AU Mod: Re-balancing the power of Armies

[Dominae]

Well, you would have ~2 Armies already by the time you build the Military Academy...

[ducki]

Oh, duh, I forgot that turns get shorter(years-wise). Darn.
Could you expire the capitol and have that only affect the Army-generation?

Damn. Stupid years/turn thing. Arg.



I sent you a PM recently Dom

[Tall Stranger]

Originally posted by alexman
The fundamental benefit of warmongering versus building is that you beat up the AI while gaining territory. This doesn't change.

But this is all is done AT THE COST of building internal infrastructure, improving your economy, enhancing your research, etc. One of the benefits of warmongering is the higher probability of getting MGLs/ Armies. You're trying to take that away.

Armies as currently implemented in C3C are a human-only toy that just makes warmongering ridiculously easy, even against technologically and numerically superior AIs.

I swear this discussion has been held before, but here goes: Then the goal needs to be helping the AI build and use Armies, not denying the human the ability to do so.

The minute you start weakening one style of play (warmongering) relative to another (builder), you're undermining macro-strategic choice. How would a builder feel if I said: "Look, the reality is winning via SS is absurdly easy. Therefore, I think certain key city improvements (say libs/temples/markets/cathedrals) should require strategic resources." Or something along the lines of "Since humans can use pre-builds and the AI can't, we can use this to our advantage. Therefore, I think we should make it impossible for humans to build more than 1 great wonder ever."

In trying to balance AI versus human, you're going to destroy the balance between warmonger and builder.

[ducki]

Tall Stranger - Yes, we've had this builder-warmonger-armies-mil.acad. discussion multiple times. Pre-C3C the problem was that warmongering was a better builder strategy due to GLs, so while we may feel like we're going in circles, C3C rewrote the book on builder-warmonger balance.

On removing the ability of the human to build armies, I'm more opposed to it than you are - I want peacemongers to be able to build them too - but without some official statement from Firaxis on whether AI Armies will be addressed or not, we get to debate this one once more.

Exasperating? Sure. Tiresome? You bet, since opinions are mostly unchanged from PtW days.
But necessary. An AI that doesn't generate MGLs and create Armies with them is crippled against a player - builder, warmonger, warbuilder, OCCer, any player - with a mind to get an Army of his own and one is all it takes to ensure the player has a gamelong advantage.

Somehow, some way, we must convince the AI that using an MGL for an army is - in the majority of cases - preferable to other options, at least for the first one.

I don't think we're ever going to get a concensus on this issue short of some very convincing experiments with reproducible results showing a way to induce the AI to make Armies. C'est la Vie.

[alexman]

First of all, there is currently no way to make the AI use their leaders to build Armies. If there were, rest assured that it would be the preferred solution.

The warmongering advantage does not get lost by limiting Armies to one at a time. Even one Army is enough to plow through any defenses, plus you can always use it to rush improvements if you want to continue to get goodies from combat.

The balance between building and warmongering is already leaning heavily towards warmongering in Civ3. Trust me, there is absolutely no danger of taking players away from war just because of this change.

[Tall Stranger]

Originally posted by alexman
First of all, there is currently no way to make the AI use their leaders to build Armies. If there were, rest assured that it would be the preferred solution.

Then (assuming we don't hear anything from Firaxis indicating this problem has been addressed), we should either adjust the number of cities required to support each army or reduce the power of armies. Limiting each civ to one army at a time is just plain silly (IMO).

The warmongering advantage does not get lost by limiting Armies to one at a time. Even one Army is enough to plow through any defenses, plus you can always use it to rush improvements if you want to continue to get goodies from combat.

First off, I didn't say it gets lost, I said it gets reduced. That's just an obvious fact.

Second, it forces a tradeoff which is incomparable to one faced by builders. For warmongers, if you choose to use the army as an army, you can NEVER rush an FP, unless you luck into an SGL. That's a huge price to pay. You're essentially forced, at some point, to gamble by destroying your army (and rushing a build) only for the possibility that you'll get another one. Since most builders will use an SGL very soon after getting it to rush a wonder (and since they can have more than one at a time), that problem does not exist for them. It would be like telling a builder, "If you use an SGL to rush a wonder, you can never get another SGL unless you destroy the first wonder you rushed." Does that sound fair?

The balance between building and warmongering is already leaning heavily towards warmongering in Civ3. Trust me, there is absolutely no danger of taking players away from war just because of this change.

That's exactly my point: the point of the AU mod is NOT (IMHO) to balance game styles. YOU may feel that the balance leans "heavily" in favor of war. I strongly disagree. I find winning via culture (below Emp), diplomacy or SS absurdly easy, but you don't hear me calling for measures that will make the builder style harder or less playable. I just ask that you show the same restraint. (Please don't take this as an angry statement. I'll even stick on a few to prove I'm not POed!!)

[alexman]

Originally posted by Tall Stranger
That's a huge price to pay.

