A Rank Corruption Discovery and Exploit to negate rank corruption

[Mazarin]

great discovery Qitai! You definetly do not need to feel guilty for posting that: Firaxis will only fix this bug if they notice that the community is aware of it.
This discovery definetly explains some of my observations in recent games where moving the palace to a remote area with very few towns of my own extremely boosted my income...and building up a second core was almost counter-productive.

 

[JJansen]

Well, Mr. Qitai, or, should I say "Bobby Oppenheimer"?

This is probably the biggest, juiciest, most delicious bug ever found in Civ III. Now that I understand it (Alex's help, his thread updated to reflect this thread's discovery, and I quote: "Qitai discovered that there is a bug in the rank calculations for cities closer to the Forbidden Palace than the Capital. In short, for such a city, the rank is given by the number of cities that are closer to the Capital than that city is to the Forbidden Palace."), maybe I can start playing the game like an experienced gamer.

Okay, I lied. I still don't get it.

But still, this is a huge thing, and congratulations on discovering this. To make an analogy, this discovery is to Civilization III what the formulation of the theory of relativity was to physics. As far as my opinion is concerned, only Alex's murderous dissection of the corruption patterns and David's development of the RCP ranks as high.

(this sort of reminds me of the tech tree- Alex's corruption begets David's RCP, which begets Alex's palace jumping exploit. The three together are more akin to Chaos Theory than anything else, I think…)

Someone please tell me, because I'm dialup and it's a damn imposition to even be surfing these boards, let alone downloading GOTMs, if I got this right:

I build rings around my Palace, then, on an iceberg deep in enemy territory near the further of the two poles, I found a city and build my palace there.

After that, I build a forbidden palace where my regular palace was, and poof- no corruption.

I know that's not what's been discussed here, but I'm asking if that would work.

Anyway, as a newb here, I must say that I'm impressed with the sheer number of graduate-level students occupying this board and sharing conjecture and hypothesis on a simple, profane video game of all things.

If this sort of dedication was focused elsewhere… well, maybe it's for the best that all you guys aren't primarily concerned with things like politics; we'd probably end up with a technocracy in a decade.

 

[Qitai]

JJansen> The usual order would be build a FP first before moving the palace. Also, if you use the extreme exploit of having no cities near the palace, then you do not need any ring concept. All cities will be rank 1 if the palace is far enough from any other of your own cities.

The above do not reduce corruption completely. It only almost completely remove one part of it, which in my opinion is the bigger contributor of corruption. This is the same part which RCP tries to reduce.

The corruption has two parts - Rank and distance. In despotism, both are equally evil. But when you move to other government forms and build improvements, distance corruption reduces dramatically. But not rank. This exploit only almost negates rank corruption. Distance corruption will still have to be fought using advance government and improvement.

If this is still not clear, I would advise that you try to understand Alex's formula first. One way to do it is to download his corruption calculator and see how corruption changes with different distance and rank, as well as the effect of various governments, improvement and things that affect corruption.

 

[JJansen]

Q-

(JJ will suffice. Minimum number of characters for a SN here is 3)

Well, I was broadly generalizing there. I have the feeling that I'm going to have to abandon that trait of mine, here at least…

I think I got it now:

So long as every city is closer to the FP than they are to the capital palace, corruption via distance is virtually eliminated.

So basically, after gaining magnetism and a GL, you could just move the GL, a Settler and oh, say, 10 Musketeers to Speck Island, Middle of Nowhere, Antarctic Circle, and put the palace there after planting the FP somewhere central in your territory, the result will be virtually no distance corruption for as long as you're away from your capital.

I hope I got it by now, because I already think I do.

Yeah, you're right, this is a dangerously easy bug to exploit. While I imagine that most players (ala www.gamefaqs.com, the 12-year-olds and the like) will never come across this idea, there still is a major, gaping hole in the structure of the game.

I was mulling over improvements to Civ III that I would make, had I any skill at Civedit.

