"those two to three months when pirated versions were unavailable must have reduced the overall level and impact of piracy" - (referring to game sales of Spyro)<p>I think that is a pretty big assumption to make. Pirates that can't afford the game at all will just wait the 2 months. And people who are die-hard Spyro fans will pre-order or buy in-store so they don't have to wait the 1 week to a few days for the game to be released on torrent sites, or deal with any bugs caused by the crack.<p>There are people who are in the middle of those two groups, who are good at pirating, have enough money to buy the game if it's not available on torrent sites, and don't mind the inconvenience / moral-dilemma caused by pirating. But I wonder if the sales to those people are large enough to offset the performance and technical cost of advanced DRM.
The article seems to be confusing "pirates" with "people who download/buy pirated games". I identify at least two groups of people -- "users" and "crackers" -- with different goals and interests. By their own admission, the "crackers" (i.e. "people who remove the copy protection") actually get a kick out of these protections schemes; the harder they are to crack, the more enjoyment they get out of it. They even state crackers seldom play the games they crack, and that most of their enjoyment comes from the actual process of finding and bypassing the DRM. You could say that for them, the "game" is actually the cracking.<p>In the end they state the effort was worthwhile, but I'm not so convinced. True, they claim the majority of a game's earnings happen in the first few weeks of a game's release, and if they can stop the crackers for that long, it's a win. But it also took them almost a month of programmer's time to stop a crack that appeared a few months after release. Also, the DRM itself made debugging and testing the game very hard, particularly because the effects of the DRM were designed to look like bugs! So at least for the players and legit users of the game, the DRM is a clear lose: a convoluted protection scheme that made the game harder to test and therefore made bugs/design errors more likely.