AU 505 The Power of Fascism - After Game Comments

[Risa]

As for unit support in Republic:
Since my land was divided into two parts, I needed to send enough forces to southern part to keep AI from attacking, as well as rushing as many as possible (ie. usually temple plus half courthouse). So my revolution delayed until entering medieval age. Even that, my sliders dropped to 4.4.2 for all those unit support and loss of military police. Later when marketplaces and cathedrals built in northern part and courthouses in southern part, the sliders gradually returned to normal 3.6.1 or 2.7.1.

All in all, the first impression is it's quite close to 1/2/2 model in late-ancient-early-medieval period in a glance. That is good, I guess.

[joncnunn]

In that case, humans research those techs not requiring Nationalism while waiting for AI to reach it if behind or slightly ahead.

If you really want to force the human to research Nationalism when slightly ahead without waiting for AI, every single tech would have to be placed such that it's a direct or indirect pre-req of Nationalism.

Originally posted by Risa

That is, after Nationalism, AIs tech choosing returns to normal. The only problem left is while AI are all busying Nationalism, player can go with main bulk without hindrance. Thus I suggest adding Nationalism to prerequisites of Industiallization, which is the first thing jumping into my mind.

[Dominae]

Originally posted by joncnunn
If you really want to force the human to research Nationalism when slightly ahead without waiting for AI, every single tech would have to be placed such that it's a direct or indirect pre-req of Nationalism.

...starting with Industrialization.

[joncnunn]

The Theory of Evolution - Hoover's beeline is outside Industrialization and so are Replaceable Parts & Sanitation

Still lots of techs for the human to reserach waiting for AI to get Nationalism.

[Dominae]

dp

[Dominae]

Originally posted by joncnunn
The Theory of Evolution - Hoover's beeline is outside Industrialization and so are Replaceable Parts & Sanitation

Still lots of techs for the human to reserach waiting for AI to get Nationalism.

We have to give the player some measure of choice; it's not very strategic to have to research Nationalism at the beginning of the Industrial every single game.

While I agree that the Theory of Evolution beeline is almost always the correct play, it's not quite as powerful if you have to backtrack through Nationalism to benefit from Hoover's. Replaceable Parts is a very good tech, but it's not entirely clear I would always research it before feeling the pull of Factories. I never research Sanitation.

[nbarclay]

After playing AU 505, I've about reached the conclusion that trying to make Fascism better is a lost cause. I see no reasonable way to make Fascism more useful to humans than Communism in any but the rarest of situations. Even if Fascism is better in the short run, it won't take much conquest to shift the balance of advantage in favor of Communism - and if you aren't planning conquest, why choose either Fascism or Communism? The xenophobic and forced relocation flags make Fascism worse than it would be otherwise, but even if we got rid of both of them - and hence deprived Fascism of most of its flavor - I can't see myself ever using Fascism in a normal game.

A long time ago, someone suggested merging the Communism and Fascism techs into a single Totalitarianism tech, and I think that would be a good idea. Fascism would still be there if players want to use it for some reason, but AIs would never waste time researching an extra tech for it and would never be stuck using Fascism instead of Communism just because they have the tech for Fascism but not for Communism.

[nbarclay]

If human players could trade Steam Power in an even swap for Nationalism under conditions where all else is equal, I might view the good of making Nationalism a prerequisite for Industrialization as outweighing the harm. Changing a tech from being optional to being a prerequisite for a second-tier tech is a huge change from stock, but the benefit of helping AIs trade for Steam Power more quickly might be worth it.

But right now, Nationalism and Steam Power cost the same, making a straight-up trade impossible when all else is equal. Further, I suspect that AIs place a value on Nationalism for trade purposes that is significantly higher than its research cost alone would warrant. (Can anyone confirm or refute that?) Under those conditions, I would strongly oppose such a change because it would change an optional tech not just to being mandatory, but to being mandatory under conditions that favor the AIs. Yes, that's good from a "helping the AIs" perspective, but I consider the impact on the feel of the game too big a departure from stock.

If we'd reduce the cost of Nationalism to a point where straight-up trades for Steam Power would be practical, the impact on the feel of the game would be significantly smaller. Players would have a strong incentive to trade for Nationalism sooner, but would not be stuck in a situation where they have to either research the tech themselves or trade on terms that place them at a significant disadvantage. That would make the change more worth considering in my view.

By the way, another pssibility would be to make Nationalism a prerequisite for Replaceable Parts. That would provide a lot more flexibility in research paths, and would eliminate the paradox of players' being able to skip straight from musketmen to infantry without ever having a time in between when the best defender available is riflemen. Unfortunately, the physical layout of the tech tree would make that a bit ugly.