One thing would be to make a new small wonder that would require, say, 20 Police Departments and the IA, call it the "Permanent Governmental Oversight Committee", and it'll combat corruption by halfing the overall total or something like that.

Of course, I also want the Telephone between Electricity and Electronics, right alongside Atomic Theory, (and the associated GW of 'Ma Bell'), but that won't happen, either.

Anyway, sorry for digressing; do I get it now?

 

[benstandby]

Will the multiplayer community look down upon this like they do ICS or ROP rape? Maybe we should post this in the Multiplayer thread and see if it's ethical.







...or maybe I should shut up and just use it until someone says something

 

[Qitai]

JJ, you sound like you got the idea. I did not realize you are 12 years old. So, all those mathematically detail are probably too tough for you. Hope in time to come, you can understand all these

benstandby, I think you should highlight it to them. This will probably help to eventually send the message to Altari that this should be fixed.

 

[lz14]

On second thought. Do you think it will boost your income if you have TWO cores ? Moving a palace far away will increase your home core, but what do you do with the *far away* core? Does one core have more income than two cores? Maybe OCP around palace and ICS(RCP) around FP will help, but does it help much more than RCP on both palace and FP ? I don't really think so, can anyone clear this up for me.

 

[Qitai]

Did a simple test with my completed GOTM22. Had normal 2 core without extensive RCP.

Income before = 384, Income after is 472.

That game is still in Middle Age with only one courthouse and aqueduct in the empire; and with the FP being more or less in a corner. Even with that, you can see a significant gain. I suspect a fully developed empire would have at least double the production with that exploit, if not more.

I am not sure how would that compare with an OCP at palace with ICS at FP. But here are some quick comparison and stats to show you how powerful this can be

On a standard map at deity level. Under democracy, a connected rank 1 city which is 50 distance away from the FP, with a courthouse and police station suffers only 20% corruption!!!!

With the same condition, if the city is rank 30 distance 50, the corruption goes to 125%.

With same condition, a rank 10 distance 10 city would suffer 30% corruption.

 

[Yndy]

It will help a lot because around the FP, all the cities will have the same low corruption due to the fact that there is no city close to the Palace. You can potentially build (say) four rings around the FP with the corruption of a first ring city. That replaces and exceeds the advantage of a ring or two around the palace.

Irrespective of how many rings you build around the FP, the corruption stays the same because the corruption counts the number of cities closer to the Palace.

 

[Dianthus]

Originally posted by Yndy
You can potentially build (say) four rings around the FP with the corruption of a first ring city.

With the advantage that the cities in those four rings don't have to be at the exact same distance. You don't have to RCP the cities around the FP anymore, you've got complete flexibility again!

 

[Yndy]

That's right. Read 'equivalent of four rings of cities'.

 

[ControlFreak]

Great Discovery Qitai!

So, without palace jumping, starting with a small first core (FP eligible) you're army sets off to attack...the fartherst civ from your capital. A great leader builds the FP in their capital. So far, normal play, with no exploit. But abandoning all your original core makes all territory acquisitions "rank=1". Exploit. Question is, how bad is rank=2 or 3 or 4? How many of the original core cities can you leave in tact without lowering the FP core significantly? And in leaving them, how long would this strategy remain undetected as an exploit? This truely is a competition game's nightmare.

 

[Qitai]

Originally posted by ControlFreak
Question is, how bad is rank=2 or 3 or 4? How many of the original core cities can you leave in tact without lowering the FP core significantly? And in leaving them, how long would this strategy remain undetected as an exploit? This truely is a competition game's nightmare.

That's a tough question. I can only use examples again. At democracy, standard, deity - without considering improvements - 3.72% per rank until rank 13. Thereafter 7.44% per rank.

And I would look at it this way. Each rank increases corruption in all cities in the FP region, so that is 3.72% times the number of cities you have per rank before rank 13. So, I guess you can figure out yourself how big that is. Note that this is just one example.

As for intentional non-building of cities. That is easily detected, since each city has the build year to it. So, you can't hide it. What is difficult to determine is a player using OCP at palace as a play style or intentional use to get lower corruption at the FP. I suspect everyone will now use OCP at the palace.

