AU 505 The Power of Fascism - After Game Comments

[Golden Bear]

You know, it occurs to me that the AI favoring Nationalism while we are all beelining for Steam Power and TOE is an inordinate advantage for humans. It is pretty much a guaranteed success criteria for me even on emperor because I know that the AI will tie itself up researching Nationalism-Communism-Fascism while I am zooming through the required techs.

That would be a pretty straightforward way to make the game tougher on humans, if whatever flag leads the AI research were to be changed to follow the human research path.

[dmd175]

Originally posted by Golden Bear
You know, it occurs to me that the AI favoring Nationalism while we are all beelining for Steam Power and TOE is an inordinate advantage for humans. It is pretty much a guaranteed success criteria for me even on emperor because I know that the AI will tie itself up researching Nationalism-Communism-Fascism while I am zooming through the required techs.

That would be a pretty straightforward way to make the game tougher on humans, if whatever flag leads the AI research were to be changed to follow the human research path.

I completely agree. This AI behavior has guaranteed my getting ahead in what I view as key techs early in the IA. I've always taken much joy as I see AI after AI getting tied up with fascism and communism; thus fixing this behavior - pretty much an exploit - would certainly maintain the challenge to the human through the IA.

Furthermore, the AI places an inordinately high value on nationalism, and I rarely if ever have been able to trade for it with the AI - at least until it is beyond usefulness (ie RP researched).

[jshelr]

We made our landing in Germany and bumped around a little before declaring victory in this game and getting out

My main thought is that I like the game format:

1) No early wars initiated by the human player
2) No late wars initiated in democracy or republic

These two items combine to make the game interesting well into the late stages by keeping the human civ small.

I would support steps in the AU mod to enforce the necessity of using non-representative governments. These might include significantly increased unit support costs in the industrial and modern eras. Increased war weariness. Changing the hurry method in fascism to cash instead of pop. And whatever else people come up with.

Nice setup. Thanks a lot.

[JimiD]

Representative governments in similar games have been hamstrung by a senate or parliment (Civ 2), who would accept any peace treaty offered, and veto war declarations (ie cant chose to go to war, and cant pursue it successfully for very long), or by combat penalties (SMAC), either negative morale (ie vets are regs), and bonuses to 'war time' governments (vets are elites), so if not in a 'war' government then you will lose many more battles (vet vs vet becomes elite vs reg)

Civ3 endevours to acheive a balance using economy - commerce bonus in representative governments, offset by high support costs.

However the commerce potential of a representative government along with city improvements can offset the loss of free unit support, so it is all too easy (in my relatively limited experience) to run representative government and fight as needed.

WW does play a part, but a good economy and luxes (trade or conquest) can offset this, and most of my battles aim to be short and intense, so this does not play the role it might.

Has an analysis been done on the relative costs of support for a given city under different governments?

ie

A city (12) with improvements (library & marketplace) under demo, and a given corruption, might produce X commerce, and thus y gpt and z research.

Similar city under communism will produce less commerce, and gpt and research, but can support units for war.

(note that as the number of city improvements increases the commerce bonus is multiplied)

Does the demo city produce enough commerce to pay for its equivilent unit support in war, instead of the 'free support'?

I will have to investigate, but I would imagine that this has already been considered, and wont get onto a machine for a while.

Quick numbers at work:

Demo:

12 tiles worked plus city, assume each normally 2 commerce =26

loose 2 to corruption

12+6 = 18 gpt (bank) - 8 (4 units x 2gpt upkeep) = 10gpt
12+6 = 18 beakers (library)

Communism:

12 tiles worked plus city, assume each normally 1 commerce =13

loose 1 for corruption

6+3 = 9gpt (bank)
6+3 = 9 beakers (library)

4 units supported per city for free.


Is it really as simple as that or, more likely, am I missing something?

[Modo44]

Yes, it is as simple as that. Choose Republic, and never change to anything else, no matter if at war or not. This is true in most games, even with the AU Mod.

