efforts? How about B-Bomb?
Joe Cain: I can't recall seeing Sonic X-treme or B-Bomb, but one game that does come to mind is Nightmare Circus.1 I was working on it just before I left Sega.
That game was unique because of its extensive customization menus. Almost everything about the character movements and animations could be tweaked. Most games have these tune-able parameters like this, but they’re buried deep in the code and can’t be edited. A programmer has to change them, make a whole new build, solicit testing feedback, and then repeat.
With Nightmare Circus, there was a command to export whatever values you were using for your game, and those could be sent back to the developer to incorporate into the next build. It let us testers function more like game designers in the development process.
Another one that I remember was an Indiana Jones game that was never came out for the Sega CD. It was so buggy and terrible! I don't know exactly why, but it was essentially impossible to play, like everything was super slow and it would just crash all the time.
The funniest bug I ever encountered was on that game, because there was a level where not much seemed to really be going on, and I was just moving Indy about and whipping stuff and suddenly, the Model 1 Sega CD I was playing on ejected the tray right in the middle of playing! I wondered what might have happened if I was playing on a Model 2 Sega CD at that time - I know it wouldn't have forced the tray lid to pop, but whatever it was doing might have been an even more nightmarish bug on that hardware.
Q. You were often testing games that had already shipped in Japan. How did Sega handle things you found during testing? Did the Japanese teams do bug fixes? Were American programmers involved?
I would say that SOJ's games always came in really clean, at least from a bug perspective. We were shocked if there was ever even a bad build or any easily reproducible serious bugs or crashes. The original teams did the localization work. They knew the games best, so I think it only made sense for them to do it.
The biggest challenge was localization of things like RPGs with a lot of text. We never really had "All Your Base"-level grammar nightmares to deal with, but we certainly saw a lot of L’s where R’s should be and vice-versa. Sometimes you couldn’t say for sure what the actual fix should be. Outrun 2019 had a music track titled Running Preasure - was that supposed to be Running Pleasure or Running Pressure?
Dealing with localization at Sega served me well a few years later when I had to localize Tiny Tank into 5 languages - including Japanese!
Q. I'm sure you've heard from Sonic fans over the years asking about prototype versions of games that you worked on. Do you recall anything notable about your testing them? You don't have any old builds in a closet, do you?
I was one of the two lead testers on Spinball. I don't remember seeing anything in that game that was working and then got left on the cutting room floor, though there's this really dull ache in my mind about the audio in that game needing to be changed very late in the production cycle…
I was also around for testing of Sonic & Knuckles and Knuckles’ Chaotix. I don't remember exactly how we tested S&K without having final “lock-on” carts to work with. We must have had some kind of specialized hardware, because I do recall plugging in retail cartridges to make sure everything worked as intended. I remember liking Chaotix a lot because it had such a cool and unique mechanic.
Sega QA was pretty tight-fisted with ROM boards and discs. If anything ever did go missing, it was a big deal. I actually got myself in a bit of hot water early on in my career there with regards to piracy. Right after high school, a couple of my friends got wind of the "game copier" market, and we all paid something like $500 each to get a Super Nintendo copier of our own. I happened to mention this to a producer on our team, and I think I almost got fired over it!
They actually made me give up the copier during the rest of my tenure there, so it was well over two years before I saw it again. After that experience, I was extra careful not to forget to return carts or burns at the end of my shift. I still have the copier, but its RAM no longer works, sadly.
I do love seeing the dev builds of games that pop up. They're a throwback to that time and remind me of that “organized chaos” we were going through back in those days.
Q. What was the atmosphere like at Sega as you entered the mid-90s? Did it seem strange to be supporting so many systems at once?