Personally, my cities after the palace jump tends to be more sparse (before having this knowledge) due to cities being captured from AI with little incentive to pop more cities, unless AI city placement are wasting some good tiles. But with this knowledge, I may not do that anymore.

 

[blobglob]

JJ, you sound like you got the idea. I did not realize you are 12 years old. So, all those mathematically detail are probably too tough for you. Hope in time to come, you can understand all these

I think you misinterpreted what he was saying. I very much doubt he's 12.

On a standard map at deity level. Under democracy, a connected rank 1 city which is 50 distance away from the FP, with a courthouse and police station suffers only 20% corruption!!!!

With the same condition, if the city is rank 30 distance 50, the corruption goes to 125%.

With same condition, a rank 10 distance 10 city would suffer 30% corruption.

Wow. Now that really puts it into perspective. Great discovery.

It's sad though how this makes such a big difference to the game by completely overhauing a lot of strategies (including a favourite of mine - RCP). It defies logic. This'll have to be fixed.

 

[Svar]

It sounds like players who build the FP next to their first city for an eventual palace jump have been benefiting from this bug all along as opposed to people like me who use the palace as the home core and build the FP as the second core.

It would give a huge jump in productivity when the palace jump is first made but would have less effect as more cities are added around the new palace. The overall productivity would probably still increase as more cities are added around the new palace but not as much as the productivity increase around a second core FP. Players using this strategy wouldn't know the reason for difference until now but intuitively they would use the strategy that yields the higest production especially in the MA when many of the best players are winning the game.

It will be real interesting to see how this affects game play. Do you take advantage of the bug or not? If not, what will you do if you already use the FP in the first core and do a palace jump to the second core? The concept of 2 core production centers has been the mainstay of the Gotm since I started (Gotm 20).

 

[Dominae]

if (Forbidden Palace is complete) 
   RelocatePalace();

function RelocatePalace() {
      t = locate(non-Mountain land tile on a crappy remote island)
      if t is not settled
         settle(t)
      c = the city at tile t
      g = wait(for a Great Leader to appear) // g is the Great Leader
      ship g off to c
      rush the Palace in c with g
      // the AI can use exploits too!
}

Dominae

 

[Txurce]

I don't like the idea of a palace jump - or an FP, for that matter - because neither exists as used in the path of history which Civ3 seeks to mimic. A weakened or non-existent FP would also break up the current strategic stranglehold of immediate two-core expansion at the expense of just about everything else, and perhaps take the game back to a more balanced approach to development.

In the meantime, though, I naturally build the FP in every game, and will eventually jump my palace in a GOTM. I also take RCP into ever-increasing consideration. All of these are arguably exploits, but fall within the rules and within the "gray area." Moving the palace to a far-off location is clearly an exploit that could be easily detected, and therefore banned. But I don't see why using ICS in the first core and building the FP there in anticipation of a palace jump, and then using OCP around the relocated capital, is any more exploitative than ICS, palace jumping, or RCP. It strikes me as similar to how the best players already play.

 

[ainwood]

If I read it correctly, patching it to make it work how we thought it worked would be the logical step: Ie calcualte rank based on whichever is closer: the Palace or the Forbidden Palace.

 

[alexman]

That might be the most logical step, but it doesn't mean that it's the easiest to implement. Patches tend to fix problems the easiest way, rather than the proper way. (See Firaxis' "fix" for the mobilization bug...)

 

[SirPleb]

Until there is a better solution, it might work well for competitive games to remove FP from the game (voluntarily or via the editor.) I previously said that would be drastic but on reflection, is it? It would level the playing field and remove the issue from our thinking.

I think I will play GOTM23 without an FP. I've already submitted 3 games for tournament 5 so I don't need GOTM23 for that. This might give us an interesting comparison on how much difference FP makes in final score. (Without considering the bug of course - it seems a safe bet that Cracker won't tolerate visible use of it ) And this way I won't have to think about the bug I will however use RCP and every other trick I can think of...