SimCity 5 Catch-All - Offline mode coming in next update

Redwing wrote:

I'm mostly okay with that. If I can start up a small region and specialise 3 or 4 cities, I could see myself having a lot of fun with that.

For me, I had fun with 1 city. Having to start over in the second part of the region and do basically the same stuff again with the same city size restrictions just didn't seem like fun to me. Still - it's fun during the development of that first city, and if you get it at a discount like I did, I feel like that's a good value proposition.

I don't think I ever went into the red any time I've played the game.

If you are working a mining/drilling zone, it's impossible not to be in the red once you get to the advanced end of the tree ?!?!
You are still profitable though through exports.

Redwing wrote:
MoonDragon wrote:

This design decision was also why they never bothered to do a single player game. Because it makes no sense. You NEED other cities to be able to progress.

I'm mostly okay with that. If I can start up a small region and specialise 3 or 4 cities, I could see myself having a lot of fun with that.

This was also the design ethos behind SimCity 4, except that one actually gave you plenty of room for individual cities. However, you needed to work the region system in order to get the biggest populations and some of the rewards were gated behind some regional statistics.

Last time I played SC4 a few months back I finally started to embrace region play and it wasn't terrible, but it's a massive mindset change from earlier games. Which is true of a number of things in SC4 actually.

shoptroll wrote:
Redwing wrote:
MoonDragon wrote:

This design decision was also why they never bothered to do a single player game. Because it makes no sense. You NEED other cities to be able to progress.

I'm mostly okay with that. If I can start up a small region and specialise 3 or 4 cities, I could see myself having a lot of fun with that.

This was also the design ethos behind SimCity 4, except that one actually gave you plenty of room for individual cities. However, you needed to work the region system in order to get the biggest populations and some of the rewards were gated behind some regional statistics.

Last time I played SC4 a few months back I finally started to embrace region play and it wasn't terrible, but it's a massive mindset change from earlier games. Which is true of a number of things in SC4 actually.

You left out what I found to be the most interesting part of RW's comment:

Redwing wrote:

Having said that, it'd be great if the regions sprawled a bit, perhaps with small towns building up in the areas between city plots, until it all fills in to one big jumbo city with specialised sections/suburbs/districts. Almost like some sort of Coruscant simulator.

While I'm not sure I want a jumbo city, it would be nice if the regions sprawled. That seems like the next iteration to me, not squares of city after city, but a more organic plotting.

One thing that almost all city-builders do right is rural areas. In early Simcity games, farms just took up valuable space that could be used for MOAR ARCHOLOGIES! And in the latest iteration, there's just no hope of anything other than a square urban city.

I want to build a mega city that encompasses everything from a downtown economic hub, all the way out to the sprawling countryside. At most, I would want to have a regional style, but all on one screen, just linked by roads. I doubt that will be available for some time, however.

omni wrote:

One thing that almost all city-builders do right is rural areas. In early Simcity games, farms just took up valuable space that could be used for MOAR ARCHOLOGIES! And in the latest iteration, there's just no hope of anything other than a square urban city.

I want to build a mega city that encompasses everything from a downtown economic hub, all the way out to the sprawling countryside. At most, I would want to have a regional style, but all on one screen, just linked by roads. I doubt that will be available for some time, however.

You and I are, like, the same person.

omni wrote:

One thing that almost all city-builders do right is rural areas. In early Simcity games, farms just took up valuable space that could be used for MOAR ARCHOLOGIES! And in the latest iteration, there's just no hope of anything other than a square urban city.

I want to build a mega city that encompasses everything from a downtown economic hub, all the way out to the sprawling countryside. At most, I would want to have a regional style, but all on one screen, just linked by roads. I doubt that will be available for some time, however.

Someone on Reddit did a Banished town that was like that. It was nifty.

Yeah, that's the idea I went with for my larger Banished towns. Still doesn't scratch the itch of doing that kinda build, but with water, electricity and traffic in the mix

Yeah, now that would be a really cool Simcity, especially if you started out low-tech, and worked your way through the historic development of cities, or at least a vague, gamey approximation of same.

