View Full Version : Possible fix for overly large REF


HG_CassiusA
Sep 25, 2008, 12:39 PM
Try changing the values for


<Define>
<DefineName>REVOLUTION_EUROPE_UNIT_THRESHOLD_INCREASE</DefineName>
<iDefineIntVal>10</iDefineIntVal>
</Define>
<Define>
<DefineName>REVOLUTION_EUROPE_UNIT_THRESHOLD</DefineName>
<iDefineIntVal>75</iDefineIntVal>
</Define>
<Define>
<DefineName>REVOLUTION_EUROPE_UNIT_SHIP_MODIFIER</DefineName>
<iDefineIntVal>-50</iDefineIntVal>
</Define>

in GlobalDefines.xml

I'm not sure what reasonable values would be though, but I think using any smaller number would reduce the problem.

Wolf33
Sep 25, 2008, 01:13 PM
Increasing "iAIKingUnitThresholdPercent" in CIV4HandicapInfo.xml seems to work nicely as well.

dr_AllCOM3
Sep 25, 2008, 01:29 PM
seems like the REF is linked with:
- gamespeed modifier
- starting age modifier
- <iAITrainPercent> modifier (doesn't make sense)
- AI gets a small bonus (<iAIKingUnitThresholdPercent>)
- REVOLUTION_EUROPE_UNIT_THRESHOLD constant
- REVOLUTION_EUROPE_UNIT_THRESHOLD_INCREASE constant (= % of how more more bells i can produce compared to last time, before kings gets new unit. should scale with difficulty or?)

snoopy369
Sep 25, 2008, 01:38 PM
We need to look at the C++ specifically ... dr_A, the area you posted in the other thread, can you post the rest of that function, if it seems relevant?

TFVanguard
Sep 25, 2008, 01:39 PM
Can those values be moved to the difficulty setting file, or will that make things go boom?

Edit: Also, keep in mind that while <iAIKingUnitThresholdPercent> is an AI-specific function, the European powers are AI powers, and thus the REF would only apply to AI opponents. The question is how the rest of the function that was found within would look.

snoopy369
Sep 25, 2008, 01:42 PM
You can't specifically move things around like that, without modding it in the dll (in c++).

The section dr_A found was specifically for AI *players*, I think (without seeing the rest of the function I can't be sure of that, but it sure looked that way). I think this is probably very much more complicated than just changing one number, unfortunately.

dr_AllCOM3
Sep 25, 2008, 01:48 PM
<iAIKingUnitThresholdPercent> is for not human

CvPlayer::doBells()
int CvPlayer::revolutionEuropeUnitThreshold() const
{
int iThreshold = ((GC.getDefineINT("REVOLUTION_EUROPE_UNIT_THRESHOLD") * std::max(0, (getRevolutionEuropeUnitThresholdMultiplier()))) / 100);

iThreshold *= GC.getGameSpeedInfo(GC.getGameINLINE().getGameSpee dType()).getTrainPercent();
iThreshold /= 100;

iThreshold *= GC.getEraInfo(GC.getGameINLINE().getStartEra()).ge tTrainPercent();
iThreshold /= 100;

iThreshold *= GC.getHandicapInfo(getHandicapType()).getAITrainPe rcent();
iThreshold /= 100;

if (!isHuman())
{
iThreshold *= GC.getHandicapInfo(GC.getGameINLINE().getHandicapT ype()).getAIKingUnitThresholdPercent();
iThreshold /= 100;
}

return std::max(1, iThreshold);
}

CvPlayer::revolutionEuropeUnitThreshold()
void CvPlayer::doBells()
{
if (getParent() == NO_PLAYER)
{
return;
}

int iBellsRate = getYieldRate(YIELD_BELLS);
if (iBellsRate == 0)
{
return;
}
//add bells to political points
for (int i = 0; i < GC.getNumFatherPointInfos(); ++i)
{
FatherPointTypes ePointType = (FatherPointTypes) i;
changeFatherPoints(ePointType, iBellsRate * GC.getFatherPointInfo(ePointType).getYieldPoints(Y IELD_BELLS));
}

