Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Locked thread
Toady
Jan 12, 2009

This game is so unpopular that pirates don't want to crack it.

Adbot
ADBOT LOVES YOU

Toady
Jan 12, 2009

OneThousandMonkeys posted:

Oh, so this is SimCity: Ron Paul Revolution Edition.

It's SimCity: Half-Assed Edition.

Toady
Jan 12, 2009

DeclaredYuppie posted:

I guess I should only be shocked they didn't figure out a way to include Nissan LEAF a couple more times there.

It gets better. It increases happiness and uses no city power despite being a recharging station. Essentially, you're obligated to accept the advertisement due to its positive effects and no drawbacks.

Toady
Jan 12, 2009

Air Julio posted:

Or you could not and your city will be just fine without it.

This is one piece I take issue with the posters in this thread about. It's free, optional dlc that happens to be sponsored. You are no more obligated to use it than you are any other ploppable. poo poo, you don't even have to dl it if you don't want to. It is transparent advertising, yes, and that it doesn't use any power is really stupid, but among this game's many issues and faults, this is at the bottom of the list.

I disagree. I think it's at the top of the list because it suggests compromised motivations on the part of the game developers and that if an advertiser pays enough money, a large enough percentage of users will plop those buildings because of their imbalanced attributes, like how enough people respond to spam to make it worthwhile. It turns the game into an advertising platform.

Toady
Jan 12, 2009

WithoutTheFezOn posted:

It really doesn't matter to me what people think about the Nissan thing, but there's a lot of wrongness in these threads about the station.

It's not "free happiness" as someone else mentioned, and while the attributes are imbalanced, they're imbalanced on the poor side.

Unless something has changed since they put it in the Nissan charging station costs $5000 to plop, has about the same influence area and upkeep cost as a $400 park while taking up a little more room, and fewer people visit it per day than that same park. And it doesn't affect land value. Efficiency/effect-wise, it's a terrible building to plop.

Unlike a park, the recharging station doesn't take power, water, or workers away from the city, and it produces no garbage or sewage.

Toady
Jan 12, 2009

Canine Blues Arooo posted:

I feel like the changes in technology just left him in the dust. Back when the most impressive game simulation was SC2K, is where I feel he peaked.

I thought The Sims was very impressive and novel.

Toady
Jan 12, 2009

KnifeWrench posted:

I'm as disappointed as anyone, but it seems like the dogpiling is getting a bit extreme. The game was fun for a good 8-10 hours, which is what a lot of $60 games with little-to-no replay value provide. The expectation of a game in this genre is that its sandbox nature will allow a lot more replayability, but that isn't strictly necessary for most gamers' value calculation, if we're being fair.

Don't consumers determine what qualifies as value?

Toady
Jan 12, 2009

CapnAndy posted:

Portal lasts for, what? 3 hours? 4? Game length has nothing to do with quality.

But it does have to do with value and price. Portal came bundled with the Orange Box.

Toady
Jan 12, 2009

kemikalkadet posted:

I'm not really sure why this is being so hyped, I guess it got publicity at the right time just as SimCity was crashing. To me it just looks like a Settlers game where you can't easily tell what each building is. Don't get mme wrong I think it looks kinda cool and I love The Settlers and similar style games, but I'm not sure why it's being hailed as what SimCity should have been.

It's the indie-developer-versus-clueless-corporation angle.

Toady
Jan 12, 2009

Broken Loose posted:

Due to a scripting error, cars will end up blinking out of existence, causing the buildings that spawned them to never get money, so they can't pay taxes, so they end up abandoned, and since the new version requires a waiting period before demolished buildings are rebuilt, completely shutting down your economy a house at a time. Apparently pedestrian-only cities are a sight to behold!

This game is loving crazy.

Toady
Jan 12, 2009

https://www.youtube.com/watch?v=hN_s7q5Dd4E

Toady
Jan 12, 2009

xzzy posted:

People put way too much value in version numbers in general. Once upon a time I did the "every release gets a bigger integer" system too and it was absurd how many nerds were convinced I was doing everything totally wrong.

As long as it differentiates between an out of date version, who cares? :iiam:

For decades, most software has followed a convention of two or three sequence schemes to classify the degree of change. Bumping the major version number for minor changes is misleading and suggests more changes than there really are.

