[Sandwich]

Anyone have one started, or interested in starting one?

[General Battuta]

Please be warned that you should never buy this terrible and intrinsically broken game.

[NGTM-1R]

Anyone have one started, or interested in starting one?

No.

[Oddgrim]

Please be warned that you should never buy this terrible and intrinsically broken game.

It's not a bad game per say, game mechanics wise and the multiplayer addon is interesting.  But on the other hand : Extremely badly implemented DRM(always online connection) a lot smaller city plots, launch DLCS (and more to come I bet) so buyer beware!

[General Battuta]

The game mechanics are atrociously mishandled - an all residential city is somewhere close to the optimal build, and even a nominally well-designed city can arbitrarily fail due to agent pathing problems. The programmers and designers didn't understand the challenges of their own design and ultimately were unable to implement it. It just does not work on any level: it will grow less fun and less interesting the more time you spend with it.

[Spoon]

Please be warned that you should never buy this terrible and intrinsically broken game.

Indeed.
Please do not give EA any more money by support this broken product.

[BrotherBryon]

Please be warned that you should never buy this terrible and intrinsically broken game.

Indeed.
Please do not give EA any more money by support this broken product.

Too late I already did, pre-ordered it after playing the beta. I thought it was going to be a good game (I played the heck out of the beta and had a blast) and then launch day came around and I couldn't play the game for 4 days. Then when I could I felt kind of disappointed, the regions small size makes it really difficult to grow the city after a certain threshold and problems become nearly impossible to manage with larger populations and no real-estate to work with. I might give it a try again once they work out some more of the bugs.

Got a free game over the launch debacle but didn't particularly care for the choices. I pretty much had all the games I would have been interested in so I ended up getting Dead Space 3, haven't even bothered to try it yet.

[Phantom Hoover]

And also their decision to simulate individual Sims and work upwards from there has to be one of the most basic mistakes they could possibly make. I expect that kind of naïveté from a 15-year-old hobbyist who's just read A New Kind of Science and expects that ~emergence~ will make everything work out, not from Maxis of all people.

[General Battuta]

I have a friend who studies computational complexity at Google and yeah that's pretty much exactly what she said - that they should have realized the game was broken much, much earlier in the design process.

I think the Eurogamer review is pretty fair to it.

[Phantom Hoover]

It's Mass Effect 3 all over again: people take the game's most horrendous failing and hold it up as the single fatal flaw in an otherwise well-made work, when actually it was completely broken from the start.

[The E]

And also their decision to simulate individual Sims and work upwards from there has to be one of the most basic mistakes they could possibly make. I expect that kind of naïveté from a 15-year-old hobbyist who's just read A New Kind of Science and expects that ~emergence~ will make everything work out, not from Maxis of all people.

From what I've read the basic approach to handle the simulated population using an agent model isn't that bad, it's just that the lack of persistence makes it all somewhat moot. If the individual sims were to remember their jobs and homes, I think the fundamental issues with the pathfinding would disappear at the cost of increasing computational complexity. The completely abstracted model they used here (and why in gods name did they use the same broken system for sewage and electricity???) just isn't fulfilling the goal of making a living city (much less the living town they were apparently aiming for).

[Sandwich]

And also their decision to simulate individual Sims and work upwards from there has to be one of the most basic mistakes they could possibly make. I expect that kind of naïveté from a 15-year-old hobbyist who's just read A New Kind of Science and expects that ~emergence~ will make everything work out, not from Maxis of all people.

From what I've read the basic approach to handle the simulated population using an agent model isn't that bad, it's just that the lack of persistence makes it all somewhat moot. If the individual sims were to remember their jobs and homes, I think the fundamental issues with the pathfinding would disappear at the cost of increasing computational complexity. The completely abstracted model they used here (and why in gods name did they use the same broken system for sewage and electricity???) just isn't fulfilling the goal of making a living city (much less the living town they were apparently aiming for).

Do you have any links to articles or research about those issues? I'm curious.

[Veers]

I read this as a HLP religion for Simy City.

[Phantom Hoover]

The thing is, even if you fleshed the Sims out and gave them persistent jobs and homes and all that, debugging the game would still be a nightmare. As anyone who has played Dwarf Fortress can tell you, detailed low-level simulations like that can lead to all sorts of totally insane things happening, and trying to tune city-wide behaviour to the standard expected of SimCity would be nigh impossible.

[The E]

Oh, certainly, but I would argue that DF with its ultra-intricate simulation of the dwarven condition overshoots the target for SimCity by quite a margin. And yes, more detailed simulation will lead to stranger emergent behaviour, but that's something that should and could be dealt with during development. Hell, it might even be fun!

My opinion is basically that if you're gonna go and do an agent-driven simulation, you might as well do it properly and not stop halfway through.

[IronBeer]

The thing is, even if you fleshed the Sims out and gave them persistent jobs and homes and all that, debugging the game would still be a nightmare. As anyone who has played Dwarf Fortress can tell you, detailed low-level simulations like that can lead to all sorts of totally insane things happening, and trying to tune city-wide behaviour to the standard expected of SimCity would be nigh impossible.

Not to mention the computing power involved. A full 200-dwarf fortress can bring a lot of rigs to their knees.

[Apollo]

The thing is, even if you fleshed the Sims out and gave them persistent jobs and homes and all that, debugging the game would still be a nightmare. As anyone who has played Dwarf Fortress can tell you, detailed low-level simulations like that can lead to all sorts of totally insane things happening, and trying to tune city-wide behaviour to the standard expected of SimCity would be nigh impossible.

Not to mention the computing power involved. A full 200-dwarf fortress can bring a lot of rigs to their knees.

Damn... I wouldn't of thought it was possible, considering how crude Dwarf Fortress's default graphics are.

[The E]

It's not the graphics that will bottleneck DF. It's a game that is exclusively CPU-bound.

[CommanderDJ]

Out of curiosity, is DF multi-threaded?

EDIT: Apologies for further derailing the thread.

[IronBeer]

Out of curiosity, is DF multi-threaded?

No. Dwarf Fortress is single-threaded and essentially held together with the coding equivalent of bubblegum and duct tape.