Effective DRM is less about technology, and more about incentives. It would not even be hard to implement if the sole goal is to maximize revenue. Sadly, most DRM schemes has a secondary goal of control rather than revenue maximization.

So how would this work: 1) have a early launch code included in collector versions and pre-order, for something around 1 week. 2) include a special DRM system for only the pre-launch week, and announce that it will be removed after the pre-launch week.

the benefits: 1) Crackers are people, and there is little incentives for 1 week crack. A honor system could also encourage them to avoid cracking the game, if this model is made in honest. 2) Buyers are people, and those that would buy the game at launch (90-95% of revenue) will not have a pirate game to pick from. 3) pirates are people, and those that rather wait for a crack will wait here too.

Steam provides DRM and a marketplace.

All DRM is eventually bypassed, why do game studios even bother?

Because every hour after release that it's not-yet-cracked equates to a substantial improvement in lifetime revenue.

They're not stupid, they know it's getting cracked. It's just a play for more time.

Maybe the case for non EA games, but considering people won't get to play Simcity within the next 2 weeks or so, there'll be a cracked version out by then. EA have brought themselves nothing but trouble in this situation and being the out-of-touch company they are, they deserve every loss of sale they get.

This isn't like normal DRM, key parts of the game logic and performed server side so you would need to reimplement the server in order to crack it.

Think of it like trying to get a "crack" for 37s Basecamp or similar.

While this might sound as a big problem for crackers, I slightly doubt it. I suspect many crackers has related day-jobs, like reverse engineers. If their day jobs require them to reverse engineer the command structure of a bot net, or make deep package inspection rules for protocols, then reverse engineer a game server is not that complicated thing to do. They are going to have special made tools for this.

Reverse engineering the binaries and network protocols is not going to be so difficult. What would be more time consuming is reverse engineering the mathematical models used for simulation on the backend, you're going to need to observe the input->output map over a lot of possible inputs and to do that you probably need a lot of game clients connected which cost a lot of money. Not to mention not having your fuzzer detected and being banned/sued.

You might be able to eventually do a good enough job that makes your server feel quite close to the official server but then every time the game updates you either have to stick with the old way or do all of your analysis again. So you end up with a divergence between the real game and the pirate game.

It is certainly much more time consuming than figuring out how to set the "isGameValid" in securom to 1 and breaking every securom game out there. In this case you have to individually reverse engineer every single game out there, so at some point you hit the limit of the rescources that game crackers will spend on cracking. Essentially the problem with traditional DRM is that attack is easier than defence, this inverts that relationship.

Some games, WOW for example have people spending significant resources on reverse engineering them (although for different reasons) because there is a financial incentive to do it (you can make money selling rare items). There probably isn't much money in cracking simcity.

> key parts of the game logic and performed server side so you would need to reimplement the server in order to crack it

But game runs just fine for twenty minutes after disconnect (at least it simulates one city fine). You just may lose your progress because game saves "in the cloud".

In this case it's not just DRM; SimCity has social features that clearly require a server.

A social feature that isn't even relevant in solo play. You shouldn't need a network for single play.

The idea here is to force you to be reliant on the network; to move the processing onto their servers.

SimCity isn't a normal game, it's a SaaS game.

This is the problem though. Their always on DRM doesn't work, they obviously didn't have enough servers to handle the load even though they would have had sales figures before the game launched. Maybe if they launched the game with enough infrastructure to cope, there would be no issue here. As pointed out somewhere else, Blizzard have done this with their last few games; Stacraft 2 (both) and Diablo 3 and people don't complain (even though at the start there were a few teething issues).

I understand the SaaS description, it's a very accurate one but the point is customers have paid for something that is essentially an overpriced server status and countdown application. This is not acceptable, it's not like the game was given away for free. People have paid for something they cannot use.

How would you feel if you just bought a new car, drove it out of the dealership and found-out it had some amazing new technological anti-theft GPS immobiliser in your car that always needed to be connected to the monitoring company network and it was preventing you from driving it out of the dealership? The dealership refuse to refund your money even though they knew the car had this issue, but they sold it anyway so you have to wait for a fix... EA sold us a lemon, I feel very ripped off.

Do EA even test their games any more or do they just develop them, throw some really buggy DRM implementation in and sell as many copies as possible? They're already one of the most hated companies in gaming, I guess they can't really lose out any more, they hit rock bottom a while ago. Much like the music industry, they'll eventually start losing copious amounts of cash and blame piracy instead of themselves. So tired of EA's crap.

SaaS Game, very well put!

Do we need a new acronym for games that required black-box server side components (Not just DRM but entire chunks of game code) not controlled by the consumer?

GaaS? (Game as a Service)?

Ubisoft did the same thing with the last silent Hunter game IIRC.

>> SaaS Game ... (Game as a Service)

This is hardly a new concept, Eve Online is a fairly prominent example of a game that is run entirely as a subscription service. I'm sure there are other MMO games that do the same, however there are others like World of Warcraft which combine a subscription service with a "boxed" game client.

In the context of the current discussion, I think EA could have made the new SimCity game entirely service based, like Eve, giving the client software as a free download and charging for access to the server, where most/all of the actual simulation logic would live. Instead they've pulled a bait-and-switch with a boxed product.

I was excluding MMOs from the above since the expectations of those are that they are multi-player and subscription based.

When I was talking about GaaS I was sarcastically talking about ostensibly single-player games are crippled in this way.

Crippleware is a good one too :)

80% of mobile games are already precisely this, without requiring the coinage of a special term for it.

I was being facetious when I said it.

I was also thinking about PC Games mainly when I said that since it still (touch wood) a rare situation.

To be honest, mobile gaming didn't even appear in my PC-master-race-gaming mind :p