//update revolution unit bells
if (!isInRevolution())
{
changeBellsStored(iBellsRate);
if (getBellsStored() >= revolutionEuropeUnitThreshold() && iBellsRate > GC.getCivilizationInfo(getCivilizationType()).getF reeYields(YIELD_BELLS))
{
changeBellsStored(-revolutionEuropeUnitThreshold());
setRevolutionEuropeUnitThresholdMultiplier((getRev olutionEuropeUnitThresholdMultiplier() * (100 + GC.getDefineINT("REVOLUTION_EUROPE_UNIT_THRESHOLD_INCREASE"))) / 100);

if (NO_PLAYER != getParent())
{
CvPlayer& kParent = GET_PLAYER(getParent());
FAssert(kParent.isEurope());

CvCivilizationInfo& kCivilizationInfo = GC.getCivilizationInfo(kParent.getCivilizationType ());
int iNumFreeUnits = kCivilizationInfo.getNumCivilizationFreeUnits();
std::vector<int> aiUnitWeights(iNumFreeUnits, 100);
for (int i = 0; i < iNumFreeUnits; ++i)
{
int iUnitClass = kCivilizationInfo.getCivilizationFreeUnitsClass(i) ;
UnitTypes eUnit = (UnitTypes) kCivilizationInfo.getCivilizationUnits(iUnitClass) ;
if (eUnit == NO_UNIT)
{
aiUnitWeights[i] = 0;
}
else
{
if (GC.getUnitInfo(eUnit).getDomainType() == DOMAIN_SEA)
{
aiUnitWeights[i] += std::max(-100, GC.getDefineINT("REVOLUTION_EUROPE_UNIT_SHIP_MODIFIER"));
}
}
}

if (iNumFreeUnits > 0)
{
int iIndex = GC.getGameINLINE().getSorenRand().pickValue(aiUnit Weights, "Pick Expeditionary force unit");
int iUnitClass = kCivilizationInfo.getCivilizationFreeUnitsClass(iI ndex);
ProfessionTypes eUnitProfession = (ProfessionTypes) kCivilizationInfo.getCivilizationFreeUnitsProfessi on(iIndex);
UnitTypes eUnit = (UnitTypes)kCivilizationInfo.getCivilizationUnits( iUnitClass);
FAssert(eUnit != NO_UNIT);
int iNumUnits = std::max(1, getRevolutionEuropeUnitThresholdMultiplier() / 100);
for (int i = 0; i < iNumUnits; ++i)
{
addRevolutionEuropeUnit(eUnit, eUnitProfession);
}

const wchar* szUnitName;
if (eUnitProfession != NO_PROFESSION)
{
szUnitName = GC.getProfessionInfo(eUnitProfession).getTextKeyWi de();
}
else
{
szUnitName = GC.getUnitInfo(eUnit).getTextKeyWide();
}

CvWString szBuffer = gDLL->getText("TXT_KEY_NEW_EUROPE_ARMY", kParent.getCivilizationShortDescriptionKey(), getCivilizationShortDescriptionKey(), szUnitName, kParent.getCivilizationAdjectiveKey());
gDLL->getInterfaceIFace()->addMessage(getID(), true, GC.getEVENT_MESSAGE_TIME(), szBuffer, "AS2D_UNIT_GREATPEOPLE", MESSAGE_TYPE_INFO, GC.getUnitInfo(eUnit).getButton(), (ColorTypes)GC.getInfoTypeForString("COLOR_UNIT_TEXT"));
}
}
}
}
}

does anyone know how to make a debug msg appear onscreen in Col2? would help a lot^^

TFVanguard
Sep 25, 2008, 02:03 PM
Now I remember why I never got into modding... if I want to WORK, I'll WORK on my own stuff. :)

But, looks like it does come down to the "REVOLUTION_EUROPE_UNIT_THRESHOLD" as your initial bells, which is then multiplied by "REVOLUTION_EUROPE_UNIT_THRESHOLD_INCREASE" every time it trips. Sadly, this is a constant across ALL levels of difficulty. So, by default, it's 75 bells, then 84, then 92, then 101, etc...

