Apolyton University Mod: C3C version

[alexman]

Apolyton University is a school of strategy, where students sharpen their Civ3 skills and share their experiences in a series of thematic games. When playing an Apolyton University game, gaining and sharing knowledge is more important than getting a high score, a fast finish, or even winning the game. Participants are encouraged to share their strategy after the game, and even to try several attempts.

For single-player games we have developed this AU mod, whose main purpose is to challenge the player with a need for deeper strategy. To accomplish that goal, the mod tries to improve the AI and present the player with more strategic decisions, while changing as little as possible.

Philosophy

The main purpose of the AU mod is to challenge the player with a need for deeper strategy, while changing as little as possible. Deeper strategy is accomplished by presenting the player with more non-trivial decisions, and by improving the AI.

Presenting the player with more strategic decisions is otherwise known as ‘balancing’: strong elements of the game are made slightly weaker, and weak elements are made slightly stronger. The game is more challenging when the best approach is not always obvious, or at least when the best approach depends on the given situation.

Stronger opponents encourage the player to rely more on sound strategy, and less on one-dimensional approaches that work against weaker opponents. Therefore, improving the performance of the AI is one of the most important goals of this mod.

Game-play modifications that do not improve the AI nor balance the game are not part of the philosophy of this mod. However, cosmetic changes that add flavor but do not affect game-play might be considered.

Method

Most modifications incorporated by the AU mod are a compromise between a) improving the AI, b) presenting the player with more options, and c) changing as little as possible. The level of this compromise is a delicate and subjective decision, so each proposed change is discussed by the Apolyton University community before it is implemented.

A new idea for the mod, or an area where the stock version needs improvement should first be posted in this thread by any member of the community. A member of the AU mod panel (see next section) then starts a thread for discussion, where the community debates the merits of the new idea, the necessity for a change, its compliance with the philosophy of the mod, and its best final implementation. That thread is linked to this thread for easy future reference.

After listening to all sides of the debate, the AU mod panel then formulates a specific proposal, and the the issue is marked "under consideration" for a period of one week. During that period, the community is given the opportunity to present arguments for and against the proposed change.

After a week from the proposal being marked as "under consideration", the panel members get a time period (usually 24 hours) to cast their (Yes/No) votes. A minimum of 5 votes are needed within that time to make the decision official. If at the end of the time limit there is a tie, all remaining of the 7 panel members who did not vote, must then vote to resolve the tie.

The Panel

The AU mod is a community effort. However, a semi-official panel is set up to ensure this process runs smoothly.

The AU mod panel has these responsibilities:

1. Listen to input from the Apolyton University community in order to identify possible changes to the standard game that fit into the AU mod philosophy, or problems with modifications already present in the AU mod. This includes reading this thread, as well as AARs and DARs for AU courses, as many scenarios are used as tests for the AU mod.
2. Formulate official proposals for any change to the AU mod. In the case of new modifications, present a proposal to the community for discussion through a new thread, (which is to be linked from the mod’s description found in this thread). In the case of changes to the existing AU mod modifications, bump the relevant discussion thread by formulating the new proposal.
3. Vote on the resolution of each official proposal. Modifications may be incorporated on a trial basis, in case real-game testing is required to further evaluate the change. Panelists cannot abstain.
4. Implement any accepted proposal in the editor, create and upkeep the official documentation for the AU mod, and ensure that both the mod and the documentation are readily accessible the community.

The panel currently consists of these members (in alphabetical order):
alexman
lockstep
nbarclay
Nor Me
pvzh
Theseus
ZargonX

The Mod

The latest version of the Apolyton University Mod can be downloaded from the link below. You need to download the mod only if you want to start random epic games under the AU mod rules. If you are playing an AU game (such as AU 504), you do not need to download this zip file unless you want the civilopedia to reflect all the AU mod changes.

AU Mod 1.07: The latest changes voted in for the AU505 game. Contains the biq, civilopedia, and readme files. To use the mod, unzip the files into your Scenarios folder under Conquests. Then from Civ3 go into Civ-Content and choose the AU Mod from the list of scenarios and mods.

