Firaxis: Corruption Breakdown

[Yeti]

In regards to folks that want a smarter / better / tougher AI:

I don't agree that "just go play Demigod, Deity, or Sid level" is a good response.

Those levels are artificially tougher because the AI has been given unfair advantages (extra units, production bonuses, bonuses during trade, etc...). I think a large number of people would prefer to see the game scale more like difficulty levels in chess. When you play chess against the computer, you're ALWAYS both playing by the same rules. When you play at higher levels, the computer just gets better - considers more options and looks further into the future. I realize this is extremely difficult to code for a game as complex as Civ, and totally understand and agree with the need for the AI to have different rules at higher levels. But at the same time, I would consider any changes that make the AI smarter - things that allow it to be more competitive at higher difficulty levels while playing a game with rules more similar to those I'm playing by - an improvement.

My idea of an ideal type of solution to this would be along the lines of a patch message or new game version feature that says: All difficulty levels should provide roughly the same challenge that they did previously, but now the AI is playing smarter, and therefore no longer receives a production bonus.

 

[Padma]

Originally posted by Charis
Ok, am I the only one that things that a supposed 'public' beta 'patch' has *NO* official mention by Firaxis, Atari, CivFanatics, or Apolyton. We're supposed to go to the web site of some party I've never heard of, and run an executable out of a zip file that has NO shred of any Firaxis/Breakaway support from it, and with a unprofessional .bmp file the only thing I can see without launching the exe?

Even if you early adopters have good virus protection, that won't stop a trojan horse. Maybe someone who has installed it can post the offical Firaxis readme, but that list of "Beta 1.13 changes" that is floating around doesn't look anything like a Firaxis Release Notes textfile. And it's four days early??

Legit patch? Spyware/trojan/hoax? Dunno, but I'm not touching that file with a ten foot pole until Tavis or the Firaxis web site announces it.


Charis

I understand your caution Charis. But it is a patch from Firaxis. Tavis didn't post it here, because:

Originally posted by Tavis
Anyone can download it, I just didn't want to do a general post since its not an official patch (beta or otherwise).

(Emphasis mine. )

 

[#1 Person]

 

[Ozymandous]

Originally posted by Bamspeedy


How ironic that you of all people would say something like that, considering past statements you have said about the site you participate in vs. the CFC GOTM.

I already explained my statement (9 days ago), you didn't need to bring it back up again.

Past statements? Care to cite examples? If I said anything in the past it was that at RBCiv many of the exploits some use with regularity are frowned upon because of the issues they cause with game imbalance. Doesn't seem elitist to me, if indeed many such comments were made at all.

You may have explaiend 9 days ago, but I didn't see the thread at all until this past week, and I commented when I saw your post since the thread was 12 pages long at that point and I hadn't read until the end before commenting, as most do.

Regardless of your personal bias against RBCiv, your comment *did* sound elitist, so my original post stands.

 

[Ozymandous]

Originally posted by Yeti
In regards to folks that want a smarter / better / tougher AI:

I don't agree that "just go play Demigod, Deity, or Sid level" is a good response.

Those levels are artificially tougher because the AI has been given unfair advantages (extra units, production bonuses, bonuses during trade, etc...). I think a large number of people would prefer to see the game scale more like difficulty levels in chess. When you play chess against the computer, you're ALWAYS both playing by the same rules. When you play at higher levels, the computer just gets better - considers more options and looks further into the future. I realize this is extremely difficult to code for a game as complex as Civ, and totally understand and agree with the need for the AI to have different rules at higher levels. But at the same time, I would consider any changes that make the AI smarter - things that allow it to be more competitive at higher difficulty levels while playing a game with rules more similar to those I'm playing by - an improvement.

My idea of an ideal type of solution to this would be along the lines of a patch message or new game version feature that says: All difficulty levels should provide roughly the same challenge that they did previously, but now the AI is playing smarter, and therefore no longer receives a production bonus.

Everyone wants this, but until a company is willing to invest the large amounts of money needed to program such an AI, and until AI programming advances further this will not happen. Hence why adding MP support is a much cheaper way (generally) of beefing up the competition of any game since then you have actual people playing against other people, which makes the game much more complex.

 

[Bamspeedy]

Regardless of your personal bias against RBCiv

No bias on my part against the site itself, as I have no problem at all and highly respect the majority of it's members. I only have a problem with a few certain members who can't control themselves from bashing other sites (or stereotyping the members of those other sites) whenever they get a chance.

I won't go any further into details here, since this has all been way off-topic and pretty soon a mod will probably step in soon anyways if this discussion continued.