My mental model for a 2014 SimCity is being able to model the Bay Area with at least some fidelity. Not down to like, individual streets or anything, but something of that size/scale. A modern computer really should be able to do something like that, at least if they're batch/parallel algorithms, instead of the weird agent model they chose.

Yeah, it would probably take a 64-bit program running on a computer with at least 8 gigs, and 16 would probably be better, but wouldn't it be way cool to be able to start small, and just keep zooming out and out and out?

Out and out, as well as changes in building type and facility type as time goes by. I'm not familiar with the Bay Area, but a city like Edinburgh has gone through so many changes and challenges moving through the hundreds of years it has been around, over a thousand if you scale back what you call a city, and around 10,000 years if you go all the way back to initial settlement. I'm not saying you should be able to go back that far, obviously that scale will still allude our current computer tech/power to be done right (and would be very slow to begin with), but some more historical -> current -> future technology progression, while sticking with the city building aspect only (rather than any form of RTS), but on a scale large enough that you need the required diversity of urban/rural living. Old cities are forced to knock-down, rebuild and renew, purely due to age, or changes in lifestyle/requirements of the inhabitants.

Clearing areas in SC5 just doesn't feel like that, though. It feels like you're just trying to build a uniform square city with equal-height skyscrapers filling that square. Anything less than that is 'wasting space'..

One can dream

Well, I just looked it up, and the Bay Area is bigger than I thought, comprising about 7.1 million people; it goes a fair bit further north and south than I'd realized. (Sonoma County, for instance, is considered part of it, when I don't think of it that way at all, and it is quite large.) Modeling that much population and land area on even a monster PC is probably not going to work very well.

Maybe just San Francisco and San Jose might be more reasonable: it's 2797 square kilometers, or about 1100 square miles, with a lot of fairly empty space between the two cities. London is actually smaller than that, though far, far more densely populated, and if you could roughly approximate London and its environs, I think that would keep people busy for a long, long while.

Would that actually be possible? I really don't know. But a 64-bit program has easy access to gobs of memory, and I strongly suspect you could do some fairly amazing stuff. And Intel says they'll be shipping octa-core processors on DDR4 late this year, which would help more than a little.

Anyone played Cities of Tomorrow expac? Wondering if its worth $10

I haven't played it, but from hearing about it on various podcasts, it sounds like it positively received and purchase-worthy.

I bought it. In some ways it breaks the game because you can raise your towers and circumvent the real estate problem. However if you deeply hated the original it probably won't change your opinion.

Budo wrote:

I bought it. In some ways it breaks the game because you can raise your towers and circumvent the real estate problem. However if you deeply hated the original it probably won't change your opinion.

Breaks you mean make it worse? I actually did not hate the original. It was just limited.

Balthezor Yes, and it is fun. The way I play is to have zones with zero future or 100% future, and that works really well.
The mall levels on the mega-blocks are still seriously nerfed though.
I think Budo means that the tower blocks give a huge population boost. They also do nothing for your taxes as they are a contained unit.

Pretty much that. Also you can develop vertically now and not just horizontally. You have more of a 3d version of your real estate.it doesn't make it worse though, it's actually an elegant work around.

Not sure where to put this, but the 25th anniversary of the first game came and went twice (February for Mac/Amiga, October for PC) with nary a word from EA or most of the gaming press. Which I think is really depressing considering how much time I spent with SimCity Classic and SimCity 2000 in grade school.

https://games.yahoo.com/blogs/plugge...
https://medium.com/re-form/simcity-t...
http://www.gizmodo.co.uk/2014/10/25-...

EDIT: Ok, so Polygon had an op-ed that at least mentioned it.

Cross-posted from the City-building Game Catch-All:

So, I'm finally giving serious consideration to Simcity 5. I was so stoked for it prior to release that I'm only now getting the foul taste of how EA twisted the game into a shadow of it's potential self.

What is the consensus with you folks of what the game is now? I read the last year's posts in the Catch-All but it's been very quiet over there. Obviously the addition of single-player is a plus, but it sounds like the small regions are still a factor. I'm not sure if that will bother me too much. A multi-region system that I entirely control could be fun too.

As long as the major systems (e.g., water, sewer, money) are solid I think I could enjoy it.

