How I won on revolutionary..

[Gliese 581]

.. or a field study in exploit pwnage of the King of Spain.

After discussing some of the broken features with people in this forum and pondering how to exploit it I decided I might as well put it to the ultimate test and start a revolutionary game.

The settings were me as Simon Bolivar of Spain on a new world map standard speed revolutionary difficulty.

I started out by establishing a colony on Tupi/Arawak lands with my soldier and quickly found a second spot on non-indian land for a second colony with the pioneer. Both cities were set to generate maximum trade income.
First city had 2 food resources and became a food farm city and second was unimpressive in most regards but had a mountain tile for a little early silvermining.

It turned out George Washington was a close neighbour and he settled just south of my position. I quickly attacked him with my veteran soldier (that spain start with) and took Jamestown and Plymouth (which was razed by default at size 1). GW proceeded to found Roanoke 1 tile away from a scout of mine and I captured and kept that city as well for an early 4th.



After this I founded my 5th and last city in a silver/fur spot.

Strategy was to play nice with indians and train some farmers etc which I did. The indians seems to get pissed off at you easilly at revolutionary so I quickly established missions in all villages with the colonists born in my food city.
I didn't bother with much infrastructure, just got up docks warehouses a couple of wagons then set all but 2 cities on building political points which was enough to quickly secure Peter Minuit. The 2 cities where I did get some infrastructure got all carpenters and lumberjacks and tools and ultimately cranked out some cannons but most cannons were bought in europe from trade income. After I had gotten the farmers and such I wanted from the indians I killed off any nearby villages to avoid surprises later and for a little extra cash two ggs and some experience.

After I decided I had enough cannons (22) I decided to push for independence. I had gotten 2 veteran soldiers from the dock queue so I decided to buy one extra to make it 4 in total with the one I started with. I also bought an elder statesman and had a schoolhouse in my food farm teach a second.

Almost ready to push for independence:



I started the push by deleting every citizen but my 4 veteran soldiers and 2 elder statesmen who were split in my 2 infrastructure cities that had each built a printing press and newspaper in anticipation of this, putting me down to pop 6 from about 50.



I had made sure they had some spare food as well as extra wagons with food for after independence so they could starve for a while generating bells.



After deleting all unwanted citizens of New Spain I gifted my other 3 settlements to George Washington (fair is fair). You might notice that I didn't bother to build any defenses around the cities (stockades etc). There's no point at all since the king can just bombard your defenses down to nothing so either you defend with no fortification bonuses or you attack him first, luckily I was able to do the latter.
Up to this point I hadn't generated a single bell in any of my colonies and only bothered to get Peter Minuit (though my political point building had accumulated another 1000 pps or so, not that I ever got any use out of it).
The REF was thus unchanged since the start of the game the turn before I got my first bells.



It took me 6 turns to get both cities to 100% rebel sentiment and the king added 1 unit each turn.



I was a little bit concerned still because I read here and in the civilopedia that the rebel sentiment bonus only worked on defense. This however turned out to be incorrect.
As the king landed his troops I had a minimum of 75% odds for attacking his troops from inside my settlement (or just outside it). Even for unpromoted cannons (str 3) vs promoted artilley (str 4). I didn't even get attacked by the whole fleet at once, first he sent 3 ships then the last.

But funniest of all is that before I got to kill all of his land units I sunk his entire fleet of 4 man-o-wars with a combat4 frigate that had been protecting my waters from privateers all game long. The first two it took down one by one (healing in between) at 75% odds.
Then I baited the last 2 with my galleon and merchantman which they killed but wounding them in the process allowing me to attack and kill the third.
The last ship tried to escape to europe but I hunted down and killed him on the subsequent turn ending the WOI in victory!



Now I might have won a revolutionary victory with 2 puny cities in 1675 but I didn't get a very high score.