 

[Muchembled]

I have just released a new version of Civ3Ext that could help fixing the problems about corruption.
It will allow us to experiment new models of corruption, rather than debating vacuously.
My utility is designed so that C++ programmers can easily add new models, but it also provides Civ3/PTW and C3C 1.12 models.

There are currently three models :
- Civ3PTW and C3C_112 as a basis.
- Example 1 : This is an example of how to fix all bugs, ie rank sharing and separate rank lists.

Unfortunately, I don't own C3C, so Civ3Ext is only for Civ3 and PTW.

For programmers :
You only have to modify Corruption.cpp and rebuild the application. The 'CorCtxt' class provides all the information you need. You can add a new method there and reference it with a new entry in the 'models' structure.

edit : hmm I should add comments in the code... well I am ready to explain it and help everyone to build it before you run away.

Besides, when I see the code of the CorCtxt::Civ3PTW method, I wonder how I must read the license of Civ3/PTW and how I must write Civ3Ext's.

Moderator Action: Advertising your threads is not allowed, generally. However, given the nature of your utility, and the fact that it might provide more light on what is happening in regards to corruption, I will leave this post as a special case. --Padma

 

[bc1]

Why not keep it simple ? The basic rules of the games are supposed to be:

- the farthest a city is from the capital, the more corruption there is in that city
- the more cities there are above a certain number, the more corruption there is generally
- the Forbidden Palace acts like a second capital
- you can decrease corruption by building specific improvents or connecting the city to one of your capitals

Then why not use:

Corruption% = F1[ F2(distance_to_closest_capital) * F3(total_city_count) ] * bonuses

Where:

F1(x) keeps corruption within a certain range (such as 0-75%)
F2(x) is an increasing function of x, such as K2*x (with K2>0)
F3(x) = max[1 , K3 * (x - optimal_city_count)] (with K3>0)
bonuses are coefficients (such as 95%) for specific city improvements such as courthouse, whether the city is connected to a capital...

Chess shows simple rules can make a deep game.

 

[DaviddesJ]

Originally posted by bc1
Then why not use:

Corruption% = F1[ F2(distance_to_closest_capital) * F3(total_city_count) ] * bonuses

Most people don't like the result of this: if you attack a neighbor and conquer some of his cities, all of your core cities start getting less productive. Ultimately, the bigger you get, the less you will produce. If you think there's a lot of whining about corruption now (I do), imagine how much there would be in this case.

 

[bc1]

Originally posted by DaviddesJ Most people don't like the result of this: if you attack a neighbor and conquer some of his cities, all of your core cities start getting less productive. Ultimately, the bigger you get, the less you will produce. If you think there's a lot of whining about corruption now (I do), imagine how much there would be in this case.

But isn't it the way it's always been in CIV1 & CIV2 ? As your empire expands, corruption increases; should'nt that mean concern all cities ? It simply translates the loss of government focus as bureaucy increases to cope.

 

[razor1952]

I agree with bc1 corruption should be simple, as a newbie corruption in all its variations (c3,c3c etc a beta's ) it hopelessly complicated now.

corruption IMHO should be

1. related to distance from palace or fp which ever is closer and corruption % assigned to that ( depending on size of map % would increase from place and fp over distance)

2.Modifyers applied to corruption % reduced by simple % for each modifyer, eg. courthouse -15% .

thats it

players, even newbies like me could understand it, decide to place palaces and fp in the best postions and moove them accordingly.

 

[anarres]

razor1952, the problem with that design is that you get less corruption per city if you pack them in tight. This would lead to 2-city-spacing in places and very tight (close) builds as a minimum to compete in things like the GotM or agianst other humans.

Already tight builds are powerful, just basing corruption on distance would make it massively overpowered.

 

[Commander Bello]

As a lot of other members already have pointed out, one of the biggest concerns seems to be just not the *corruption*, but the *waste*, a point to which I wholeheartedly agree.
I agree to the idea that as soon as your empire grows, you as the leader will loose your tight grip to a certain extent. But the way it is calculated right now (even with the 1.15 patch) seems to be far too much for the waste component.
Additionally, I once again would like to state that the idea of "hidden corruption/waste" of xx-hundred per cent and then being capped, is unintuitive and frustrating, as it makes a lot of means to fight both components very complicated to be used correctly. This in no way means fun, but is just tedious, annoying and in no way convincing.
I really hope that this concept will be revised - at least for the setup of Civ4.

 

[HellboundHeathcliff]

As I see it, the basic problem is, and always has been, corruption based on distance to capital. It has caused numerous exploits and posts comparing it to real world examples.