[alexman]

Description of all changes

UNITS

• AI Naval exploration: The AI doesn't fully understand the power of naval exploration, and doesn't build Curraghs until it has reached an advanced point in its military and territorial expansion. Most of the time, this point occurs well after Curraghs have become obsolete. Another problem is that the AI does not perform any suicide exploration missions, not even seafaring AI civilizations that get a reduced sinking probability. The risk of a few sunken galleys is a small price to pay for the potential reward of a trading monopoly with an entire continent.
  1. Curraghs get the 'unload' flag and are given the AI Naval Transport flag instead of the AI Naval Power flag.
  2. Curraghs have the wheeled property, and Ocean is impassable to Wheeled units.
  3. Sea and Ocean tiles cost 2 MP to enter for Curraghs.
  4. Ocean tiles cost 2 MP to enter for Galleys.
• Defensive units with attack capabilities: Units such as Infantry, Paratroopers, and Riflemen are marked for AI offense, which encourages the AI to build and use them specifically for offense, even though their attack factors are weak for their cost.
  1. Increased Infantry attack to 8.
  2. Increased Paratrooper attack to 6.
  3. Increased Modern Paratrooper attack to 8.
  4. Removed AI offense flag from Rifleman
  5. Removed AI offense flag from Mechanized Infantry
  6. Removed AI offense flag from Musketeer
  7. Removed AI offense flag from Impi
• Cavalry: A large proportion of Civ3 games effectively end when the human player gets Military Tradition. Such a powerful unit makes it hard for the human to resist the temptation to beeline for Military Tradition and roll over his neighbors for an easy domination victory over AI, which often has been researching the upper branch of the tech tree and does not even have Gunpowder.
  1. Reduced attack strength of Cavalry by 1 and renamed to Light Cavalry.
  2. Added a new unit, Heavy Cavalry (6-3-3), which is identical to Light Cavalry but with greater attack strength, and requires Nationalism. Heavy Cavalry is not available to Russians or Ottomans.
  3. Reduced attack factor of Sipahi by 1.
• The Conquistador: A UU is supposed to provide an advantage to its civilization, but the occasions where a Conquistador is worth building are very rare.
  1. Reduced shield cost to 60.
  2. Increased defense strength to 3.
  3. Added ability for Spain to build Explorers when horses are not available
• Balancing Airpower: C3C has introduced lethal bombardment to air units. While this change has added a new dimension to military strategy between humans, the AI does not cope well. The AI does not build enough air units and builds almost no AA units, so it's easy to use a fleet of bombers and a minimum ground force to defeat larger and technologically superior AI foes, taking almost no damage in the process.
  1. Removed lethal land bombard from Bombers.
  2. Gave Helicopters bombard strength 6, ROF 3.
  3. Reduced Helicopter defense to 1.
  4. Gave Helicopters lethal land bombard.
• The Keshik: The Mongol UU is a Knight replacement with a reduced defense and cost. That means that Keshiks can't attack as effectively without defensive support, so they have to advance more slowly. Not only is this unfair to the Mongols, it's also historically inaccurate.
  1. Added zero-range bombard strength of 2
• Modern Armor: Modern Armor is the best ground unit in the game. Its high attack and defense encourage players to follow the one-dimensional strategy of building exclusively Modern Armor as their ground units. You get almost the same defensive value as Mechanized Infantry, but with awesome attack capabilities.
  1. Reduced defense to 14
• Wheeled units and Impassable terrain: Wheeled mechanized units and artillery units add some depth to military tactics. Special forces become the strongest units in mountains and jungles, which encourages players to research their optional tech requirements more often.
  1. Add wheeled property to Artillery, Radar Artillery, Tank, Mechanized Infantry, Modern Armor, Panzer.
• The Privateer: This unit's sole purpose is to attack undefended transport ships, costs as much as a Galleon, but has less than a 50-50 chance of victory against either a Caravel or a Galleon. During war time, the Frigate is a superior ship because of its bombardment ability. During peace, the shields lost by unsuccessful Privateer attacks make it better to invest in infrastructure than in Privateers.
  1. Privateer does not require support.
  2. Privateer, Frigate, Man-O-War unbuildable when Destroyers are available.
• Modern Paratroopers: This unit is weak for its position in the tech tree and its cost.
  1. Marines upgrade to Modern Paratroopers.
  2. Increased attack of Modern Paratroopers to 12.
  3. Added stealth attack (vesrus all units except leaders, air, and naval) to Modern Paratroopers.
  4. Added amphibious ability to Modern Paratroopers.
  5. Renamed Modern Paratroopers to Special Forces.
• The Guerilla: The AI builds Guerillas, which cost the same as Infantry, even when it has access to rubber.
  1. Increased attack to 8.
  2. Added ZOC.
  3. Added ZOC to TOW Infantry.
• The Chasqui Scout: This unit is worse than a Chariot for combat because of its upgrade path, and worse than a Scout for exploring because of its double cost.
  1. Reduced cost to 15 shields.
• Balancing Ground Unit Bombardment: Ground unit bombardment is more powerful in C3C, yet the AI still doesn't make use of it on offense. This is most evident in the age of Artillery, but even Catapults and Trebuchets are more cost-effective than Cannons.
  1. Increased cost of Catapult to 30.
  2. Increased cost of Trebuchet to 35.
  3. Reduced bombard strength of Artillery to 10
  4. Added build-never flag to artillery for all AI civs.
• The Enkidu Warrior: This unit has only the defense strategy marked for the AI. That means that the AI does not have an offensive unit available at the beginning of the game, so it doesn’t get any bonus offensive units at higher difficulty levels.
  1. Added AI offense strategy.
• The Army: The AI does not know how to build, populate, or defend against Armies in C3C. By contrast, humans rely on armies more than ever before, to the extent that using an MGL to make an army has become a no-brainer, and waging war with Armies against the AI has become shockingly easy.
  1. Army: Added 4 bonus HP
  2. Army: Reduced transport capacity to 1
  3. Pentagon: Replaced 'Build Larger Armies' flag by 'Increased Army Value' flag
  4. Pentagon: reduced cost to 300 shields
  5. Reduced cities needed to support an Army to 1