So there's the problem. The multiplier is far, far too low, since your own liberty bell multipliers (from, say, the Printing Press) are likely to be over DOUBLE the threshold required (much less if you have Newspapers, etc). Very quickly, you can actually outpace the King's threshold.

The 75 to start seems fine, but the "REVOLUTION_EUROPE_UNIT_THRESHOLD_INCREASE" really needs to change (and also be dependant on difficulty level). It should probably be much higher, perhaps 50 or so, to start with.

Unfortunately, I'm still stealing minutes. Can anyone check if this value can be dropped from 'global' and moved to the difficulty XML? If it can, this would do wonders for the game all by itself.

Reveilled
Sep 25, 2008, 02:21 PM
TFVanguard, you said that when you tried changing iAIKingUnitThresholdPercent you saw a difference, right?

If that's an AI value, why would that be?

Is it possible there's a bug causing that variable to be applied to Players on top of their own modifiers, which could be responsible for the large REFs?

dr_AllCOM3
Sep 25, 2008, 02:21 PM
Unfortunately, I'm still stealing minutes. Can anyone check if this value can be dropped from 'global' and moved to the difficulty XML? If it can, this would do wonders for the game all by itself.

maybe i can rig <iAIKingUnitThresholdPercent> and use it as REVOLUTION_EUROPE_UNIT_THRESHOLD_INCREASE. has no purpose anyway.
i just getting the stuff for compiling a .dll.
you can think of fitting multipliers^^

TFVanguard
Sep 25, 2008, 02:27 PM
TFVanguard, you said that when you tried changing iAIKingUnitThresholdPercent you saw a difference, right?

If that's an AI value, why would that be?

Is it possible there's a bug causing that variable to be applied to Players on top of their own modifiers, which could be responsible for the large REFs?

Well, there's a possibility that it's running it all down TWICE, one for England itself, then boosting it again when hitting your colony. That is, it's winding up doubling over these routines because when running colonies, it checks the parent - then it actually DOES the parent (which is an AI player). So, this'll hit the routine once regardless of if it was supposed to.

dr_AllCOM3
Sep 25, 2008, 02:43 PM
something must be wrong. with REVOLUTION_EUROPE_UNIT_THRESHOLD_INCREASE=1000 the kings buys tons of cavallery or artillery, but like no soldiers.

snoopy369
Sep 25, 2008, 03:33 PM
I don't think using iAIKingUnitThreshold is a good idea - it has a purpose. If we're changing something in the DLL, just add that tag to the Handicap XML file REVOLUTION_EUROPE_UNIT_THRESHOLD_INCREASE and REVOLUTION_EUROPE_UNIT_THRESHOLD. It's cleaner and better to do it that way.

dr_AllCOM3, if you can figure out the best numbers for each I'll get it into the XML file in my patch-mod this week/weekend and we can go from there.

jimjet
Sep 25, 2008, 07:46 PM
Its the 'REVOLUTION_EUROPE_UNIT_THRESHOLD' value the one you have to change NOT the 'REVOLUTION_EUROPE_UNIT_THRESHOLD_INCREASE'.

For example if you set the value to 'REVOLUTION_EUROPE_UNIT_THRESHOLD' to 1000, you will see theres almost no REF grow. From what I understand the 'REVOLUTION_EUROPE_UNIT_THRESHOLD' variable works with REF something like this (taking 1000 for example and being simplistic):

Each turn you generate bells, each time you accumulate 1000 bells the king adds a unit to the REF. Of course there are some variables and percentages that are multiplied to that 1000, but from what I could see the base is that value.

That means that the more the value the less the REF grow.

I just finished a game on Pilgrim with the value set on 150 and won. The REF was pretty easy. Of course I didnt build any bells until turn 175 and rushed for 50%. But I think 150 its a good place to start.

dr_AllCOM3
Sep 25, 2008, 08:31 PM
REVOLUTION_EUROPE_UNIT_THRESHOLD_INCREASE also increases the amout of troops the king will get.
i modded the .dll, now i can see the numbers ingame.