Why not just do away with it? Make it a random event or base it on distance to *closest* city. The farther it is from any other city the less corruption it has (i know, not real world realistic). Or conversely, if it is *too* close to another city, both cities corruption increase.

The corruption model should encourage OCP not discourage it.

 

[Commander Bello]

Originally posted by watorrey
As I see it, the basic problem is, and always has been, corruption based on distance to capital. It has caused numerous exploits and posts comparing it to real world examples.

Why not just do away with it? Make it a random event or base it on distance to *closest* city. The farther it is from any other city the less corruption it has (i know, not real world realistic). Or conversely, if it is *too* close to another city, both cities corruption increase.

The corruption model should encourage OCP not discourage it.

About the distance factor:
I think that it is a proven fact from history that through most eras, the farther a city was from the residence of the central power, the more likely it was to submit less income, that is, taxes. So, for the corruption aspect, I think the approach of linking it to distance from the capital is valid.
In Civ, it has been overextended by far.

And about the OCP:
I think, there should be no other OCP strategy than to place cities in a way that the least tiles are unused and as less as possible tiles are shared. This approach is obvious, easy to understand and would give no chance for "exploits" (at least I can't see any)

 

[HellboundHeathcliff]

CB: Using any sort of distance factor from the capital is what causes all the problems. It may be 'real world true' but there is no way to include it that doesn't present the desire to build compactly around the capital. Increasing corruption for cities that have overlapping city radii *may* be the only cure. Which, i suppose, negates the 'no way to do it' comment

Basically, just count the tiles that are within a different city radius as well and increase corruption in the first city by a predetermined amount. Each city would have to be calculated individually.

(i think i'm on to something here... not bad for half asleep)

 

[Slax]

Why isn't corruption based on happiness? Happy people produce more.

 

[Commander Bello]

@Watorrey:

Although I agree with you that it is the distance calculation, that causes the most of the problems we currently have with the corruption/waste issue (at least, as far as I understand that issue) we shouldn't close our eyes that most gamers expect exactly this concept. And in terms of 'realism' or 'authenticity' of the game, they are not wrong. At least not for the most time of known history.
I fear, any attempt to change this would cause a major uproar.

But, as far as I understand it, most of these people are more feeling mad about the waste component, than about the corruption component. To be honest, I tend to that attitude as well.
It has been very likely through history for a distant city to contribute less to the wellfare of the state than others, being closer to the central authorities.

Where the current concept seems to be completely rubbish is the fact that you can get 'infinite' corruption/waste at a given distance. By that, most of the obvious means to fight this are just failing and the player - as long as not having read and learned all the articles about that topic - just has no clue, why it didn't work.
This is minimizing the fun.

So, once again, the solution I would prefer is to split both components. As long as nearly everything is based on shields production, the "waste component" should be significantly lower than the "corruption component. Furthermore, there should be an absolute limit to both, may it be 80/60 in per cent (corruption/waste) or any other reasonable combination. If, at a given distance D, those 60% are reached then there should be no further increase, even at a distance of D+5 or D+10. So, any means you apply to corruption/waste would show an immediate effect, what is absolutely what a player would expect.
I think it is this attitude of "I'll get you from behind" which is incorporated in some if not many micro-concepts in the game (if not incorpored intentionally, then at least it seems to be perceived that way), which makes players feel unhappy.
This may very well be the result of some "easy" principles being clumped together. But, as we learn from the game, not always the combination of 6 or 7 "easy" concepts is easy to handle later on. Neither for the player, nor for the game engine. The result of that is just confusion on both sides, as we had to learn.
So, a more elaborated algorithm to calculate the effect of increasing waste/corruption is very likely to be easier to handle at the end. Again, for both sides.
Let's hope, that our concerns about that topic are read and will influence at least the planning for Civ4, in case that there shouldn't be any hope to find a better, more understandable and acceptable concept in a later patch. There will be still two years for Civ4 to be released, if at all.

 

[HellboundHeathcliff]

Originally posted by Slax
Why isn't corruption based on happiness? Happy people produce more.

There are enough luxury wars in the game now

 

[HellboundHeathcliff]

@CB

Yes, i agree that the waste is what most players dislike the most. I suspect it seems or is worse because there is also another way to reduce it separate from corruption.

I also have no illusions that there will be any major changes to the corruption model for C3C. It is, IMO already too lenient.

Hopefully they will soon start developing a new game engine for civ4 and seriously rethink the whole corruption model. Many good ideas have been presented in various threads. If they get rid of distance as a factor for civ4, i doubt there will be any more of an outcry over that than other changes they will surely make. Gameplay should take precedence over reality simulation when needed. After all, this is *not* a simulation game, it is a TBS based on history.