So what to say? Didn't the beta uncover these exploits? This clearly shows how fundamentally flawed two portions of the game currently are. I didn't even micromanage very well in this game and could probably have ended the game much earlier if I had focused from the start on what to do but I made up particulars of the strategy as I went along and thought about it. This is the first time I finished the game so obviously this could be developed even further..
The two main concerns:

1) The ridiculously bad defense of AI colonies. I might only have temporarilly stopped GW in my game but he was totally crippled after this and behind me in score all the way right up until I deleted all those citizens. Meanwhilst Peter Stuyvesant was accumulating FFs like no tomorrow which ultimately didn't do squat for him.

2) The REF can not be tied to bell generation like it currently is. There needs to be a linear progression of unit buildup that is intense enough at revolutionary that you need to exceute some remarkable fortune and empire building skills in order to overtake the kings buildup progress, declare independence and fight of his army before the time runs out or you are defeated by him or an AI colony.
In contrast that should be very easy on pilgrim of course.

Edit: Added save, just attack the MOW to win.

[Viperace]

It took me 6 turns to get both cities to 100% rebel sentiment and the king added 1 unit each turn.

Great job in showcasing the REF flaw


Now we know how to take down the guys, lets spread the news ! Fellow rebellion!!

[Andvare]

You could super charge it, and stockpile weapons and horses in your smaller, size 1, cities. Then when revolution comes, take the 2 pop per city, and turn them into dragoons.
But it seems that the two-city gambit works well.

[GenocideBringer]

So your evidence that the game is flawed is that you can go through very elaborate means to exploit the REF system and get an incredibly tiny force to deal with?

Please. Exploits exist in all games, and simple fixes (not elaborate ones) are all that are needed to fix them.

[Gliese 581]

Originally Posted by Andvare

You could super charge it, and stockpile weapons and horses in your smaller, size 1, cities. Then when revolution comes, take the 2 pop per city, and turn them into dragoons.
But it seems that the two-city gambit works well.

Yeah I could actually have gone with just 1 city since the REF landed rather than attack amphibiously. I was a little concerned that the king might take 1 city by gathering all his forces in one place and attacking it from his MOWs.

I lost 1 cannon in the WOI.

Genocidebringer: I'm not so sure it's going to be a simple fix here. The REF system needs to be completely redone from the ground up and while it might be simple to put something like a fixed buildup in place, balancing it for the different difficulties is going to require alot of work. Essentially firaxis need to put the game through the whole testing period a second time if it's to be done right imo.
Most games are exploitable to some degree and in some of them, that's expected in order to win on the highest difficulty. But CIV is a good counterweight example here. Despite having owned it for years and now playing on immortal I've not been confident enough in my knowledge of the game to attempt a deity win, because unlike in col2 that is actually difficult even if you know about all the possible exploits (or advantages you have over the AI).

[Viperace]

Originally Posted by GenocideBringer

So your evidence that the game is flawed is that you can go through very elaborate means to exploit the REF system and get an incredibly tiny force to deal with?

Please. Exploits exist in all games, and simple fixes (not elaborate ones) are all that are needed to fix them.

This is clearly a glaring exploits. I am thankful that Gliese581 showed us how to make it works( just like I am thankful to those various high-difficulty guides in civ4).

By showing the exploits exist, he has taken the initiative in perfecting the game. Next is to figure out a better rule to fix the REF scaling.

[mzprox]

This is clearly an exploit, but: in most case when you know about an exploit you can easily not do it and enjoy the game. In this game however it is very hard to avoid it because you know the mechanism, you know:
-smaller cities are better- so you will make huge cities just to be sure it's not exploit?
-capturing the ai cities very easy- so you wont attack them even if you are a warmonger type?
-rebel sentiment should be raised as fast as possible- well, you set the challenge rating here not when you pick a difficulty in the beginning, but if you planning to win you you have to know the mechanism-you have to be fast..
-etc

[Ellestar]

Originally Posted by Gliese 581

Genocidebringer: I'm not so sure it's going to be a simple fix here. The REF system needs to be completely redone from the ground up and while it might be simple to put something like a fixed buildup in place, balancing it for the different difficulties is going to require alot of work. Essentially firaxis need to put the game through the whole testing period a second time if it's to be done right imo.