Actually, it's not.

Rushing the FP is not nearly as important as it used to be. I got many leaders in AU 501, but still built my FP from scratch. The reason (apart from the fact that Armies are so powerful) is that the FP now needs to be built in a city that's already half-way productive, if you want a productive second core.

[ducki]

I am sure I've mentioned this one, or something similar, but here goes, since I can't find it.

What if the Military Academy was similar the Manhattan Project - any civ that builds it opens it up for any other civ to build it? Or (probably not possible) any civ having an Army in the field enables any Civ with Mil Trad to build the Academy?

Something along the lines of if one civ opens the can of worms, all civs get to play with them?

Just thinking out loud, really. Anyone know of a way to create a Manhattan Project style trigger for Armies/MilAcad?

[pvzh]

What about removing requirement of victorious army from HE (make it require Literature, so AI will have some productive cities)? AI will definitly build it and have higher chances for leaders.

[pvzh]

And adding to this, although it is somewhat big change...

What if we take out ability of GL to build armies and replace it with ability to build a high HP unit. Because there only one unit can be build by leader, it has to be resource dependent and upgrdable.

Let say we start with a uber-Horsemen, than (uber-AC?), uber-Knight, uber-Cavalry, uber-Tank, and finally uber-MA. Because it is not technically an army it can not benefit from army and Millitary Academy boni and we are free to set cost (for upgrade purposes because these are not-buildable), att/deff/move, HP etc. I envision something like +1..3 att/def depending on tech.

Normal armies will be buildable with military academy.

[alexman]

Originally posted by pvzh
What about removing requirement of victorious army from HE (make it require Literature, so AI will have some productive cities)? AI will definitly build it and have higher chances for leaders.

Sorry, I'm not sure I understand how this would solve the problem. The problem is not that the AI doesn't get leaders, it's that it doesn't use them to build Armies.

What if we take out ability of GL to build armies and replace it with ability to build a high HP unit.

The AI will not use leaders to build this uber-unit either, since it doesn't use them to build Armies. Am I missing something?

[pvzh]

Yes that is true, but these uber-units will not be as powerful as armies; thus, player will not have a devastating weapon agains AI. I do not think AU can make AI build armies, so we must seek ways to reduce their power.

Let say uber horseman will be 6HP 3/1/2 (regular) this is not as good as horsemen army (3/1/3, 12 HP, blitz, healing in the enmy territory, radar).

These units have to be designed carefully because they are upgradable and could be put in the actual armies later.

[ducki]

I don't really follow - how do you build an army if you don't have an army to make the Military Academy available?

If you mean to remove the "Requires Army" flag from the Academy, I've been suggesting that for some time and with the new C3C changes, there's not much support.

It did make it into AU:PtW primarily, I think, because the best builder strategy was to be an all out warmonger - MGLs were way overpowered. Now that MGLs are limited to Small Wonders and Armies, the freely buildable Academy lost much of its support.

As hard as it is (for me) to generate an Army, I really dislike the idea of completely outlawing early armies for the warmonger. Maybe that's just me, though.

[nbarclay]

Originally posted by Tall Stranger
I find winning via culture (below Emp), diplomacy or SS absurdly easy, but you don't hear me calling for measures that will make the builder style harder or less playable.

And you win those ways without doing any significant fighting? For me, no matter what victory condition I ultimately pursue, I tend to do a good bit of fighting along the way. (My first diplomatic victory came when I eliminated the other civs in my hemisphere, setting up a situation where the only civs left in the game were half a world away and got along with me nicely. )

[nbarclay]

Here's a radical idea: could we take away the ability of military leaders to rush improvements with interfering with the uses of scientific leaders? It's not an idea I really like, but it might be a way to get the AIs to build armies.

By the way, from a game I've been playing, I get the strong impression that saving a SGL to use later prevents a player from getting a MGL. (Either that or I got remarkably unlucky with leader generation in dozens of battles won by elite tanks.) If having an SGL really does make it impossible to get a MGL, that seems a bit absurd to me.

Nathan

[Jaybe]

Here's a radical idea: could we take away the ability of military leaders to rush improvements without interfering with the uses of scientific leaders?

I tried to test this, but I could not place a SGL on the map. Placing a Leader applies a MGL.
Would alexman have the patience to create a scenario and wait for a SGL to test it? I know I don't.

[Aqualung71]

Originally posted by nbarclay
Here's a radical idea: could we take away the ability of military leaders to rush improvements with interfering with the uses of scientific leaders?

That sounds pretty interesting. And simple. And it may be a compromise for some of the radically different opinions we've generated here, as long as you accept that you have to build your FP yourself unless you use an SGL.

Is it possible?

[alexman]

Originally posted by nbarclay
could we take away the ability of military leaders to rush improvements without interfering with the uses of scientific leaders?