Ellestar
Sep 26, 2008, 04:56 AM
REVOLUTION_EUROPE_UNIT_THRESHOLD_INCREASE also increases the amout of troops the king will get.
i modded the .dll, now i can see the numbers ingame.
Yes, REVOLUTION_EUROPE_UNIT_THRESHOLD_INCREASE only matters at the very start, until you'll get about 4*ln(2; REVOLUTION_EUROPE_UNIT_THRESHOLD_INCREASE) increases in a number of REF units. After that it doesn't really matter. Only REVOLUTION_EUROPE_UNIT_THRESHOLD does matter.

I think the bigger problem is that REF increases linearly with a number of produced bells. At the same time, each turn you lose 10% of your accumulated bells in each city. I think it's much better to count total accumulated bells towards REF (sum of CvCity::getRebelSentiment()) instead of total produced bells. As it is now, most founding fathers give you PENALTIES instead of bonuses as their benefits don't seem to outweight the increase in REF :crazyeye:

dr_AllCOM3
Sep 26, 2008, 05:24 AM
[...]I think it's much better to count total accumulated bells towards REF (sum of CvCity::getRebelSentiment()) instead of total produced bells.[...]
no, REF would stop then at some point and you could outbuild them.

Ellestar
Sep 26, 2008, 05:45 AM
no, REF would stop then at some point and you could outbuild them.
If you have a bigger total population then you need more rebel sentiment in cities so to declare a revolution. So, all your soldiers will also increase a REF army. If you use CvCity::getRebelSentiment() then REVOLUTION_EUROPE_UNIT_THRESHOLD defines the ratio between your total population and a number of units in a REF army. So, it's possible to balance it across different difficulty levels. As it is now, you can't balance REVOLUTION_EUROPE_UNIT_THRESHOLD for both higher and lower difficulties - playstyle affects REF significantly more than that variable, and players shouldn't be punished for their playstyle on lowest difficulties.

Dale
Sep 26, 2008, 06:00 AM
That just opens a MASSIVE exploit in that you keep a mass of units as colonists/pioneers/soldiers that are never counted in the REF calculation.

Once your colonies reach 100% sentiment the threshold will plateau. Thats the extreme the other way cuz the player can have a small colony population with colonies of 100% but massive outside colony population, whilst the King has a tiny REF in relation to the tiny population.

Unless I'm missing something I don't see that working.

Ellestar
Sep 26, 2008, 06:54 AM
That just opens a MASSIVE exploit in that you keep a mass of units as colonists/pioneers/soldiers that are never counted in the REF calculation.

Once your colonies reach 100% sentiment the threshold will plateau. Thats the extreme the other way cuz the player can have a small colony population with colonies of 100% but massive outside colony population, whilst the King has a tiny REF in relation to the tiny population.

Unless I'm missing something I don't see that working.
If it worked that way, then it should have been exploitable in the original Col2.

CvTeam::canDoRevolution(), CvTeam::getRebelPercent(), CvTeam::getTotalPopulation(), CvUnit::changeTotalPopulation(), CvUnit::updateOwnerCache()

Every unit that can found a city, including soldiers, counts in getTotalPopulation() and so in canDoRevolution() too. So, if you have 40 colonists/pioneers/soldiers and 10 citizens, you'll have 20% rebel sentiment at best. Cannons aren't counted, but that is true for an original Col2 too. Also, we may mod it to include cannons as well if nessesary.

TFVanguard
Sep 26, 2008, 08:37 AM
Seems to me that there are three things that need done.

First, up the 'threshold change' dramatically for lower levels. This may require a DLL chance since the variable is not set as a difficulty flag (though it should be, and this is an incredibly bad oversight).

Second, recode the entire 'add bells' routine so that it ONLY works if the player being checked is not a parent empire. There seems to be a double whammy in the code if I'm looking at it right.

Lastly, modify the 'threshold change' amount by the current mood of the kind. For 'friendly' and 'pleased', this should be faily high (maybe 1.5x), for 'cautious' it should be about 1.2, for 'angry' it should be 1.0, and 'furious' should be .8 or so.

snoopy369
Sep 26, 2008, 09:25 AM
TFV, Dale and I went over this at quite a length last night (mostly Dale, as I was fighting with C++ to compile). There are actually many things that affect the REF, and they are tied to difficulty level. If you look in the C++, quite a few things go into the Threshold variable, many of which are tied to handicap.