Well, with my mod you'll face 74-77 units instead of 26 http://forums.civfanatics.com/showthread.php?t=292708

[Andvare]

Originally Posted by Viperace

This is clearly a glaring exploits. I am thankful that Gliese581 showed us how to make it works( just like I am thankful to those various high-difficulty guides in civ4).

By showing the exploits exist, he has taken the initiative in perfecting the game. Next is to figure out a better rule to fix the REF scaling.

Hell, he didn't even exploit it to the max. He didn't stockpile food to instantly create a new army or create a ton of size 1 cities to get a horde of colonists for free (you can, if you have the weapons for it, get 3 soldiers for every city you have the turn after you declare independence, imagine this with a population of ten, spread out on ten cities, but then, you could just turn on the world builder instead, exploiting this game is too easy, and my will is weak ).

[obsolete]

Hmm, I had a feeling he was playing something like this, since Gliese seems to hardly get any Fathers. Anyhow, I myself have been keeping my empire small, to get around the broken Bell system.

I almost wonder if Firaxis made it screwed up on purpose, just to prevent the "land is power" kids from playing in a straightforward manner :P

[Andvare]

Hehehe, Futurehermit wont like this.

[misterslack]

100% Bolivar ownage. Awesome.

[Gliese 581]

Originally Posted by mzprox

This is clearly an exploit, but: in most case when you know about an exploit you can easily not do it and enjoy the game. In this game however it is very hard to avoid it because you know the mechanism, you know:
-smaller cities are better- so you will make huge cities just to be sure it's not exploit?
-capturing the ai cities very easy- so you wont attack them even if you are a warmonger type?
-rebel sentiment should be raised as fast as possible- well, you set the challenge rating here not when you pick a difficulty in the beginning, but if you planning to win you you have to know the mechanism-you have to be fast..
-etc

Exactly. This is very different from for example selling the AI in CIV a resource it already has (back when that was possible) or a glitch like the one demonstraded in the vampire masquerade speedrun where you can skip an hour or so (in speedrun time) of the game by entering an area that is supposed to be off limits until much later.
This has to do with some of the fundamental characteristics of the game. The way AI colonies react to warfare and the way the REF is determined. Now I'm sure that some people will have no problem avoiding these things nevertheless and conduct something akin to a roleplaying game in CIV but for me these kind of strategy games are only enjoyable if I'm doing my best to win, that's why I play them.
Right now this game is ruined in that way for me.

As Andvare says, this could be taken even further. I did stockpile small amounts of food 150-200 per colony but you could make that a goal from the start rather than an afterthought and have a 1000 food or more per city. I also only had 100 extra horses and guns per city instead of huge numbers. But Bolivar's trait is apparently so powerful that you could fight of a significantly larger force. I could probably have taken out the MOWs with frigates alone if I had 5-7 of them with an upgrade or two (easy since the AI sends mass numbers of privateers on this level without regard for your frigate patrols).
Short of testing out the Agg AI and OCC options with another leader I don't see how to increase the difficulty level any further and please observe that I've never been a colonization expert, I played the old one on a mid-level. Despite this, I'm confident that I could easily win with those extra challenges tossed in, using these exploits.

Ellestar: Interesting I'll have to give that a look.

[Delm]

Great job Gliese581, I hope there will be a patch out soon =/ or at least a decent mod

[Lord Shadow]

Hmm. I noticed something off about the combat bonus from rebel sentiment: the general sentiment was 88%, yet your units only gained 76% bonus. It's still considerable, but shouldn't it be 88%, just like the sentiment?

[GenocideBringer]

Ellestar: Interesting I'll have to give that a look.

That pretty much proves my point. A relatively minor mod can fix glaring exploits, it doesn't require an entire revision of the liberty bell system.

[Gliese 581]

Originally Posted by Lord Shadow

Hmm. I noticed something off about the combat bonus from rebel sentiment: the general sentiment was 88%, yet your units only gained 76% bonus. It's still considerable, but shouldn't it be 88%, just like the sentiment?