But then, my brain keeps telling me to just wait for Skylines and to not fixate on the Simcity name. Also, it seems to currently be $30 pretty much everywhere. Think it's worth that much at this point?

Hmm. That's discouraging. I was hoping that after so much time, it would be straightened out more.

I think I'll wait for a sale.

Don't get me wrong, I haven't played in a long time. But ironing out a few bugs and removing the online-only requirements didn't fix what I saw as the major issues of the game anyway. It may indeed be better/smoother now, but I just can't deal with running out of land an hour or so into a new city.

D-Man777 wrote:

Hmm. That's discouraging. I was hoping that after so much time, it would be straightened out more.

I think I'll wait for a sale.

Above and beyond the bullsh*t that was jammed down our throats to serve as DRM, the fundamental design of the game is not workable: they're trying to model every citizen, and as we talked about in the agent-based sim thread, that doesn't scale very far. Typically, the upper limit on that sort of game is a few hundred agents, and Simcity was trying to do many, many thousands of them. This meant they had to be impossibly stupid and totally abstract. (and then, of course, EA lied about individual people existing and being tracked, like with so many other things they lied about on that game. You could tell when they were lying about Simcity by whether or not they were talking.)

The model is just broken. The only reasonable way to do that kind of a game is with abstraction and blocks, not with individual agents. They needed to do something like Simcity 2000, scaled up: computer power hasn't advanced anywhere near far enough to do what they wanted. Barring some fundamental advance in physics, it's quite likely that consumers will never be able to buy machines fast enough to run that model at any kind of real scale.

edit: to be fair, however, I blasted them at the time for restricting city sizes so that they could charge more. Most or all of the OTHER restrictions were so they could charge more for it later, but probably not city size.

D-Man777 wrote:

Hmm. That's discouraging. I was hoping that after so much time, it would be straightened out more.

Speaking as an outsider on SC5, this played out basically the way I expected it to. Once the sales dried up (and presumably Cities of Tomorrow didn't sell well) EA pretty much dropped support for the title, much like what happened with SimCity 4 + Rush Hour. EA doesn't support their titles indefinitely unless they have a sustainable sales model (see The Sims and it's plethora of DLC, "Stuff" Packs, and Expansion Packs). So while they did put out an offline mode, it was too little too late to keep the game afloat.

At this point I would grab it for $20 or less and only then as a curio in the franchise.

Malor wrote:

The model is just broken. The only reasonable way to do that kind of a game is with abstraction and blocks, not with individual agents. They needed to do something like Simcity 2000, scaled up: computer power hasn't advanced anywhere near far enough to do what they wanted

What's funny about the whole thing is that they had a perfectly workable model they had done for the past two decades. Why they decided to change it to agent based (and then using multipliers on the agents once the population passes a certain threshold) is probably one of their most bizarre decisions ever.

That's interesting, and quite believable. I just finished, or perhaps a better word is abandoned, a game of Banished that exceeded 900 people--who ARE modeled individually. My PC is no powerhouse, but the game became increasingly choppy beyond that number. That EA was even trying to do that on a city-size scale seems foolish.

Actually, I'm kind of impressed that you were able to run 900. The Banished dev must write very tight code.

Malor wrote:

Typically, the upper limit on that sort of game is a few hundred agents, and Simcity was trying to do many, many thousands of them.

It's compounded by the fact that a lot of these "agents" are taken up by stupid stuff like electricity and waste. These things are better abstracted and the agent simulation just does them un-intuitively at best and wrong at worst.

For example, connecting to the power grid doesn't immediately power your building, instead it causes a power packet to be sent to your building from the reactor and when the packet arrives then the building has power. It's very much like Banished, where the building gets fire wood or whatever and saves it until it runs out and requests more. This is not how power works and it's just wasting agent processing to manage it like that.

Frahg wrote:

For example, connecting to the power grid doesn't immediately power your building, instead it causes a power packet to be sent to your building from the reactor and when the packet arrives then the building has power. It's very much like Banished, where the building gets fire wood or whatever and saves it until it runs out and requests more. This is not how power works and it's just wasting agent processing to manage it like that.

