SimCity and its Scientific Accuracy

[meowza 230]

Economy, Sociology, Demography, Climatology and Urban Planning are probably just some of the scientific branches the new SimCity will be tackling.

Spore's "mistake"?

The original concept for Spore was more scientifically accurate than the version that was eventually released. It included more realistic artwork for the single-celled organisms and a rejection of faster-than-light travel as impossible. However, these were removed to make the game more friendly to casual users.

- Seed Magazine

Do you think Maxis can more or less accurately reflect real life concepts through the new game, without compromising marketability and gameplay for both rookie and hardcore players?

  Edited March 26, 2012 by meowza  

[Zelvik 0]

I dont know how scientifically acurate the models are that Maxis use to simulate something like traffic but they use underlying models that somewhat resemble what we see in real life. Considering how complex entities cities are they do a good job at representing the basic forces that make a city run - though like with the spore example they omit certain areas or simplify them to a point where its playable as a game. The budgeting process is simplified, as is administration in general.

[A Nonny Moose 8,261]

Do you think Maxis can more or less accurately reflect real life concepts through the new game, without compromising marketability and gameplay for both rookie and hardcore players?

Not if the game is to run in any reasonable time on a PC, no matter how big. Even on a huge mainframe simulating all this would never fly. There are far too many variables. Heck, even the weather guys using a big Cray system in Montreal don't run the simulation to completion. They just pause it every once and a while, do a read-out, and let it continue. And they are only simulating the weather trying to get a prediction off for Canada. I imagine NOAA does the same.

Now, if you toss in all aspects of running a city including the world economy, climate, etc. what kind of a machine do you think you'd need? Something bigger than a home appliance, yes? Just starting it up would take about half-an hour to populate the initial matrices. People have to be reasonable. PC's are a lot more powerful than the original 8086's, but they still don't have the bandwidth to handle things that are very serious without a backup to a more powerful box somewhere on the net.

That is why I favor a distributed model for this game, especially with all the features people think they want. It is going to be expensive to get all that.

We've been pampered too long by the custom content for SC4. At the time this was written a plugin suite of several GB was unthinkable. Managing such a thing boggles my mind. Some people must be just plugin collectors who don't use everything they've got, and may not even play the game.

[Catmando 7]

We've been pampered too long by the custom content for SC4. At the time this was written a plugin suite of several GB was unthinkable. Managing such a thing boggles my mind. Some people must be just plugin collectors who don't use everything they've got, and may not even play the game.

I'm definitely that kind of person. I have 6.34 gigs right now (down from 6.4 gigs last week; I've been cleaning out a few stray installers and stuff that the LEX left me), and it's unwieldy in ways I can't properly describe. I think it was mrbisonm that takes the record, though (am I remembering his name right? I always got him confused with you). Last I remember he maxed out at about 20 gigabytes and then went on a two-year organization binge and cut it down to like 12-15 gigs. That was years ago, though, so I may be remembering wrong.

What I'm actually going to start doing now, I think, is reorganizing my whole plugins folder into a series of much smaller ones that the game can actually handle: nature, city, surburban, and a folder full of all the shared files, like my terrain mod, water mod, NAM, and so on. This is probably an even bigger undertaking than all my previous organization binges combined, however, and I'm not sure it's worth the effort: I can already cut my folder by 20% just by moving the Miscellaneous Skyscrapers and NDEX folders to some safe spot, and probably upwards of 50% if all I want to do is make a nice semi-urban town. All things considered, however, at least half of the content in my plugins folder is simply not used by me. I'm a hoarder at heart.

[frdrcklim 139]

Accurate or not, there is no such thing as a right or wrong model, only one that works and one that doesn't. If Flight Simulator X was portrayed flying aircraft accurately, then I should've been a pilot by now . Games can't be too real or they lose some of their appeal. One example is Operation Flashpoint Dragon Rising. It's supposed to be a military simulator. And it is, but it was boring and it lost its appeal. Had lousy AI too.

But maybe to put it into context, I'd like to discuss Tropico 3 and 4 as an example. There is a certain depth of simulation of Economy (Industry, Tourism, etc.) , Sociology (Civil war, rebellion, coup, etc.), Demography (Important for checking the age of your workforce), Climatology (for Agriculture) and Urban Planning (Services being accessible, etc.). Of course, it's not realistic and AI is sometimes mind-scratching, but it works and it's fun.

  Edited March 27, 2012 by frdrcklim  

[croxis 150]

Humans are really good at finding patterns in things that they don't exist in. One of the dirty secrets of computer game design and programming is the use of psychology to trick the player into seeing complicated behaviors when the simulation behind it is very simple. The opposite is true too. If there is a complex simulation but only a simple presentation to the player (ie your AI has 12 emotional states but there are only happy, sad, and angry face animations) the player's internal model of the game is only going to have the simple version and you wasted programmer dev time.

The social mobility discussion is a really good example of this. If the player drops a school in a low income house area and watches some houses change to mid income. The player will apply a pattern (social mobility or gentrification), but the simulation behind it is as simple as the school sending education resources to the nearby houses, and the houses follow their unit rules that change the house and occupents when enough education resorces are stored up.

[tSlater 134]

^ And the above is why I am not concerned with their simulation approach. The simulation will be fine.