People are upset about the lack of a single player offline mode, which absolutely is about preventing unlicensed copies. There is no technical reason for its absence.
The cloud based stuff sounds great, and if it were an optional extra that was unavailable at launch due to technical difficulties, I doubt you'd see a hundredth as much bad PR.
I see this sentiment expressed pretty often, but in this case it seems flawed. From what I understand, you can't build a city that does everything, instead you rely on the other cities in your 'region.'
The new Simcity is no more single player than World of Warcraft. The previous games were single player, as they focused on a monolithic city. This game is designed from the ground up with the assumption that you will play with other people. There's no technical restrictions preventing Blizzard from offering single-person servers for WoW, it just doesn't follow from the game design. The same is true here.
WoW relies on active human beings to make its world lively and addicting. I don't think that need is as prevalent in a sim game, where the game AI itself can...well...simulate the regional qualities that would affect a player's city well enough.
How is this any different from a single-player RPG where you control a party of characters which all depend on one another's unique abilities?
The previous title in the series quite happily did this all locally. Log in and be presented with the area and all the city locations in a grid, pick one and build.
> We can argue about whether these things should be computed and stored in the cloud, but I think there is a plausible, if imperfect, argument for doing so
Conceivably true? We could be in a world with no DRM, but there would still be a market trend moving games towards a more social, connected and integrated experience; and some developers will be looking at ways to fundamentally build their new games in a way that takes advantage of this. Why not SimCity? The Sims 3 already indicated where the Sims franchise might be heading with regards to improved connectivity with friends, I don't think it's an entirely unreasonable leap to get to where SimCity is now.
Which isn't to say that DRM wouldn't have been a business consideration at an executive level; but there's definitely an argument that SimCity's design choices, even if flawed, can be justified without considering DRM.
If the problem was "going social", you'd see more p2p and lan-gaming features, rather than "let's all connect to my servers, creating huge scalability problems which may or may not be solved at some point".
this is not preposterous at all. if all the simulation happens on the server side, its not necessarily trivial to have that run on the client. maybe the server requires unix. maybe the server requires an oracle database.
That said, you can imagine a situation where doing these computations locally did compromise the experience for a certain spec of machines. Maxis has already explained that city size limits were imposed because the simulation didn't run sufficiently well on the low-mid spec machines above the allowed size.
As the owner of a higher spec machine, I find it somewhat silly that this couldn't just be a preference item (as AA and resolution are), but I can understand the decision to cut down the computational demands to keep gameplay smooth.
Games that have launched with framerate issues recently have met a huge amount of flak--see what's going on in the Planetside 2 community. You can argue that what's happening to PS2 is far worse: they released a poorly optimized game, and those issues prevented people from playing for months.
I find it hard to believe the game is so complicated a laptop will struggle, but EA can efficiently load it onto servers? That sounds pretty hard to believe. Even if the target player is using a crappy Netbook, you're talking about a dedicated compute server for every 4-20 players. That's insane.
Intel players will need, minimally, a 2.0 GHz Core 2 Duo, while our AMD players will need at least a an Athlon 64 X2 Dual-Core 4000+.
Those sound pretty capable of a significant amount of simulation themselves...
But beyond the "feather that broke the camel's back" rationale for offloading compute, keep in mind that there are issues around shared data that would vastly improve performance of region-level computations in the cloud. If the region state is maintained in the cloud, any computations on that state will be far more quick when done in a centralized way.
Further, there are fewer challenges around handling client desynchronization, which is a classic problem in large scale strategy games. Reconciling one or more desync'd client's state can get very difficult. Under a centralized system, the master copy reigns.
Obviously there are downsides, but it's not like it's a crazy idea to offload some of the compute to the cloud.
As far as the multiplayer, I don't think anyone has a problem needing servers for multiplayer.
All I'm saying is that the suggestion that computations need to be done online to save local CPU, even on a singleplayer game, is obviously ridiculous.
When you centralise you also have other advantages, such as being able to cache the results of previous computations. So if you can structure the computations in such a way as to be fully or partially reusable you can memoize and scale much better.