Toady
Jan 12, 2009

Mr. Wynand posted:

I've said this previously in this thread, but no, this is not the case. For a continuous release process, this is exactly what you do - bump the major version number each time (bump the minor version only for hotfixes).

The majority of software really doesn't follow such a scheme. Usually, the major version number is bumped for big changes and the minor for small patches. However, these updates aren't feature releases; they're just bug fixes.

quote:

This whole marketing conspiracy angle is stupid and unlikely.

A public "version number" is a marketing name unrelated to the actual internal version number, so I don't see why it's so far-fetched that they might feel obligated to imply that major changes have gone into the game since the disastrous launch.

Toady fucked around with this message at 00:53 on May 24, 2013

Toady
Jan 12, 2009

Mr. Wynand posted:

Because it's weird. Who the hell bases their purchasing decision based on version numbers?

I get the implication that maybe they think seeing bigger numbers on the update page might lead some people to believe that more fixes have been performed then really have been, but what I don't get is who this fictional customer is that is informed enough about the game to know that it needs lots of fixes but not informed enough to actually read the patch notes that go with said number. Sure the marketing team might just be dumb about this but again, it's just a weird thing to be dumb about. Dumb marketing dudes make dumb marketing dude mistakes like trying to push a mega-lame facebook participation campaign, not futz around with version aesthetics.

It just seems so far fetched compared to the much simpler theory of "they said they wanted to do continuous releases, they appear to be doing continuous releases, they picked a versioning scheme that is commonly used with products doing continuous releases". Both Chrome and Firefox have very pushy auto-updaters (not unlike Origin) and rely on... surprise, continuous releases. Yes this isn't really done for games but it is hardly a stretch for a game on an auto-updating platform like Origin or Steam.

This is just my impression as a guy that has sat in on a lot of stupid meetings where everyone argues about the versioning scheme. I don't think I've ever heard of a marketing guy even being present for such things...

I don't think anyone has suggested that people are basing their purchasing decision on SimCity's version number. I think it was more of a passing observation that when people read on a game website that version 4.0 of SimCity is out, they're going to assume far more changes than they would if it was 1.04, and that could have been a conscious choice on the part of Maxis. As in, it was an amusing example of another misleading thing about SimCity, albeit a trivial one.

Most software with continuous updates doesn't bump the major version for every patch. I'd feel dishonest if I called my app 2.0 for a bugfix patch, not only because that's not the conventional versioning scheme I'm used to but also because I know that when people see 2.0 on the iTunes Store, they'll assume I've made major changes.

Toady fucked around with this message at 01:07 on May 24, 2013

Toady
Jan 12, 2009

WeaponBoy posted:

People getting angry about version numbers is definitely one of the more entertaining moments of :goonsay: I've seen in a while. This thread goes to some insane lengths to hate SimCity sometimes.

Except nobody's angry? :confused:

Toady
Jan 12, 2009

triplexpac posted:

Saw this on Reddit: https://www.youtube.com/watch?v=5bZcMi5zasM

Simcity 64DD. I've never seen it before, looks pretty neat.

Interesting to note that SimCity 3000 was originally supposed to be 3D like that and let you go to street level to talk to citizens, but it was scrapped for not being up to the visual standards of the PC.

Toady
Jan 12, 2009

eXXon posted:

It sounds like at rush hour, each vehicle picks their destination before they leave. So... why the gently caress didn't they think of this, like, a year ago? Why can't specific workers have their own homes and workplaces now? These and other questions answered never because who cares anymore.

The wording is "far in advance", so I don't think it's happening before they leave. I wonder if agents are just looking some number of intersections ahead now and pre-filling a sink before they actually reach it.

Toady
Jan 12, 2009

Fallom posted:

Isn't this exactly what some goons were saying was impossible with the agent system or far too computationally difficult?

That was in response to the suggestion of having every agent perform individual pathing and maintain persistent destinations. I don't know the details of this change, but I wonder if it's letting vehicle agents look ahead some number of intersections to fill a sink before they actually reach it, so that the conga line breaks up a little sooner.

Toady
Jan 12, 2009

Happy Noodle Boy posted:

Didn't someone here suggested that fix (pre-determine jobs/homes at time of departure) months ago?

