Deus Ex was a masterpiece of its time and I am amazed at how well it still holds up today, the thing is that it was so under the radar that the only ones to know how good it was beforehand were the guys testing it. While we were raving about it, everyone else wasn’t paying it much attention. The press pretty much ignored it and even the Eidos marketing department didn’t really push it. The game’s success is largely due to the word of month that spread the week it was released. Something I am thankful for as it otherwise would have been a “cult classic”.
I did get to meet Warren Spector on the day of Deus Ex’s release, he was visiting the Eidos offices and he stopped by my area for a chat. He was extremely good natured and even shared a joke with me at Daikatana’s expense although he was saddened by the news that Eidos had just announced that the portion of Ion Storm (i.e. the Daikatana team) had been let go. His branch in survived the cut however and although they produced the ill received Deus Ex 2 they are now looking to make a come back with Deus Ex 3.
As a quick note, swimming is not useless in Deus Ex at all. There are many shortcuts and hiding places that can only be reached with it. If memory serves me right you can even get an augmentation canister much earlier than land lubbers as well.
After the disappointment of Daikatana I was looking for a new team to join but I was getting worried that my six months contract was about to expire. Despite the fact that no one had shown even the slightest bit of dissatisfaction with my work I was getting worried. They had been pretty honest about the whole keeping you on a rolling monthly contract after the initial six months were done. While it was true we had a lot of turnover at the time the people were leaving of their own accord rather than being let go. Even so it was great relief when I was given another title to work on. This meant that they wanted me to stick around at least till it was done. The game was Hitman: codename 47 which was the first game made by the then completely unknown Danish outfit, Io Interactive.
Being assigned to Hitman also gave me a huge opportunity, one I grasped with both hands. The lead tester, Nick had announced his intention to leave a couple of weeks after I was assigned to the project. Being the only other person involved with it this put me in prime position to take over (muh ha ha). With that in mind I nervously emailed the producer to ask about the gig. He then blew a huge hole in my plans by announcing that too was leaving later that week. I politely waited a week before emailing the new producer, although I was half expecting him to leave the company as well. He was silent on the matter though, remaining so even after a follow up email was sent.
I decided it was time to be a bit more proactive and went up to see my boss to have a chat. As soon as I mentioned lead testing for Hitman he immediately offered me the position. I was taken aback but he explained that he had been looking for something for me to do for a while. Technically this was my first promotion in my working career.
Now that I had the lead I had work to do. I quickly found that lead testing is a whole different kettle of fish when compared to being a humble tester. You do a lot less actual testing for a start. Instead your time is filled with report writing and keeping the testers in line. Sometimes you can sneak in a bit of testing now and again but this only tends to happen towards the end of the testing phase when there is lot less micro managing to be done.
A lead tester is normally assigned before there is any actual code to test. This means writing a test plan from the design documents which can be either a laugh or nightmare depending on the developer. Pouring over the design docs does give you an excellent knowledge of the game which can come in useful when you get the playable code and you need to finalise the test plan. Once it is done you hand it out to your testers and assign them their roles. Progress is always slow to begin with as they will need to learn the game but within weeks they will be able to zip through the title at ninja speed (remember, these guys are playing the game for at least eight hours a day).
Once the test plan has all been completed and all of the sections checked off you then move onto “adhoc” testing phase, where the testers simply play the title in any manner they wish. This type of testing tends to find logical bugs such as “drinking a healing potion in the corner of this room causes your stats to triple” rather than any coding bugs which should (in theory) have been caught during the test plan phase. Previous bugs that were found and have now been fixed also need to be retested to ensure that they are fixed, you would be amazed at what developers try to get away with. This is called regression testing and as a lead you need to keep on top of this as there needs to be a balance between regression testing and spending time finding new bugs you need to maintain.
Towards the end of project playthroughs are started. These are to make sure that the game can be completed and that there are no nasty surprises waiting for the player from star to finish. At this point (again, in theory) there should be few new bugs found and most of the old bugs fixed. The pressure is on at this point as any new serious bug could potentially delay the shipping date. This makes the atmosphere very tense and it is not uncommon to see the lead tester and the producer locked in an argument over whether a bug is important or not.
On the last day of testing the QA manager asks you a very important question, is the game good enough for release? At this point you are an excellent position to know the answer to that question. Although in my opinion it is nothing more than a formality, I didn’t see any lead tester successfully argue against releasing in my whole time in the games industry.
So I had all this to look forward to while I lead tested the minefield that was Hitman: codename 47.