IMPROVEMENTS AND WONDERS

• The Colosseum: Colosseums cost as much as two full-price Temples to build and maintain, but produce half as much culture and don't allow Cathedrals like Temples do. Their cost is prohibitive for an ancient city build. Even with the luxury scarcity in C3C, Colosseums are one of the least built city improvements.
  1. Reduced maintenance to 1 gold per turn.
  2. Reduced cost to 110 shields
• Balancing Naval Bombardment: Coastal Fortresses do not serve their purpose of protecting a city from naval bombardment, especially since it’s so easy to avoid the shots from their ZOC. They are the least often built city improvement in Civ3.
  1. Increased Bombardment defense of Coastal Fortress to 32.
  2. Increased bombardment strength of Battleship to 12
• Longevity: This Wonder does not provide the builder with a noticeable advantage because it comes at a time when most cities have already reached their maximum size, so they do not benefit from double population growth.
  1. Changed required technology to Sanitation.
  2. Reduced cost to 800 shields
• Cure for Cancer: This Wonder does not provide the builder with a noticeable advantage because with it comes at a time when most cities have marketplaces and access to multiple luxuries, so a single happy citizen rarely makes a difference.
  1. Increased happy faces to 3
• The Knights Templar: This Wonder is considered weak compared to other Wonders of its period. Also, the AI doesn't understand that it needs to build a barracks in the Knights Templar city.
  1. Added 'veteran ground units' flag.
  2. Crusader upgrades to Guerilla.
• The Statue of Zeus: Due to the clustered distribution of luxury resources, the player with ivory has an enormous advantage. The AI doesn't understand the value of ivory, and when it has ivory, it doesn't understand the value of Mathematics. The AI also doesn't understand that it needs to build a barracks in the SoZ city.
  1. Increased cost to 300.
  2. Added 'veteran ground units' flag.
  3. Removed bonus HP from Ancient Cavalry.
• The Oracle: This Wonder is one of the weakest in the game, yet it is usually one of the AI’s top priorities.
  1. Reduced cost to 200 shields.
• The Military Academy: Armies are extremely powerful in C3C. Their power is magnified against the AI because the AI refuses to attack Armies in the field, it does not use leaders to build Armies (so it cannot build the Military Academy), and does not build armies even when given the Military Academy.
  1. Removed victorious Army requirement.
  2. Removed 'increased army value' flag.
  3. Removed the 'build Armies without leader' flag.
  4. Added ability to spawn an Army every 25 turns.
  5. Reduced shield cost of Army to 1.