I think they gave agents the ability to look ahead some number of intersections in the graph and fill sinks early, thereby updating the intersection weights earlier to help break up the herd of agents. I'm interested to know how well this would fix things.

eXXon posted:

There's no technical reason why they couldn't store a workplace for every sim, especially given how small most cities are. Even having 2-3 potential routes stored between every sim and their workplace wouldn't be that demanding.

The technical reason is that predetermined locations for each agent would require individual routing of potentially hundreds of thousands of agents through dynamic traffic conditions. Perhaps a hybrid statistical approach would have been better, using agents only for visual feedback, like in SimCity 4 but without fading in an out of existence.

Toady fucked around with this message at 22:27 on Aug 9, 2013

Toady
Jan 12, 2009

eXXon posted:

Huh? Agents are already being individually routed during every rush hour, and the patch notes make it sound like they will adapt to traffic conditions as well. Storing a few potential commutes for every resident would require an extra few tens or hundreds of MB of memory usage, but it can't possibly be more demanding than re-calculating routes every day like the sim does now.

Agents have no data regarding their destination. The game steers them using weighted values calculated for each intersection from the available nearby sinks (empty slots in buildings). When agents fill a sink, the weights are updated after a short delay. So, the agents head in a conga line to each building because they're following the same map of available sinks and have no knowledge of where they're headed. It's like a pachinko machine.

Toady
Jan 12, 2009

quote:

The official EA forums are overrun with Mac gamers desperately trying to get their purchases to work correctly on computers ranging from brand-new Retina MacBooks to iMacs that are just a few weeks old. I've been told by one player that in order to even download the game, he was forced to reinstall EA's Origin download application three times.

gently caress yeah, awesome.

Toady
Jan 12, 2009

Eric the Mauve posted:

That was my original point: Civ V was released before it was ready and with obvious features (like religion) stripped out for the express purpose of selling expansion packs.

That's a strong accusation to make without evidence. I've never felt like Firaxis was the kind of company to rip people off, especially with the level of support they give the Civ games for years after release.

Toady
Jan 12, 2009

Kyrosiris posted:

I see no one on Twitter throwing this back in that clown's face and that makes me sad. :(

Several people are calling him out on it. His response is that, while he didn't claim offline mode was impossible, "clearly, I shouldn't have opened my mouth".

Toady
Jan 12, 2009

Combat Pretzel posted:

So how can this guy do 51000 cars doing their (ostensibly) own pathfinding on a huge roadgrid on a single thread, and Maxis can not?

In fairness, that guy's JavaScript game doesn't exist yet.

ToxicSlurpee posted:

From a computer science standpoint it really depends on what algorithm they're using. I'm going to go ahead and guess that not only were their agent algorithms really loving stupid but also poorly optimized and didn't get much time put on them. It's certainly possible to do a lot with agent or agent-like systems, look at games like Tropico, Caesar IV, and Dwarf Fortress. However, if your algorithm sucks it's going to eat more processing power.

The agent system was explained at a GDC 2012 presentation. To lower resource costs, a lot of work was put into making the agents lightweight at the cost of making them dumb. They aren't aware of their destinations and are steered along a weighted decision graph that directs them at each branch in the graph (a traffic intersection in SimCity) based on the capacity of nearby sinks, like balls falling through a pachinko machine. Because agents of the same type are following the same graph, and there's a delay when the graph updates to reflect sink availability changes, agents move in clusters and fill up sinks one by one instead of heading out in scattered directions like they would if they were real people driving to different destinations.

Toady
Jan 12, 2009

YourAverageJoe posted:

What I keep getting from this thread isn't just that agents are a bad idea, it's badly optimised for the wrong reason. Imagine if a sim had a house and a job. They take the same route every day because the destinations are always the same unless something changes dramatically. This would mean that you could easily cache that route to disk. Not only that, since SimCity encourages you to make neighbourhoods, you have multiple agents taking the same routes; you could break routes up between zones and have multiple agents reference them in their routes. Of course, a system that cross-references path caches like that might be difficult to make, but might not be all that taxing on hardware.