Dale did comment that the add bells routine does check for 'if you have a parent', though I didn't notice one way or the other. It may be later on in the code check, though, as he only found it later in the investigation.

Your third point is another interesting way to deal with this, though I suspect tying it to the gold-given is probably the most efficient way to accomplish this (and you do accomplish roughly the same thing, as that's the main way you piss off the king). It's easier to make changes when certain events occur, but certainly that's another thing to take into consideration.

Ellestar
Sep 26, 2008, 11:08 AM
By the way, what is a good size of REF compared to the total population for various difficulty levels? Say, piligrim - half of the total population, Explorer - equal to a total population, Revolutionary - double total population?

snoopy369
Sep 26, 2008, 11:13 AM
That's one of the difficult questions to answer, particularly given different strategies. There are ways to play the game currently that make REFs perfectly manageable (such as the wait-for-SoL strategy). Making the normal REF smaller will make those strategies even MORE powerful, and let you DoI earlier.

This is why this question isn't one with a trivial answer ;) I suspect we need weeks or even months of playing in order to get a reasonable test done.

Dale
Sep 26, 2008, 11:41 AM
I think you guys are aiming at too complex an answer. Complex answers cause complex balance issues. :confused:

Also I think you guys are forgetting that the positive of generating bells (prod bonus, FF's, culture) is still very powerful. From what I'm reading in this thread you are aiming at a large reduction with no offsets. It's almost like you can't see the forest for the tree in front of you, don't forget the rest of the game. :)

In my proposed solution I'm offering a positive to the player for doing a negative action (giving in to the King increases the amount of time between REF increases). So based on how much gold you end up giving to the King, you can earn more bells before the threshold kicks in. So if you piss off the King and generate a lot of bells, you will still see a big REF. But under those circumstances I see that as justified. You deserve a big army attacking you. That money you've saved can go towards buying your army. But keep the King happy and kissing his pinky, you effectively reduce the size of the REF.

Aside from the fact it also gives a reason to buy off the King. :)

TFVanguard
Sep 26, 2008, 12:03 PM
Dale, you're all about "The game balance should be exactly as it is now," when it's clear that the majority of the people on this very forum regret getting the game. The game balance, as it stands, is too punitive for the REF, period. It needs to be tweaked down, particularly on the easiest difficulty levels.

You're an 'expert' on this game, very very few other people are, and fewer still are even remotely interested in the meta-gaming solutions that you've put forward thus far. Given that, the REF amount is what needs tweaked. Period. End of story.

The most logical answers, to me, are to adjust the threshold increase rate given any particular difficulty level, and modify even that based on the king's feelings towards you (which currently does nothing).

Ellestar
Sep 26, 2008, 12:12 PM
I think you guys are aiming at too complex an answer. Complex answers cause complex balance issues. :confused:

Also I think you guys are forgetting that the positive of generating bells (prod bonus, FF's, culture) is still very powerful. From what I'm reading in this thread you are aiming at a large reduction with no offsets. It's almost like you can't see the forest for the tree in front of you, don't forget the rest of the game. :)

In my proposed solution I'm offering a positive to the player for doing a negative action (giving in to the King increases the amount of time between REF increases). So based on how much gold you end up giving to the King, you can earn more bells before the threshold kicks in. So if you piss off the King and generate a lot of bells, you will still see a big REF. But under those circumstances I see that as justified. You deserve a big army attacking you. That money you've saved can go towards buying your army. But keep the King happy and kissing his pinky, you effectively reduce the size of the REF.

Aside from the fact it also gives a reason to buy off the King. :)
I don't think so.

What's most important on the lower difficulties? It's most important to make the game winnable even if a player doesn't really know what is he doing, after all these difficulties are made for new players who're still learning the game. In your case, player may not pay the king that minor amount of gold and he'll not be able to win the game at all because of that sole reason, your mod doesn't change anything here.