A Republic can usually handle 20 turns of war with ease. Heck, even a Democracy can, if the war is guaranteed to last only those 20 turns. If a player masters quick wars (BOOM -> we have 3 cities -> 4 turns later we have peace -> the AI pays tribute), and keeps the other AIs at peace, no War Weariness ever happens. The lux slider is much more powerful for those governments, offsetting the lost MP. And, the units not used for MP can go boot someone, or can be disbanded for a massive decrease in upkeep cost. Democracy is only dangerous against many opponents (and wars). Republic seems able to handle even that.

Yes, Fascism or Communism can be fun, but they are really useful only in very special conditions. Fascism - for a small core, preferably matros. Communism - for a huge empire. Both need to keep everyone at war, to stay alive in the tech race. Losing the cash (huge impact to research ability), and x turns to Anarchy is too big a hit in most cases. About the only thing those two governments are good at are extreme wars (many turns, many casualties, many cities changing hands), that would generate huge amounts of War Weariness.

Unfortunately, unit counts don't matter either. A player has a huge tactical advantage over the AI. This means, that he can often win with a fairly small force, that would be easy to sustain even with high upkeep costs. Those would only help against a massive build up of forces, which is often not needed, especially after Railroads.

[Risa]

I learned several points from this course.
1) In a game of no early conquest, gaining a city or settler from hut is very helpful, even if it's hopeless corrupted. I have played twice, both got New York by south of Rome. Helped by rice on grassland and forest-chopping, I claimed a fairly large area around - even larger than home part on the second try.

2) Longevity is very important for Fascism warfare. It helps the recovery from the population hit on government change, and more importantly, speeds up culture expansion of conquered land immensely.

3) Airfield is so cheap, to a point that there's almost no reason to build airports in cities.

And some observations:
1) Compared to Communism, Fascism is an ineffecient wartime government type. It's corruption model is inferior to Communism's. SPHQ is only useful to save a Ironwork city, cities around it are of course not total corrupted, but still too corrupted to be useful. And culture expansion in conquered land is too slow.

2) AI don't give a high enough priority on Computer (and to a less point, Synthetic Fiber). I enjoyed a quite long period of MA vs. infantry/TOW infantry warfare.

[Risa]

Originally posted by Dominae
Nationalism is not the prerequisite of anything all that useful, so you could just wait until the end of the Industrial era to research it. Not a huge change, if you ask me.

What if making it the prerequisite of Industialization?
It is a big change to alter tech tree, but is not the first time AU Mod do this.

[pvzh]

A more logical move would be to place all industrial governments after Espionage since it has industrialization as pre-request, so AI might go for something else after Nationalism.

[Dominae]

Originally posted by Risa

What if making it the prerequisite of Industialization?
It is a big change to alter tech tree, but is not the first time AU Mod do this.

Yes, a pretty drastic change (at least, compared to alexman's idea regarding Motorized Transportation). I like it, though.

[joncnunn]

Doubt it; if the AI is anywhere near my tech lead, I'd still be content researching something else and trading it to the first AI getting Nationalism.

With a big enough tech lead, sure I might make Nationalism my second tech (right after Steam Power.)

I've tried my all optional techs are now required mod. The affects on my research plans for Nationalism/Communism was competely unaffected. The most noticeable affect of this was a delay at the end of each era researching the bypassed techs.

Originally posted by Modo44
The point of Nationalism as a required tech is to make the human go after Communism/Fascism/Espionage some more. Those three techs often get completely neglected otherwise, especially because you need to research/trade for Nationalism to get them. You can often beeline for Rep Parts and completely forget that part of the tech tree - a part that just as often stalls the AI for a while. Making Nationalism required could break this human habit, at least to some extent.

