nipsen said:
.. I think his point was that in order to get it to work, it would have to involve basically streaming the entire game off a server, as well as have programmed into it so complex routines that the work on the actual game would take a back seat.
I'm pretty sure that wasn't his point. His argument rests in flags. Flag off: Enemy is not present. Flag on: Enemy is present for assassination. And so on. Every game can have thousands of those flags, easily, and they're very small pieces of data. All the graphics, the sound, everything else can be on the disc. That's the big stuff. The stuff that's streaming are the tiny data points that govern the game's logic.
I quote Shamus:
"Your local copy of the game is missing key bits of game logic. Out of the box, the game doesn't know where characters are standing, where the cutscenes are triggered, what items are in the area, or even where the player should appear. All of that is on the server, and the server doesn't send it until the moment that you need it. That information is small (easy and lightweight to transmit) and if it's wrong the game will break."
There's also the problem that - as long as they're not on a 100% protected platform, the data in the game has to have been complete at one point, and that data can be duplicated. Such as coding a video, and expecting it to be impossible to copy, even though you can, at some point, see it uncoded with your eyes.
Sure, that's roughly the point of his original column, "Impossible DRM [http://www.escapistmagazine.com/articles/view/columns/experienced-points/5930-The-Impossible-DRM]". It's very difficult to somehow control how a system is used when the user is in complete control of that system. That's why conventional DRM cannot work.
But this time, part of the game
is on a 100% protected platform: The Ubisoft servers. That's the whole point; they've taken a major element of control away from the user.
If that isn't an option, then the external system can be duplicated, with the logic and the resources being spoofed. Even if it's dependent on individual keys - there's nothing stopping you from duplicating a single session, and then duplicating the responses from that play.
Nope. Here I quote Shamus Young again. This is a long one, so I'll use the quote tags.
Shamus Young said:
But it's possible to make this process a gigantic and time-consuming pain in the ass. For example, if the triggers all behave slightly differently on different difficulty levels, then the cracker will need to play through the full game on every difficulty to get all the information they need to make a complete server.
Then the killing blow: Make the various triggers dependent on branching player behavior ... Suddenly the cracker needs to play the game all the way through on every difficulty and following all of the possible branching paths if they want all the data they need to make the game work.
Cracking is fun and exciting now because the cracker can get the game a day or two before release and have it cracked before launch. They get to "defeat" the DRM-authoring numbskulls at SecuROM and feel like heroes. The adventure becomes a lot less fun if they have to wait until the servers go live at launch, and then they have to labor for weeks or months and play the game until they're sick of it. And when they're done, they'll have a crack for an old game that nobody cares about anymore.
Anyway, back to your post.
nipsen said:
The point is that the only way to make a system like this work, is making it so complex and intricate that it will take too much work to crack, for anyone who isn't working full time on it.
The lemma to that would be: "at this point, the drm is more expensive than the actual game".
And that's not an exaggeration - it just means that the degree of complexity a system like that would have to have - would involve it being part of the program-logic, and be part of it in such a complex and manually created way that it isn't viable to try to circumvent it.
I don't doubt for a moment that publishers would want that to happen, though, or that they would easily spend money on it. I'm just doubting whether anyone on any kind of project would appreciate being stuck with having to design drm-tricks in between every quest-item.
Right on the first point at least. That's the whole goal: Not to make it impossible to crack, but to make it ridiculously slow, boring and painful to crack. Delay is the deadliest form of denial. No security system is impossible to beat, but you can make it not worth your while to beat it.
Too complex to work? Maybe, but I'm unconvinced. If the DRM production software is built into their own development tools, then it would take very little extra effort at all: The game itself would be looking for the flags. Remember, building in flags is normal development practice, it's just normally to govern gameplay rather than prevent copying. It's possible to conceive of a system that builds the DRM entirely on its own, based on nothing more than normal game development. So the real complexity is in building that first system, and while not a programmer, I don't think that sounds undoable.
Again, I wish to add: I'm a detractor of this DRM system. It's unduly intrusive, like all DRM systems punishes the paying users instead of the pirates, and is of dubious use in lifting sales. (I said earlier that I was unconvinced that AC2's DRM hurt sales. That's true, but I'm also suspicious that it helped. My guess is that it was a wash, neither helping nor harming overly much. But it's only a guess.) But let's give credit where credit's due: It's a system so unbelievably horrible that it actually works. That's more than any other DRM system can claim. And it's why, I fear, it will be imitated by other game publishers.