I've been following SimCity with interest because I've wanted to build agent-like sims and it's an instructional failure. This is a really good example of why: the plus side of agents is that you get emergent effects like a broken bridge automatically affecting traffic flow across the network; locality and position and individual units can matter. But sometimes they're the wrong abstraction, and they're prone to unpredictable strange attractors.

Which is fun in Dwarf Fortress, because you can examine each step in the chain and discover why your fortress is inevitably doomed this time, and less fun in a less accessible system where things that are supposed to be stable mysteriously aren't.

I think that Sim City 5 gave a fairly bad rap to the idea of agent-based games used at this level. The basic idea of using agents, and then after you hit your agent limit being able to have enough agents to use them as samples for more statistical based modeling, is a completely valid strategy. (Assuming that you can manage enough agents to actually make statistically valid output at the population limits of the game) It failed, IMO, purely based on execution. The agents were so stupid that they modeled a city and its populace very poorly, so that the optimal strategies that were developed didn't look like cities any more. In addition, the scaling once you hit the limit wasn't done very well. It just started multiplying the population it showed you by an invisible modifier, but no other logic was changed.

As an example lets say the agent limit was very low, so the scaling started really low. You put down one house of 8 people, you would have 8 agents and 8 population. If you put down a second house you would have 16 agents and 30 population. Instead of allocating it's agents better it has created a system where a town that is supposedly 5 times bigger plays exactly like a town twice as big, the feeling of growth and development has been sabotaged. If they had been more sophisticated they would have done the reverse. Take away agents from the first house, so the first house has 5 agents even though it adds 8 population, then put 5 agents in the next house. Now you have 10 agents mimicking 16 people.

You then scale up the traffic, waste, job, tax, etc, etc of those agents by 1.6. Do you have 1 unemployed agent? That's 1.6 unemployed sims, etc, etc. The downside is that you can no longer go to a house and be guaranteed to have 8 meaningless people there to follow around all day. The upside is that your city can organically grow and grow and keep changing in a much more realistic way. And that upside is far more important. A city that is 5 times larger actually has 5 times the buildings, the traffic, the power, etc, etc, etc, at the cost of decreased granularity, which is fine, as long as your granularity is still good enough to populate an effective statistical modelling.

And yes, wasting their agents on electricity and sewage packets was ridiculously stupid. A much, much, much more complex way of modelling something way worse. I don't know if they've fixed it, but at one point you had to over produce electricity, etc, etc, because all the electrons would bounce around the city in their stupid drunk walks and only occasionally make it somewhere that needed them.

Yonder wrote:

I don't know if they've fixed it, but at one point you had to over produce electricity, etc, etc, because all the electrons would bounce around the city in their stupid drunk walks and only occasionally make it somewhere that needed them.

I fired up SimCity today and nothing has changed. Packets will still get consumed by the first building that needs it causing you to have some buildings with a ton of power/water and remote buildings with no power/water instead of uniform distribution. And the solution is to overproduce, like you said.

The always online or 'multiplayer' aspects didn't hugely bother me. It's just the crippling city size that really restricts what you can do with a city. In previous iterations, I could have agricultural areas far out of the city, with small suburbs before hitting the main city area, all in one 'map'. And that was without getting into the multi-region deal that SC4 introduced.

Playing as Supreme Ruler of a whole region in SC5 might get towards that same feeling in some respects, but it is still disjointed and far removed from my expectations from a city builder or the scale of previous Sim City titles.

In short, I got about 20 hours out of it, at launch, before finally giving into just how restricted it is. I also disliked the SuperHero nonsense that was involved and required completing tasks as both the hero and the villain. Get out of my City Simulation!

Looking forward to Skylines with fingers well and truly crossed. Please, please don't balls this up.

As for SC5, if 20 hours or so for $30 represents value to you, you may get some enjoyment out of it. The graphics are nice, and the representation of power and waste/water is well done, imo. Last time I played there were some issues with 'convoy-ing' service trucks that would bunch together heading to the same target, meaning they weren't servicing other targets in the city (this was true for fire, ambulance, garbage collection).

It is very pretty, it's just too small scale.

Edit: fixed typo.