C3C Corruption: Whoa!!!

[vulture]

I'm not a big fan of this chosen 'fix' for RCP. RCP has to go, but this is the wrong way to do it, and the FP thing definitely looks like a bug. It sounds like the best you can do building the FP is to get a total civ productivity about the same as you had without the FP, and without good positioning you do worse.

I think it's fair to say that between this and the gpt bug, I'm not going to bother with conquests until a patch arrives for it.

[Harovan]

A RCP fix with tied cities solved by a foundation counter (not "time", as more than one city can be built in a given turn), seems the most logical fix. The counter should even be global (only one for all civs), since cities can be gifted, be taken or culture flip.

Don't forget with all the fix proposals the completely screwed FP.

[DrSpike]

There are several intuitive ways to implement the fix. Just break rank ties by order of building, or average the rank of the ring cities. With the latter you'd have to decide which rank to give an even number of cities, but you may as well round down.

The second leaves some RCP elements in place (albeit greatly reduced effects), but given the way it was implemented first I think that's probably a good thing. I mean, (as in method 1) should the rank of your new cities necessarily be increasing for EVERY city you build? I don't think so.

[Provost Harrison]

I think that idea is the best Sir Ralph, it would certainly make sense. To me, it would have made sense to use a purely distance-dependent system in the first place, of course taking into account distance from the palace and the FP and both being additive, dependent on map size, etc, etc...

[Stuie]

Originally posted by notyoueither

Originally posted by Dominae
Given all the stuff in C3C (the Conquests, for example!), I'm quite happy that this is the biggest patch-worthy problem encountered thus far.

I think the gpt bug is bigger. Unless you think that AI civs should have 20 or 30,000 gold in the bank.

Must agree with nye here. If I had to choose, I'd want the gpt bug fixed.

[Jon Miller]

I woudl prefer a pseudo random solution

like use the rank thing, and than add a +# where # depends on the number of cities at that rank

Jn Miller

[Doc Tsiolkovski]

Yes, RCP was "fixed" after the beta test ended

Not true. This bug was already in the up-to-now playable beta version, and submitted. It especially makes Conquests with no proper ways to fight corruption (Middle Ages) a major p.i.t.a....


But for me, this is good news, since now I know I can wait some more weeks for C3C without problems, and concentrate on the next DyP release.

[Slax]

Originally posted by bongo
The simplest way to fix the RCP-exploit is to base rank on foundation time.

Do cities placed in the editor have the same foundation time?

The 'best' fix would be to give best rank to the city that would benefit most from it. So if two cities are at the same distance, the one with the highest productivity gets the best rank.

What about ties in production?

[bongo]

Do cities placed in the editor have the same foundation time?

Never used the editor...Either add foundation time to the editor or use some randomness

What about ties in production?

With productivity, I was originally refering to the 'F11' kind of productivity but as it includes food surplus(which isn't affected by corruption, at least in PTW) a better solution would be to take the combined shields+gold+science output of a city. In case of a tie, flip a coin or use age, culture, whatever.

[Fosse]

I never used RCP anyway, favoring a "natural" city placement scheme. This means I never bothered counting tiles before, but when I get Conquests I had sure as spit BETTER count tiles to make sure I don't have any "rings."

And the last thing I want are more corrupted AI civs.

I'm going to hold out for a patch that addresses this (and this after my PtW CD just shattered (in the drive!!)).

I favor the build order fix.. it's simple and intuitve, and the best human and AI cities will often tend to be the earlier ones that have had more time to grow and build infrastructure, so it nets the best advantage for both.

edited for typo

[Arnelos]

Concerning the RCP "fix"...

They need to have more statisticians and fewer programmers making these rules

The solution should be obvious.

Say you have cities at the following distances:

3, 3, 3, 4, 5, 6, 6

Under the old rules, it would have produced either:

1, 1, 1, 2, 3, 4, 4
(sum of ranks is 16, where it should be 28)

or (I'm not sure from what you guys have presented):

1, 1, 1, 4, 5, 6, 6
(sum of ranks is 24, where it should be 28)

The new system with release C3C is the following:

3, 3, 3, 4, 5, 7, 7
(sum of ranks is 32, where it should be 28)

What it SHOULD be is that if you have a tie, as with ranking with a data set with certain statistical methods, you just add together the ranks that those data points would have (if they were at different distances) and average them - then assign the average rank to each data point in the tie.

So...

(1+2+3)/3 = 2 and (6+7)/2 = 6.5

thus...

2, 2, 2, 4, 5, 6.5, 6.5 (assuming this is possible with the way the math is done)

1+2+3+4+5+6+7 = 28
2+2+2+4+5+6.5+6.5 = 28
(the sum of the ranks in both cases is 28)

Since you're averaging them out, it's no more or less advantageous to build one city each at ranges 3, 4, and 5 than it is to build three cities at range 4. Thus you would be neither rewarded nor penalized for RCP, as it should be.

[Dominae]

Just watch Firaxis hire a statistician in the upcoming weeks!

[Fried-Psitalon]

Arenlos -

I certainly respect the logic of your position, but remember that we don't know how exactly they have such things coded, as outsiders looking in. Your "simple solution" may be as simple to them as, say, one of us being told the simple solution to our problem is to hang a cow by a piece of twine from a stoplight during the busy time for traffic.

[Arnelos]

Originally posted by Fried-Psitalon
Arenlos -

I certainly respect the logic of your position, but remember that we don't know how exactly they have such things coded, as outsiders looking in. Your "simple solution" may be as simple to them as, say, one of us being told the simple solution to our problem is to hang a cow by a piece of twine from a stoplight during the busy time for traffic.

Originally posted by Arnelos
2, 2, 2, 4, 5, 6.5, 6.5 (assuming this is possible with the way the math is done)

That was basically what I was trying to point out with that parenthetical note there. I recognize that using fractions may not be easy with the rank system, depending on how it is coded.

Sorry if I didn't make that more clear.

That said, if it's possible to code it this way, it would certainly work better.

[GhengisFarbâ„¢]

I have a off the wall question.

What happens if you mod the rules to allow more than one FP? Does it triple the rate at which the cites are corrupt, sorta average the productive capacity amongst the three palaces?