When I was there (from 1992 to 1995), the atmosphere was very… self-confident. I don’t want to sound overly critical, but the company was maybe a bit too smug about its success. Like, Sega was definitely doing some things really right to compete with Nintendo at that time, but I always saw it as something more of a "style over substance" battle. I figured it would ultimately lose on that basis because there were just more higher quality experiences on the SNES. (Full disclosure: I am, and always have been, a Nintendo fanboy)
We all absolutely felt like Sega was trying to do too much too fast, and it definitely felt like a house of cards to me. I just didn't believe that you could convince people to keep spending money on so many different game systems all at the same time. Were you really going to buy a Genesis to play at home, then upgrade with Sega CD and a 32X? Get a Game Gear for the car, and a Pico for your toddler?
There were all these peripherals and accessories that they expected people to buy, too. The Activator was clearly just a gimmick, VR wasn't ready yet, Menacer felt like a total knock-off of the Super Scope. All of this stuff made all of our lives miserable in QA, because they all had to be tested.
I don't know how many people know what an address checker is, but we had to run every game we released through one at some point and the more bits and pieces you support with a game, the more likely that game is to throw an illegal memory address access error. We had to have checkers for every console, Genesis/32X, Game Gear, Saturn, etc. Trying to ensure quality across all of these systems was a huge undertaking!
Q. You spent time at 3DO and Atari, two famous casualties of the console war. Did their difficulty competing affect your work?
At 3DO, it felt to me like the issue was growing too quickly. They made some money and tried to do too much with it. Some really promising projects were coming online right as the doors were getting ready to close, but the company couldn't finish them before funding ran out. It all just imploded and was a really bad ending for a number of folks who unfortunately went down with that ship.
There was a similar lack of resources at Atari. When I started there, Atari had not turned a profit in 5 years, so they were very much fighting with one hand tied behind their back. Then PlayStation came out and just flat-out ate their lunch - and pretty much everyone else's.
The issue there for me, at least in hindsight, is that I think many of the upper management folks knew the writing was on the wall, but they weren't going to say that to anyone when there were still development contracts to fulfill and games to (hopefully) get released. I think they wanted products in the pipeline in case things miraculously turned around.
You knew something was wrong because good work wasn't being rewarded. I was getting desperate calls from devs asking for their next milestone payment so they could keep developing these games they'd been working feverishly on. I even remember one conversation happening about how we might get a particular team into contract breach so we could basically yank their tech for free.
I wasn't very keen on that plan, so in a way I was relieved when I was let go because it felt like everything about that situation was wrong and only going to get worse for anyone who stuck around. You can read Clipped Claws for more good info regarding the end of Atari.
Q. How did your background in QA influence your later involvement in design and production? Do you still see things the way a tester would?
If there's one thing that made me a successful tester, it was attention to detail and being able to kind of notice the little things that other people might not catch.
Obviously bug testing is different than game design or production, but certainly being exposed to so many different games was an education. And watching development cycles play out showed me that you can’t just throw together a game on the strength of a good concept and some capable devs.
The experience really opened my eyes to what successful commercial game development was all about. Making a great game (which is hard!) is only a piece of the whole puzzle - marketing, branding, merchandising, and pricing all matter. And some good fortune never hurts! I had naively thought that coming up with good game ideas was enough to guarantee success.
Testing taught me that there's a lot of trial and error involved in everything, including building a career! I had down times (as we all inevitably do) but I kept looking for opportunities to prove that I could bring value to a team, and I was just fortunate enough to survive that without having to give up on my "dream" career path. I had down times (as we all inevitably do), but I kept finding work and kept learning and trying to prove that I could bring value to a team.
I definitely still have a lot of that tester mentality in me, not just in my day-to-day job (which does sometimes involve testing what I'm working on), but also in other parts of my life. Being a tester is about finding problems, which really unlocked my ability to look ahead and try to prevent problems from happening in the first place. You can't avoid problems entirely, and solving them does get easier with experience, but I definitely wouldn't have developed the skills to do that effectively if I'd never been a tester.
Thanks for reading! Follow Joe Cain on X / Twitter for more from him, and stay tuned to Rings of Saturn for more.
Nightmare Circus did finally get released, sort of. It came out in Brazil, and a version of it was distributed by Sega Channel.