It should yes, I didn't check my general sentiment and growing extra citizens with food actually retarded my army by lowering the value I guess. We'll have to wait and see if Firaxis comments on rebel sentiment mechanics. From the civilopedia it's only supposed to work on defense which seems underpowered, otoh working on offense as well seems overpowered..

Genocidebringer: Perhaps but notice that he did completely change the REF generation system, removing bells from the equation. But from a quick look it's clear that this mod is just in the beginning stages and we'll need to test it out thoroughly to get good values for all difficulties and speeds, the mod creator acknowledges as much himself insofar as it not being yet compatible with all speeds and the specific numbers being largely untested.
I don't think this is something that the players should be burdened with, Firaxis should take the time and effort to correct the problems. There might be no choice for us but to trust in the community for this though, I don't know.

[Lord Shadow]

Hmm. The only thing I can think of is an artificial cap on the bonus, imposed by the Revolutionary difficulty level.

[DrewBledsoe]

Originally Posted by Gliese 581

It should yes, I didn't check my general sentiment and growing extra citizens with food actually retarded my army by lowering the value I guess. We'll have to wait and see if Firaxis comments on rebel sentiment mechanics. From the civilopedia it's only supposed to work on defense which seems underpowered, otoh working on offense as well seems overpowered..

Genocidebringer: Perhaps but notice that he did completely change the REF generation system, removing bells from the equation. But from a quick look it's clear that this mod is just in the beginning stages and we'll need to test it out thoroughly to get good values for all difficulties and speeds, the mod creator acknowledges as much himself insofar as it not being yet compatible with all speeds and the specific numbers being largely untested.
I don't think this is something that the players should be burdened with, Firaxis should take the time and effort to correct the problems. There might be no choice for us but to trust in the community for this though, I don't know.

Many thanks for this thread Gliese. I was still "on the fence" about whether to get Col II yet (still a holdout). Your thread is probably the final example I needed to keep waiting.

I've modded, tested and helped code various incarnations of Civ IV over the past few yrs, and while enjoying it in the main, its too time consuming a process. This time, I want a game that works properly from the day I buy it. Dale, Snoopy and Ellestar (to name a few) are each attempting various fixes. I suppose I'll have to look at it with the view, that the game is only really being beta tested now
. When its a finished product, I'll go buy it.

Thanks again.

[Blackmantle]

Originally Posted by GenocideBringer

That pretty much proves my point. A relatively minor mod can fix glaring exploits, it doesn't require an entire revision of the liberty bell system.

Even though i do agree with your base premise that it could be fixed to be fun! with a few rather simple fixes, i strongly agree with Gliese 581 that it should be rebuilt from the ground up. Because that whould make the game even more fun. And far easier to differentiate between difficulties. Which is badly needed.

But i have to actually disagree with Gliese 581 that it whould be soo a hard thing to do codewise. See Dale's modification idea (I proposed a changed base-system (including a treshold increase for bell-Ref after each ref-increase, and including maxpop, number of cannons, economy and bought guns + if doable max food around into the equation.) for the game minding all the major exploits sorts of. In just that thread. Shouldn't be all that hard to do i reckon...) and Ellestar's linked here.


The truly hard work imo rather lies with making the actual AI better (now thats somewhat hard allready for a game like civ4, but much harder for a game with so many variables and layers as is Civ4:Col.).
Now even in Col 1 the AI wasn't a real huge problem (by design, sorts of. The game is not all about wars, at least until the WoI. And imo shouldn't be.) as a dangerous competitor and not utterly necessary to mind very much even at viceroy if you really didn't want to and didn't get very unlucky.

But at least it wasn't overrun by a few dragoons later in the game at higher difficulties (or utterly securely by paying a few hundred gold to the indians to have those guys wipe them out for that matter. There the AI actually started to wipe the Natives rather early into the game with their troops. Not the other way around.).
Unless by chance at the very start (thats somewhat ok i think but not for all AI without any serious fuss.). That actually worked better in Col 1 were you got quite some colonists that way, because the AI was fond to return with more colonists 'for you'.