[nbarclay]

Originally posted by Dominae
Like I said way back (before I left the panel), the solution to making the representative governments (and their trade bonus) more balanced is to make them pay through the roof for unit support costs.

C3C tried to do this with Republic with 2gpt support instead of 1, but this had little effect because of the boosted unit support values for cities and metros. The AU505 tweak of granting Rep. and Dem. +15 or so free unit support is definitely a step in the wrong direction.

The "builder" governments should have enough money to support a defensive standing army. Waging war while in Republic should be difficult. If ever diplomacy fails, the player will have to think long and hard about his or her government choice: do I risk getting overrun while staying in Republic, or do I defend myself properly with Monarchy/Communism, etc.?

It seems to me that the balance between builder and warmonger governments is deeply and fundamentally broken due to (1) the trade bonus's being so large compared with a civ's base income and (2) the time required to change governments being so long for nonreligious civs, especially on higher difficulty levels. Given those limitations, I see no way to get a really good balance between governments without altering the feel of the game more than I consider acceptable and creating some nasty effects for the game's enjoyability (at least from my perspective) in the process. The last thing I want is a situation where the only ways I can do serious fighting during even a portion of the medieval era involve either spending the entire medieval era in a war-time government or going through at least one extra period of seven or eight turns of anarchy. I don't view a better balance between Monarchy and Republic as being even halfway close to worth such a cost.

[Dominae]

Originally posted by nbarclay
The last thing I want is a situation where the only ways I can do serious fighting during even a portion of the medieval era involve either spending the entire medieval era in a war-time government or going through at least one extra period of seven or eight turns of anarchy.

This is because you are used to a Republic that offers far greater rewards than its "costs" are supposed to offset. You therefore consider it bad strategy to use any of the wartime governments almost at all, and who can blame you?

While I view this situation as a major flaw in the game (the choice of governments is not exactly strategic, is it?), I also appreciate that it is one of the defining characterstics of Civ3 as we know it. This is one of those cases where the rule of "staying close to the feel of stock" is actually a hindrance to the quality of this mod.

[Dominae]

I had my computer crap out on me a couple of days ago, which means I will not be able to play the last 15 turns of my game. Too bad: I had just switched to Fascism, meaning the fireworks were finally about to begin...

One comment I would like to throw out here concerns the AU mod's version of the Military Academy. I do not think the change (grants free Army every x turns) is doing what it is supposed to do.

In this game, I saw:

1. A 1x Medieval Infanty Army attack a Rifleman.
2. A 1x TOW Infantry Army attack a Mechanized Infantry.
3. A 2x Cavalry Army (wow!) attack an Infantry.
4. A 1x Tank Army attack an Infantry.

Of these, the first three Armies lost their combats (as expected), and the last won, only to be taken down in a counter-attack.

From this data, my observations are that: one, the AI does not load Armies sufficiently when it has them available; two, the AI wastes the special benefits Armies confer in the field.

Now, I believe the idea behind the Military Academy change was to promote the AI's use of Armies by putting many of them at its disposal. From my observations in this game I do not think that giving the AI more Armies to play with does all that much for it beyond being slightly cute. While I imagine that sometimes the AI makes a really awesome Army and parades it on the front lines, the fact of the matter is that it's going to fall almost as easily as any other combination of units.

So, having the Military Academy provide a free Army every x turns does not really help the AI. Does it help the human player?

For me, in this game, definitely yes. But I admittedly this was the best type of game for me to abuse the AU mod Military Academy: my self-imposed restriction to wait until the very end of the tech tree to attack meant that I would have many Armies at my disposal, at zero Shield cost (barring the cost of the Military Academy). Nonetheless I imagine that in more normal games, the AU mod version is not that far off from the stock version: all the human player needs to abuse the AI is an Army or two, and waiting for these (if necessary) is pretty painless.

The big difference between the stock and AU mod versions, of course, is that in the mod version the human player cannot stockpile Armies as quickly. But there is a simpler solution to avoid this: just make Armies cost more to build. By doubling or even tripling the cost, Armies would be a lot less cost-effective for what they do. I'm sure a happy ground can be found to avoid stockpiling but still make Armies a worthwhile thing to build if the game lasts well into the Industrial era.