On average/higher difficulties it's better to expand player's strategic choices and to make a game challenging. I bet on highest difficulties it's always better to start making bells only right before the revolution, your mod or not. So, there is no strategic choice, there is only one best option. Even if you decide to make bells earlier, the best choice is to buy off the King, and an advantage of it is unclear. In any case, you can't really use bells for a major Founding Fathers race - it isn't really efficient, so an important aspect of the game is neglected.

My way, on a lower difficulty player always faces a consistent but beatable challenge, which is a good thing. On a highest difficulties, you're not forced to use the same trick with zero bells before revolution all the time. You may decide for yourself when do you want to start producing bells for their advantages - they aren't cheap to produce, and you may make extra money or units instead of it so it's not as easy to make a decision as to do "never produce bells before revolution" or "always buy off the king if you make the bells".
At the same time, you face a consistently high challenge. It doesn't really matter if REF will be lower than some ridiculously high REF you'll get if you'll produce the bells too early - that is an unwinnable situation, so noone will fight it anyway. However, with the right number tweaking you'll face a REF that is big enough for a challenge. The point is to win before AI and before the time limit - that's the challenge. "Never produce bells before revolution" is not a challenge at all.

Dale, you're all about "The game balance should be exactly as it is now," when it's clear that the majority of the people on this very forum regret getting the game. The game balance, as it stands, is too punitive for the REF, period. It needs to be tweaked down, particularly on the easiest difficulty levels.

You're an 'expert' on this game, very very few other people are, and fewer still are even remotely interested in the meta-gaming solutions that you've put forward thus far. Given that, the REF amount is what needs tweaked. Period. End of story.

The most logical answers, to me, are to adjust the threshold increase rate given any particular difficulty level, and modify even that based on the king's feelings towards you (which currently does nothing).
Nah, just reducing the numbers doesn't really remove a meta-gaming solution (i don't like it as well). That makes REF manageable without a meta-gaming solution, but it's even better with a meta-gaming solution - REF will be a total joke! That fixes one thing and breaks another one. I think the better way is to remove a meta-gaming solution at all.

Dale
Sep 26, 2008, 12:23 PM
Dale, you're all about "The game balance should be exactly as it is now," when it's clear that the majority of the people on this very forum regret getting the game. The game balance, as it stands, is too punitive for the REF, period. It needs to be tweaked down, particularly on the easiest difficulty levels.

No I'm not all about that. I was just pointing out that I think you're looking for a solution which has the potential to unbalance other stuff.

TFVanguard
Sep 26, 2008, 12:25 PM
Nah, just reducing the numbers doesn't really remove a meta-gaming solution (i don't like it as well). That makes REF manageable without a meta-gaming solution, but it's even better with a meta-gaming solution - REF will be a total joke! That fixes one thing and breaks another one. I think the better way is to remove a meta-gaming solution at all.

I don't think you can, short of rewriting the game. The main problem right now is that the REF is wholly dependant not on the political situation of the new world, of the military threats of other empires, the size of your forces, or even the amount of 'rebel sentiment' in your colonies - but wholly and completely on liberty bell production.

The quick 'difficulty slider' fix, then, is to fix the REF count's threshold percentage to begin with. After that, start looking at the meta-game exploit and see if there's a thematic way to address that.

Reveilled
Sep 26, 2008, 12:28 PM
No I'm not all about that. I was just pointing out that I think you're looking for a solution which has the potential to unbalance other stuff.

But only in favour of the player, which is exactly what's needed at the moment regarding the difficulty at lower levels.

On pilgrim difficulty, there really should be next to no penalty for Liberty Bell production.

DrewBledsoe
Sep 26, 2008, 03:20 PM
--Never producing bells until late, by effectively "hiding" your soon to be bell producers, then suddenly having a mad rush-- may work, but its not only incredibly cheesy, its also completely ridiculous. Just stop and think about it for a second. Add that to the fact that the manual tells you to make bells asap without delay, add to that that ais seem to love bells and will get the asses handed to them without a doubt, and surely then Vanguard's point is valid.

When an absolutely basic game mechanic is broken, it isn't a point of thinking how to slightly tweak it, and how this may effect other game balances. That's a bit like having your arm hanging by a few sinews, and someone commenting "oh I think you're getting a bit of a rash on your arm". Liberty bells should be producable right from the very start at all levels, its an integral part of the game in so many ways. Then balance the rest of the game from there.

