I'd like to attempt to clear up a couple misconceptions about piracy in this thread. Both of them have been repeated often in this thread and others.
1. "Piracy doesn't cost sales because pirates wouldn't actually buy the game anyway."
In response, I'll quote an article (link below). Piracy does cause economic loss.
The argument is straightforward and both intuitively and logically sound: for every pirated copy of a product, there is some potential loss of income to the producer of that product. This is not the same as saying that every pirated copy is a lost sale. What it actually means is that firstly some proportion of the people who are pirating a game would have bought it in the absence of piracy. Equally as important however is the fact that even those who would never have paid the full purchase price for one reason or another may still have paid some lower amount to purchase and play the game which they pirated. This is because by the very act of obtaining and playing a game, they've clearly demonstrated that they place some value on that game. After all, if something is truly 'worthless', consumers won't bother to obtain or use it in the first place, regardless of whether it's free or not. Even if a game only gives the pirate a few hours of enjoyment, that's still worth something. In the absence of piracy they may have purchased the game at a discount several months after its release, or bought it second-hand for example. So the existence of piracy results in some loss of income to PC game developers, publishers, retailers and even other consumers.
Pure economic loss is actually very difficult to calculate in precise terms because it's largely hypothetical - there's no way of knowing exactly how many more units of a particular product would have sold if piracy did not exist, or how much money various people would have paid over time to buy discounted or second-hand copies in the absence of piracy for example. However examination of piracy figures combined with sales figures for similar products which are less affected by piracy does provide some indication of the scale of loss.
2. "DRM is useless because all games get cracked eventually."
This is simply a misunderstanding of the purpose of DRM and copy protection of all kinds. Do you stop locking your car simply because a determined thief will break a window, or do you remove the locks on your house because physical locks are ineffective at stopping thieves? No. Neither the lock on your car nor the copy protection on a game is intended to be foolproof--it's impossible. The goal is twofold: keep honest people honest, and delay the eventual crack.
Primarily, game companies want to avoid day-zero (pre-release) and day-one cracks. Imagine if DA:O, or any major title, were cracked before release. Unfortunately, I know a number of people who certainly wouldn't have waited for release to play the game. If the game is cracked a month after release, that's still a success. Some people like to play games they didn't pay for, but they're also impatient and are more likely to just buy a game if the crack takes too long. It's the initial month or two after release that is the most important for sales, and that's the time period DRM and copy protection tries to cover.
If anyone is interested in reading an article that covers this issue in depth,
click here.