GOVERNMENTS

• Republic: The Republic has always been the most flexible government, and now that free unit support has been added in C3C, it's even better than before.
  1. Reduced the free unit support to 0/1/1 per town/city/metro
  2. Added 18 flat free unit support.
  3. All AI civilizations favor this government.
• Feudalism: There is little point to Feudalism, as it's worse than Monarchy in almost all situations.
  1. Reduced corruption to minimal.
• Democracy: Democracy does not provide a significant enough advantage over Republic to justify a non-Religious switch.
  1. Increased the free unit support to 0/1/1 per town/city/metro.
  2. Added 18 flat free unit support.
• Communism: Communism is too powerful with the new corruption model and the addition of the SPHQ.
  1. Removed Secret Police Headquarters Small Wonder.
• Fascism: Fascism is almost never better than Communism as a war time government, yet the AI consistently choose it during war.
  1. Increased Fascism corruption level to problematic.
  2. Added Secret Police Headquarters Small Wonder.
  3. All AI civilizations shun this government.

TECHNOLOGIES

• Philosophy: The free technology granted by Philosophy in C3C has added a 'no-brainer' decision to technology research: If you have the opportunity to learn Philosophy before all other civilizations, you should always try to do so.
  1. Not required for era advancement
  2. Removed half the cost of the Republic, and added it to Philosophy.

AI

• How to help the AI with happiness: The AI doesn't use the luxury slider. As a result, the AI assigns an inordinate number of entertainers, which cripple its growth, economy, and production. The fact that the AI pop-rushes and drafts more often than most humans, makes the problem worse. Also, the AI's combat tactics often result in war weariness, which forces a switch to inefficient wartime governments and a serious research setback late in the game.
  1. Doubled the value of entertainers.
• How to help the AI with research choices: The AI’s research priorities are predictable. Human players with experience can research technologies left by the AI until late (e.g. Polytheism), in order to get maximum trade value from their research. The AI also does not take into account its traits when selecting the next technology to research.
  1. Added flavors, additional technology properties, and dummy resources, as described in the related thread.
• AI build priorities: The AI build priorities don't always match the human build priorities. This is expected, as it is often necessary to compensate for the AI's poor military tactics. However, even taking into account the differences between AI and human optimal strategies, there are some areas where the AI build priorities could be improved.
  1. Added Offensive units to all militaristic civs and to Iroquois.
  2. Removed Defensive units, Growth, and Explore from all civs.
  3. Added Workers, Air Units, and Trade to all civs.
  4. Added Happiness to all religious or non-scientific civs that didn't have culture flagged.
  5. Added Culture to all scientific civs that didn't already have both Science and Happiness flagged.
  6. Added Production to all civs with fewer than 6 build-often items.
  7. Removed items so that all civs have a maximum of 6 items (Science from China).
• AI aggression levels: Adjusted AI aggression levels to be more in line with each civilization's traits, and less based on history.
  1. Greece from 3 to 2
  2. America from 3 to 2
  3. Babylon from 3 to 2
  4. China from 2 to 3
  5. Russia from 4 to 3
  6. Iroquois from 2 to 4

RESOURCES

• Resource Scarcity: The goal is to have resources which are scarce enough to provide a challenge, while they are plentiful enough so that they don't force players to fight a certain war or lose the game. So we want to increase strategic options while keeping the challenge offered by scarce resources.
  1. Increased the frequency of coal from 120 to 160.

MINOR CHANGES

• Removed 'Explore' ability from Settler, Worker, all ground bombardment units, Tactical Nuke, Leader, Army, and King units.
  These units were not meant to explore the map, in fact, their zero defensive strength makes it dangerous for them to do so. Accidentally sending these units exploring in harm's way is quite common, especially since the 'E' key is right next to the 'W' key.
• Added airlift flag to all ground bombardment units, scout, Explorer.
  If airlifting Modern Armor is allowed, why not Artillery? It's a pain to have to use transports only for these units, when there is no clear reason to do so in terms of game play.
• Added foot unit flag to scout and Explorer,
  so they can be transported by Helicopters, just like Infantry. This restriction rarely comes into play, but there is no clear reason why it should be there.

[alexman]

Changes in 1.07 (included above)

Cavalry:

• Renamed Cavalry to Light Cavalry.
• Restored Cossack attack strength to 6.
• Added a new unit, Heavy Cavalry (6-3-3), at the end of the Cavalry upgrade chain, and available with Nationalism. This unit, which is identical to Light Cavalry but with a higher attack strength, is not available to Ottomans or Russians.