Probably 75+% of people who buy this game, will try at an easy level, get slaughtered, maybe try again a few times then give up in frustration. They will tell their friends "it's impossible, don't bother getting it" or something similar. Hardly good for marketing and sales. Of the other 25%, some of these may visit these forums. An even smaller percentage, will attempt to play at the higher game levels....and an even smaller number (we are getting quite weeny in the number category now) will use any and every exploit left in the game. And your telling me, that those are the guys we have to make sure its hard enough for?

Hmmmm.....


N.B. Dale you're doing a sterling job as always, just hope this one isn't too broken to fix...

Franks
Sep 26, 2008, 03:21 PM
Completed my second game. First one New France got independence before me. Second one, New Spain declared independence but got whiped by their King, then I declare independence, I have about 10 dragoons, 12 soldiers, 4 cannons, 2 galleons (wow!) and 1 Ship of the Line (didn't have time to make more, I was too nervous at Holland getting independence before me). The king had something like 80 soldiers, 30 artillery, 40 dragoons and 12 warships. Result: Way too easy. However, I lost the game, because the King was taking too long to bring his troops in the blood bath ... I was 4 turns away from Europe and had two small colonies on a small island that I used at first for furs but then I abandonned them, the King took them of course, but I coudn't care less. The rest of my colony consisted of 4 cities. When I lost the King had 10 soldiers left, 2 artilley, 4 dragoons and 12 warships. After my loss I opened world builder and saw the King landed 4 troops near a native settlement (???). Apart from that it was ok, way too easy, but not enough time to enjoy building the colony. I think 50-75 more turns would be perfect.

Ellestar
Sep 26, 2008, 03:52 PM
I made a first version of my mod http://rapidshare.de/files/40558186/Ellestar_REF.rar.html

I'm still unsure about a good REF size compared to a total population.

snoopy369
Sep 26, 2008, 04:06 PM
Linking REF to population is a very bad idea, because it's like linking AI army size in Civ4 to your military count - it too strongly encourages a particular playstyle (low population), no differently from how it currently encourages low SoL. I'd keep it the way it is now, just alter the ReF threshold based on difficulty level.

smoke9999
Sep 26, 2008, 04:27 PM
Also I think you guys are forgetting that the positive of generating bells (prod bonus, FF's, culture) is still very powerful. From what I'm reading in this thread you are aiming at a large reduction with no offsets. It's almost like you can't see the forest for the tree in front of you, don't forget the rest of the game.

ok, the production of liberty bells is "§$"§$ essential, there are multiple things bound to it.

YOU HAVE TO PRODUCE LIBERTY BELLS FROM START.

Founding fathers, borders, production bonuses all bound to bells. It is a bug if u cant build a decent amount from start cause if your borders dont grow and u got some other europe near u ur §$"§$"§$ed. And no founding fathers for u if u dont generate some bells. This founding father system is big crap also cause u can actually rush them, look at col1 how its way better :(

why make things much worse in a "remake"

If producing liberty bells from start makes the game impossible to win, it is a bug! look at col1. Sorry dale but ur still defending this horrible bug.

Its not a feature. its bad balance and a bad system made by the programmers.

Ellestar
Sep 26, 2008, 05:00 PM
Linking REF to population is a very bad idea, because it's like linking AI army size in Civ4 to your military count - it too strongly encourages a particular playstyle (low population), no differently from how it currently encourages low SoL. I'd keep it the way it is now, just alter the ReF threshold based on difficulty level.
No, REF has a base value of about ~20, so it's advantageous to have more population. Also, original Col2 mechanics works exactly the same way - you need a certain number of bells so to get 50% rebel sentiment, and bells create more REF units. So, double population = about double REF army increase. You can't just reduce REF threshold, see post #27 for details.


Its not a feature. its bad balance and a bad system made by the programmers.
For the sake of programmers, i say it's a bad system made by game designers. Game designers design such things. Programmers just make it happen so they're not responsible for a bad gameplay.

snoopy369
Sep 26, 2008, 05:17 PM
REF should not be tied solely to population, though. That means SoL percent is irrelevant - since you HAVE to get 50% SoL before independence, yes, you know the MINIMUM, but now there's no disincentive to getting a HUGE SoL level (as many bells as possible). If that were the ReF determining factor, I'd say go for 100% SoL at all times.