The AI already is in love with Nationalism, and it does research Communism and/or Fascism very often anyway, so this change wouldn't hurt it at all (which is good)

[joncnunn]

I think what would help the AI the most would be removing the Communism & Fascism techs, moving Police Stations, and the goverments themselves into Nationalism. (Increase the cost of Nationalism)

Originally posted by Golden Bear
You know, it occurs to me that the AI favoring Nationalism while we are all beelining for Steam Power and TOE is an inordinate advantage for humans. It is pretty much a guaranteed success criteria for me even on emperor because I know that the AI will tie itself up researching Nationalism-Communism-Fascism while I am zooming through the required techs.

That would be a pretty straightforward way to make the game tougher on humans, if whatever flag leads the AI research were to be changed to follow the human research path.

[joncnunn]

Originally posted by The pirate
12 tiles worked plus city, assume each normally 2 commerce =26

loose 2 to corruption

12+6 = 18 gpt (bank) - 8 (4 units x 2gpt upkeep) = 10gpt
12+6 = 18 beakers (library)

Communism:

12 tiles worked plus city, assume each normally 1 commerce =13

loose 1 for corruption

6+3 = 9gpt (bank)
6+3 = 9 beakers (library)

4 units supported per city for free.

Is it really as simple as that or, more likely, am I missing something?

Yes, but a 24 out of 26 commerce city is extremely high; even with court houses. It's either right next door tio the palace without a CH (first 4), or within the first 8 cities with one. With Police Stations could be out to 12.

Offest by a non-FP / non-SPHQ Communist city getting 12/13 commerce would also be a really, really tiny communist country.

The one thing you did miss is that with a SPHQ, A communist empire would out shield produce a Democracy/Republic overall, but the manner in which it's done is more helpful for communists building a lot of units & cheap improvements, and counter-productive for wonder building outside P/FP/SPHQ.

Democracy would stilll be able to pay for the large army within reason with the commerce bonus at normal costs.
Republic does run into the support problem much quicker with the double cost over the limit.

[joncnunn]

My prediction:

All AIs would perform the following beeline:

1. Industrialism
2. Esponage
3. Communism.

Originally posted by pvzh
A more logical move would be to place all industrial governments after Espionage since it has industrialization as pre-request, so AI might go for something else after Nationalism.

[pvzh]

I do not think Espionage will be picked right away. At least now, AI does not pick it right away in AU mod, plus AI's will have Steam power (rails) and Industrialization (Factories/Plants) before they start to entrench themselves in government techs. Besides, after governments AI goes about replacable parts, maybe hiding goverments behind espionage will send some AI's for Replacable parts.
There are flavours in AU mod attached to techs already, so maybe some adjustment with tech costs like just increasing Espinage cost will discourage AI from researching it right away despite the fact that it allows government techs.
For example in Ancient Era, Religious AI (in AU mod) goes straight for Monarchy because all bottom branch is religious. Militaristic civs tends to go for Wheel, IW, HBR. Scientific AI if it has writing frequently picks Literature as next tech. So, it might work in Industrial Era, as long as goverment techs are not reachable right after Nationalism and enough diversity is present in terms of flavours for Industrial Era tech.

[Risa]

In my game AIs tech choice was pretty well already.
I was the first civ entering industrial era, followed by Germany, Netherland and Celts. My researching route was Steam Power, Industrialization and Eletricity, while Germany's free tech was Nationalism. Netherland and Celts's first tech was no suprise Nationalism, and Germany's was Communism. By that time war among Germany, Celts and Russia had already broken out, so their choice was understandable.
Later when I was about halfway to Industrialization, Russia was nearly finishing their medieval age too (judging by how they valued the last tech). Fearing their free tech would be Steam Power, I sold the tech to three indust-civs first, managed to buy Nationalism back, and not enough for Communism. It turned out Russia's free tech was also Nationalism.
By the time I had got Electricity, Germany researched Industrialization, Netherland got Medicine, Russia got Fascism. The latter two were branch lead.

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.