The Statue of Zeus and Knights Templar:

• Added the 'veteran ground units' flag.

Governments:

• Republic: Increased flat free unit support to 18 units.
• Democracy: Increased flat free unit support to 18 units.

[alexman]

Changes under consideration

The Colosseum:

• Restore cost to 120 shields and flag as militaristic (voted in for 1.08)

City Names

• Extended the list of city names for each civilization (voted in for 1.08)

Discussion that has not (yet) resulted in changes

Amphibious units
The Explorer
Post-sail ship movement
The Javelin Thrower
The Jaguar Warrior
The Agricultural trait
The Expansionist trait
The 3-Man Chariot
The Carrier

Discussion that has resulted in changes to stock C3C!

The ToE-Hoover beeline: Removed prerequisite of Electronics from Motorized Transportation

[ducki]

Reading the "Method" section, I hope I understood correctly that this is the place for this suggestion. If not, I'll repost elsewhere.


I'd like to propose, at the very least, a discussion on, somehow rebalancing Granaries. With the advent of Agricultural civs, the ultra-early game is even more important, and I think if we can somehow temper the advantage a human has with the ultra-early granary while boosting the AI's use of early granaries, the Agricultural trait might not seem so overpowered.

That's my theory anyway, though I'm not experienced enough with the editor to know if there's a way to reduce the granary without hurting the AI more than the player.

I doubt this one will be popular.

[Theseus]

alexman... what can I say...

[ZargonX]

Wow... thanks alexman! This will make keeping track of things much easier! :salute:

[Buckets]

Have you considered using Flavours to enhance the ai's behaviour? I've thought there are some techs that the ai overemphasises (looking at you, Nationalism) and things the ai neglects (factories, granaries). If i understand correctly, it should be possible to use flavours globally to encourage and discourage improvements, techs and unit usage. If there's any consensus on which direction the ai needs to be pushed in, adding Flavours should be a piece of cake. Thoughts on this yet?

[donZappo]

I thought that the flavors discussion would fall under "How to help the AI with research choices." That's one of the threads that I'm eagerly awaiting creation. Let's get the ball moving panel members!

[sabrewolf]

yes , using flavours really does open a whole new perspective on enhancing the AI level.

they can be "motivated" to research their UU-tech and their trait-matching techs sooner. eg: medieval days: militaristic civs with cavalry-replacing-UUs prefer beelining for MT while commercial, scientific and religious take the upper half (rel: monotheism, democracy; com: banking, economics; sci: education) etc.

the great thing about these flavours are is, that you can really fine-tune some civs. the only disadvantage: only 7 flavours possible... but 8 traits... so it needs some intelligent combining which probably will take several cycles of balancing!

[nbarclay]

Originally posted by Buckets
Have you considered using Flavours to enhance the ai's behaviour? I've thought there are some techs that the ai overemphasises (looking at you, Nationalism)

Nationalism is a bit odd in that human players tend to care little about it but it's absolutely critical to AI survival. In fighting an offensive war, Nationalism matters little (unless perhaps a player wants to use rifleman/cannon tactics instead of either blitzing with cavalry or waiting for infantry/artillery). But for an AI threatened with a cavalry blitz from a heavily offensive-minded human player, the advantage of defending with riflemen instead of musketmen is enormous. So the AI habit of making Nationalism top priority in the industrial era is not something I view as needing fixed.

[alexman]

Just a reminder: this thread is not for debating proposals. Save your arguments for the appropriate discussion thread, when it opens.

[donZappo]

Alexman, are all of the panel members going to be monitoring every thread or, at the very least, one panel member making sure to check up on all of them? The AU modifications seem quite dependent upon each other, and it would be a shame to get into some pointless arguments in one thread due to decisions made in another. For example, if you make light and normal cavalry (the latter of which would be available under Nationalism) then that doesn't make Nationalism as valuable to the AI since it's not a priority unless the human also gets "normal" cavalry. This would change the nature of the AI research path thread because nationalisms benefits would have to be reconsidered. I just want to make sure that the word gets around when certain decisions are made so that other discussions can be more worthwhile.

[Dominae]

I'm glad and proud to have nominated alexman for the panel. Where would we be without ya!?




Dominae

[sabrewolf]

(bump)

can someone top this thread?