REF needs to be tied to bells, in some fashion, in order to have some reason not to get them. I agree there needs to be a solution to the REF problem, but I think making it effectively static is not the solution.

Dale
Sep 26, 2008, 05:22 PM
ok, the production of liberty bells is "§$"§$ essential, there are multiple things bound to it.

YOU HAVE TO PRODUCE LIBERTY BELLS FROM START.

Founding fathers, borders, production bonuses all bound to bells. It is a bug if u cant build a decent amount from start cause if your borders dont grow and u got some other europe near u ur §$"§$"§$ed. And no founding fathers for u if u dont generate some bells. This founding father system is big crap also cause u can actually rush them, look at col1 how its way better :(

why make things much worse in a "remake"

If producing liberty bells from start makes the game impossible to win, it is a bug! look at col1. Sorry dale but ur still defending this horrible bug.

Its not a feature. its bad balance and a bad system made by the programmers.

I'm sorry, but I don't think you understand what I'm saying. You keep focusing on the bell production. You don't HAVE to make bells from the start. There's no set in stone rule saying you must. You're encouraged to by the massive advantages, countered by the disadvantage of more REF. What I see proposed in this thread is that if you make bells from the start, you get all the massive advantages, with a small disadvantage. This is what I'm trying to point out, don't forget the rest of the game when trying to fix this. You are going from one extreme to the other, from "to win hold off bells" to "to win rush bells from turn 1". That situation is no better than what we have now.

That is in NO WAY defending what you believe is a bug (technically it isn't a bug, as it works correctly, its a balance issue which aren't bugs).

BTW, you CAN get founding fathers with zero bell production (and it's actually more efficient as I pointed out in a different thread). Build politics points in a colony. You gain politics points (3 per hammer) without REF increases.

Lord Shadow
Sep 26, 2008, 05:44 PM
Seems like everyone's ignored Franks' post. :rolleyes:

Apparently he totally owned the REF despite being outnumbered 1 to 6. Sounds like Pilgrim, and besides the King's men came in waves, which makes them easier to defeat them.

Again, I must ask, how many of you have actually fought a really big REF? The big number is intimidating, so I suspect some quit in despair without even trying to face the King's regulars. How many of these complaints are backed by solid experience instead of speculation? It doesn't matter if they're 200, 500 or 1000 if they come in waves AND drop like flies (someone mentioned it's easy to heal inbetween waves, especially with units with Medic promotions).

Let me clarify that I'm speculating as well, taking the experiences of others into account, as I've been pretty busy to complete even a single game.

smoke9999
Sep 26, 2008, 06:11 PM
ok then tell me plz what i can do if some bell producing french settles next to me an constantly increases his territory over culture (e.g. bells).

i ultimately have to fight him or a huge REF. awesome choices there.

i just believe they messed this game up, this is one part of it.

Southern Hunter
Sep 26, 2008, 09:33 PM
I appreciate Dale's comments and willingness to engage in this discussion, but I agree with the critics that the mechanism seems busted (bad game design).

There SHOULD be advantages to generating rebel sentiment, and there are.

There are also COSTS, it takes time, food, money to get Statesmen, etc, etc. This can all be spent on other things like producing goods, units, etc.

The huge negative incentive of the REF is totally out of whack, unexplained in the manual, doesn't scale for easy difficulty levels and not necessary.

If it was totally missing, there is still a dynamic between building Bells and building other things, also useful.

The fact that forests and carpenters can produce many more political points than statesman is just a bizarre exploit / bug and should probably be removed.

I appreciate those working towards a solution. That the game designers and beta testers did not work this mechanic out is disappointing, but will also hopefully be addressed by them in a quick patch.

Lord Shadow
Sep 27, 2008, 12:56 AM
Political points aren't rebel sentiment. Statesmen create political points as a by-product of the liberty bell generation, which is the one that produces rebel sentiment.