StarCraft II Cost $100 Million To Develop
UgLyPuNk writes with news of a report that Blizzard has spent over $100 million developing StarCraft II. Initial development on the game began in 2003, and it's due to be released on July 27th. Activision Blizzard CEO Bobby Kotick "described StarCraft as one of the company’s seven 'pillars of opportunity' (where each pillar has the potential to deliver operating profit between $500 million and $1 billion over its life span)." The finalized system requirements for the game have been released, and players planning to buy the digitally distributed version can download it now, though it won't be playable until the 27th.
If the crackers find a way to play before the start date.
The main reason that I would suspect a distinct pre-release cryptographic mechanism is that such have been seen before(I believe some Steam titles have used them) and that they are so utterly trivial yet so functionally unbreakable.
You would simply take the release installer, and encrypt a copy with a key known only to you(and probably stored on a securely-locked-away air-gapped medium, to prevent leaks.
Add a little stub program that does nothing but check your website for the key, decrypt the installer binary, and start the installation.(Because a key doesn't need to be all that long in order to be functionally unbreakable, it is even practical for those without web access to type a suitably encoded version of the key in manually).
Absolutely no "innovating" needed. Basically any encryption method that isn't declared "deprecated" will work, and implementations of most of them are available under pretty much any license you want. The total implementation time will be a few hours for a competent programmer(and it need not be a competent programmer who has any knowledge of the project, this is quite a generic thing), possibly a man-day or two if the decrypter needs QA on 15 different Windows localizations and some attractive splash-screen art. And yet, despite the ease of implementation, even three letter agencies won't be able to get to it until you release the decryption key.
Aside from the fact that it is easy and robust, the main reason to use a separate system for the "release date control" vs. whatever DRM is used post-release, is that market research suggests that the financial damage of having your DRM cracked tapers off fairly rapidly post-release. Having would-be early adopters downloading pre-release cracked copies instead of buying $150 "platinum packs" with a couple of useless trinkets is financially painful. Having cheapskates a year from now picking up off the Pirate Bay rather than Ebay is virtually irrelevant. In between, the value falls over time, fast at first, and gradually tapering off.
If the installer binary is encrypted, would-be DRM-hackers don't even get to look at the DRM until release day(whereas, if you depended on the release-DRM, they would have the extra 10 days of hacking done before the game is even supposed to be released). This means that the chance of a pre-release pirate version(barring a penetration of your systems by hackers or inside guys) will be impossible, and the time-to-working-crack will be 10 days longer than it otherwise would be...