Now luckily for Firaxis / Take 2 i belive the REF-issue is the bigger problem in terms of overall fun in a game like Colonization. Even without highly competent AI the game can be very much fun due to a solid economic and native part. (And the most awful points about the AI should be easier to adress again then the finetuning.) Col1 proved that.


What i find truly glaring is that obviously the devs + testers have hugely overestimated the value of bells, sentiment and FFs (as seen in Dales comments as a tester in one of the more 'rough' threads around here) and harshly underestimated / neglected the power of the disband-button + cannons at the same time.
As a direct result of that, the AI seems quite fine at getting the liberty-bells + FF-aspect. And to disregard the rest more or less.



What the AI needs to grasp is:

a) military (defensively as well as at least some offense. Especially if the natives / their Ref comes knocking. Also they should be able to finish off a native tribe if they really aim to do so (especially spanish AI). Needn't be a few-turn affair. But no chanceless (colonial power) Bug vs. (native) Godzilla battles anymore. Please?) Thats by far the most important point for AI-improvements imo.

b) naval movement and combat + harassment (especially the part with hauling off treasures with Galeons... Should be doable and Galeons are not all so a bad investment. Also caravels shouldn't behave like slow bait if privateers are around. Run for the hills (settlements)!)

c) Economy (perhaps it does that allready. Doesn't much seem so though... Perhaps at govaneur + above its a tad bit different.)

d) focus less on bells and FFs because spreading borders angers natives which leads to dead AI. At least until point a) is fixed and the REF generation is done in a more organic way, which is less totally focused on Bells.

If you take above example im utterly sure it whould have perfectly worked even without Peter Minuit and massive ref bonus (if taken to another level by massing food and taking "all men are free") if some other "tricks" whould have been used.


I whould go as far as saying sentiment is downright unnessecary (for fighting the battles / Winning the WoI). As are FFs. You just need to throw 3 times the troops at him (works for the Ref in waves at the very least. Why shouldn't it work for you with a concentrated force, or harassment or a mix of the 2?).
And lots of cannons included in that troops of course. Its much easier to do than it sounds actually (i rather think 5:1 is doable. 10:1 if you count rabid armed + mounted natives into the whole thing and missions all over the place spawning converts all the time).

But the biggest feature-addition i could imagine diplomacywise (and i was sure whould be in. Thats perhaps the biggest disappointment for me) whould be to implement inter-native rivalry. Then you whould have real diplomatic beartraps and much backstabbing, fighting and things to mind.
Or am i wrong and the natives battle it out between themselves and draw you into their wars at higher difficulties?

Also in addition to delivering potentially very fun gameplay its very much accurate (Azetecs anyone?) so no big stretch here.
If that is seen as a problem for multiplayer balance (very valid argument for a game so obviously designed around multiplayer. Not sure if that was such a wise design-choice after all with the kind of marketing surounding CiV:4Col... but thats another matter) that should make for excellent mod-mod material. (And i reckon not one of the hardest changes to do... Just include other natives as possible targets of DoW for native tribes.)


@ Andvare to continue here on our previous discussion / case-study:
The settlements strewn all over the map whouldn't even need to be Pop 2 because if you pick: 'all men are free' you get 2 pop even in your 'sacrificial goats-settlements' of which you can settle 1 just when the ref comes near so they don't raze them. So even much less time needed to get this sentiments up to 50%, another point . Inland you just need the food as mentioned not even those 2 free guys each.


Add to that you get one bell for free each turn without Ref (measured for the whole colony not each settlement. Source: Dale) and you could even misuse a convert / servant / criminal to pump the sentiment up in those 1-pop settlements one by one with very few REF-interference.
(no Ref whatsoever from 1 criminal and 1 Ref for 75! turns (normal gamespeed i reckon) if the one unit belling is a convert / servant. 37 Turns if its a non-statesmen doing the politics. Sounds not all so horrible with Pop-1 cities.)