The effect of this change on the AI (I imagine) will be that the AI will rarely build Armies. Given what the AI actually does with them, I do not really consider this a bad thing. We will see Armies when the AI gets them via MGLs, but Military Academy Armies will be a toy for the human player exclusively (albeit a great cost). While not perfect, I consider this a far better state of affairs than giving the AI tools it cannot use well, and giving the human player something so powerful at no Shield cost. Furthermore, by just making Armies cost more, the AU mod would be far closer to stock than where it is now with respect to the Military Academy.

In conclusion:

1. The AI is not more competitive with free Armies.
2. The human player use of the Military Academy can be hampered far more successfuly (and non-drastically) by simply making Armies cost more to build.

[nbarclay]

Originally posted by Dominae

While I view this situation as a major flaw in the game (the choice of governments is not exactly strategic, is it?), I also appreciate that it is one of the defining characterstics of Civ3 as we know it. This is one of those cases where the rule of "staying close to the feel of stock" is actually a hindrance to the quality of this mod.

That depends on how you define "quality." In my view, the whole design of the anarchy system in Civ 3 is seriously flawed, making changes in government for nonreligious civs a whole lot more painful than they ought to be. But because players can avoid that flaw by just changing governments once or twice in the entire game, the flaw doesn't undermine enjoyment of the game anywhere near as much as it would if players were pressured to change governments more often. The kind of change you want would make questions of what government to use when more interesting, but would do so at the expense of making the anarchy situation a much bigger pain. Whether that combination would constitute an improvement in "quality" is very much open to debate, depending on how one defines "quality."

[Dominae]

Originally posted by nbarclay
The kind of change you want would make questions of what government to use when more interesting, but would do so at the expense of making the anarchy situation a much bigger pain.

The only reason Anarchy is such a big pain is that there is one best government; the opportunity-cost (turns in Anarchy) to switch between governments is therefore rarely acceptable (put another way, if Democracy were leaps and bounds better than Republic, I'm sure you could get used to a bit of Anarchy).

You can solve the problem in two ways: one, make Anarchy last fewer turns, to encourage strategic switches between governemnts; two, make the governements relatively equal in power level, to encourage strategic choice between governements.

The AU mod can only employ one of these: AFAIK, Anarchy durations are hard-coded for the human player. So it's a question of changing the game drastically but for the better by tweaking governments, or leaving it as it is.

[nbarclay]

Originally posted by Dominae


The only reason Anarchy is such a big pain is that there is one best government; the opportunity-cost (turns in Anarchy) to switch between governments is therefore rarely acceptable (put another way, if Democracy were leaps and bounds better than Republic, I'm sure you could get used to a bit of Anarchy).

I view anarchy as a huge pain in absolute terms due to the hassle of micromanagement and related issues. Food shortages and happiness issues both tend to make periods of anarchy a lot of extra hassle, above and beyond the opportunity cost in lost productivity.

I actually do change to Communism fairly often these days because when I'm playing for domination, I like the feel of having my conquered territory be useful, and a large communist civ can research about as effectively as a republic once its outlying areas get developed sufficiently. But that just involves one extra change of government, not a need to keep switching back and forth.

You can solve the problem in two ways: one, make Anarchy last fewer turns, to encourage strategic switches between governemnts; two, make the governements relatively equal in power level, to encourage strategic choice between governements.

The AU mod can only employ one of these: AFAIK, Anarchy durations are hard-coded for the human player. So it's a question of changing the game drastically but for the better by tweaking governments, or leaving it as it is.

It's not that simple. Right now, Republic actually serves two roles: (1) it functions as the ultimate science-oriented government until Democracy and (2) it also serves the function of a government that can support a balanced playstyle that includes both research and occasional fighting. With the kind of arrangement you seem to want, we would be left without any government that supports a balanced style of play, at least until Communism (which can be a fairly potent builder government if a civ is sufficiently large). Republic and Democracy would be good for research but lousy for warfare, while Monarchy and Feudalism would be better for warfare but lousy for research. So while the choice of governments would become more interesting, a lot of strategic choices that require a balanced government in order to work correctly would disappear or be seriously undermined - except maybe in the case of Religious civs, which could more easily change back and forth.

[Risa]

Originally posted by Dominae
In this game, I saw:

1. A 1x Medieval Infanty Army attack a Rifleman.
2. A 1x TOW Infantry Army attack a Mechanized Infantry.
3. A 2x Cavalry Army (wow!) attack an Infantry.
4. A 1x Tank Army attack an Infantry.

I have even seen a warrior army in industrial era in my game!

In conclusion:

1. The AI is not more competitive with free Armies.

There's one thing to consider: AI totally refuses to attack a high defense army (probably player's) with normal unit, but it will attack it with an army.