Individual pathfinding for hundreds of thousands of agents would be expensive, especially in a dynamic system with traffic congestion, constantly changing zones, and an infrastructure that can be modified by the player at any time.

Toady
Jan 12, 2009

Dark_Swordmaster posted:

That's why you assign a job and a house, just like real life. Occasionally they might move, but you just path ONCE a-to-b and then let traffic laws dictate how things flow. You aren't continually pathing, just having natural traffic laws dictate how it goes.

Assigning unique destinations is individual pathfinding. The agents would have to recalculate their paths in response to traffic conditions and other dynamic elements.

YourAverageJoe posted:

Those hundreds of thousands of agents can easily be grouped by the common routes they usually take, and while the infrastructure can be modified at any time, it usually isn't, as far as an agent is concerned. Because social mobility isn't modelled and companies don't change buildings often.

What does "grouped by the common routes they usually take" mean? How would you determine those routes without unique pathing?

Instead of forcing a square peg through a round hole, the agent system should have been used for visual feedback while relying on traditional methods for the simulation. This might mean you can't follow Peggy the Sim on her daily commute to work, but who cares if it means you can build New York City?

Toady
Jan 12, 2009

YourAverageJoe posted:

Let me break it down for you:
There are fewer roads than there are agents.
There are fewer residence buildings than there are agents.
There are fewer places to work than there are agents.
There are fewer neighbourhoods than there are buildings.
Furthermore, the game encourages you to group buildings together into neighbourhoods, ensuring there's fewer of them than possible.

As the game is now, agents already share routes. It's statistically inevitable. Not only that, they each re-calculate their routes every time they spawn (when going to-and-from work). This is wasteful and unnecessary. In a system where agents have dedicated residences and employment, the routes they take to-and-from work will be similar, sometimes identical. Make the agents remember their routes, with the possibility to divide those routes into segments that are called upon in succession. Then make a system that detects similarities in those segments between agents, and a group of agents with similar routes can replace the route segments they have in common with pointers to a single, identical route segment. This saves memory.

Also, you assume that SimCity ever did pathfinding accommodating for traffic. It doesn't.

Like DarkJC said, agents don't calculate routes or have knowledge of their destinations. Instead, each intersection weighs its attached segments according to available nearby sinks, and the agents are steered to the most appealing segment when they reach that intersection. It's like marbles falling through a maze of tubes, steered by gates. This is the reason agents are so dumb and why they move in herds, filling each sink one by one. I'm not sure what a route-combination system would accomplish. It would be much more complicated and expensive than you're making it sound.

The problem with the notion of persistent routes is that SimCity is dynamic enough to negate any advantages. If a thousand sims decide to take the same road to work, and they all congest it, you have to recalculate again to determine alternate routes. Along with changing zones and player modification, there would be constant adjustments by the sims to the changing environment. If you don't allow the sims to respond to changes, then you have the problem of dumb sims again.

Agents shouldn't have been the basis of the simulation. They should have been visual flair, reflections of the state of the city but not beholden to accuracy, like the pedestrians and vehicles of SimCity 4.

Toady fucked around with this message at 02:33 on Mar 11, 2014

Toady
Jan 12, 2009

That page looks like it's describing HPA*.

Adbot
ADBOT LOVES YOU

Toady
Jan 12, 2009

This was the Emeryville studio. The Walnut Creek Maxis that people are reminiscing about was closed in 2004. From a former employee:

quote:

These laments are a bit overblown.

I worked for Maxis during the big closure in 2004, when the company was in Walnut Creek. A small group moved to Emeryville, many people (including the Sims teams) moved to Redwood Shores, and some some people didn't make the transition. Think of it like layoffs, following on the colossal failure that was The Sims Online (although, ironically, what's left of that team had already moved to EARS).

This is pretty much the same thing. Spore was a disappointment, SimCity was never a massive moneymaker, and whatever games they've been working on since got cancelled. The Emeryville shop has had a decade to produce a hit and it hasn't worked out. Some folks have already been laid off. Some folks will get moved to EARS. Some folks will be looking for a new job. The cycle continues.

It's unlikely that any of the individuals responsible for the specific versions of whatever old game you fondly remember playing are still around. At the end of the day, what you're lamenting is the loss of an office building and a logo.

  • Locked thread