Unfortunately not. The AI leader strategy can be selected only if the unit has the ability to build Armies and rush improvements. If the unit doesn't have the AI leader strategy, the AI does nothing with it.

By the way, from a game I've been playing, I get the strong impression that saving a SGL to use later prevents a player from getting a MGL.

Yes, this is a fact. I have tested it.

[Tall Stranger]

Originally posted by nbarclay

Quote:

Originally posted by Tall Stranger I find winning via culture (below Emp), diplomacy or SS absurdly easy, but you don't hear me calling for measures that will make the builder style harder or less playable.

And you win those ways without doing any significant fighting? For me, no matter what victory condition I ultimately pursue, I tend to do a good bit of fighting along the way. (My first diplomatic victory came when I eliminated the other civs in my hemisphere, setting up a situation where the only civs left in the game were half a world away and got along with me nicely. )

Not always, but sometimes. I've played games where I restricted myself to one, ancient era war and then only defensive wars (no territorial gain). I've also played purely peaceful (and nasty warmonger) games. It depends on the game and my mood. My point was that there seems to be an underlying, and IMO deeply flawed, set of assumptions floating around this discussion that:

1. Warmongering is, in and of itself, too easy;
2. The added strength of armies makes warmongering even easier; therefore
3. We need to make it harder for the human to win this way.

My point was that one can make a case that the builder-style game is no more challenging than the warmonger-style. Limiting warmongers to 1 army is, IMO, blatantly unfair, as there is no corresponding limit on builders.

My recommendation is that we not make any significant change to armies until we are certain Firaxis is not going to fix the problem of the AI not creating them. If it becomes clear that they aren't, then we make some comparatively minor changes (raising the cities/army ratio, for example). The restriction that alexman has proposed is, IMO, just too radical for the AU mod.

BTW, interesting that the presence of SGLs makes it impossible to generate an MGL. Never would have thought they were related.

[ducki]

Can't you have multiple SGLs at a time? Or am I imagining that? If you can, then the SGLs blocking MGLs makes even less sense. If I'm imagining it, can I have some salt for my foot?

[alexman]

The rule is this: you can't get a leader from a battle if you have any kind of leader active. It doesn't make much sense, but it happens because all types of leaders are actually the same unit.

[nbarclay]

Originally posted by alexman
The rule is this: you can't get a leader from a battle if you have any kind of leader active. It doesn't make much sense, but it happens because all types of leaders are actually the same unit.

Is that something that it would be practical for us to fix in the AU Mod? To me, that feels a lot more like a bug than like a feature that exists for good reasons, especially when having two SGLs at the same time is possible. (I've had two SGLs at the same time before myself.)

[ducki]

This sounds like some hard-coded spaghetti to me - a remnant of the old GLs where you could only have one at a time. Instead of adding a new unit type, Firaxis seems to have just added code to the old unit type with some sort of flag/checking on the method of generation - if I was timecrunched, I'd probably have done something similar, though I'd hate having to for reasons we're finding now.

This should be consistent, IMO - either you can have multiple leaders at a time or not(with the current code) or the same options on a per leader-type basis if we could get Firaxis to separate the two.

Personally I prefer one at a time on a per type basis, but from the little poking around I've done in the editor looking for a solution to the missing AI armies, I don't see a way for us to fix this. N.B. I am not an editor wizard, so I might have missed something.

[lockstep]

I'm not sure if this has been mentioned already, but another (and quite conservative) method to reduce the power of armies is to remove the 'increased army value' flag from the Military Academy. This should reduce the attack/defense bonus by one third (i.e., from 75% to 50% for a 3-unit-army, and from 100% to 67% for a 4-unit-army).

[alexman]

That's a good idea.

However, I think that the best approach here is to wait until a patch teaches the AI how to build and use Armies. If anybody else from the AU mod panel thinks otherwise, please put a proposal under consideration.

[Aqualung71]

Do you know whether Firaxis has a "fix" under consideration for a patch?

[alexman]

No, I don't. I just hope.

[ducki]

We could give each AI one Army as a starting unit, except that they load a single warrior and garrison the capitol with it. Not sure if they would ever use it on offense, since I abandoned that Debug-game after a few thousand years.

Should their capitol ever get attacked and the army win, though, they could build the MilAcad.

Here's hoping Firaxis has this fairly high on the list.

[lockstep]

Originally posted by alexman
I think that the best approach here is to wait until a patch teaches the AI how to build and use Armies.

I agree. Any changes to armies and the Military Academy ( ) should be 'deferred' for now.

[jshelr]

Good discussion above.

Philosophically, the AU mod departs from a product that would be sold to newbies because we want more challenging conditions rather than a fun game. (When we want a "fun" game, we could always play stock rules.)

I think the basic problem here is the leader concept itself. If there were no MGLs, competition with the AI would be enhanced due to the humans better ability to place the FP and use armies. (SGLs probably help the AI at higher levels, at least until the industrial era.)

I would not recommend eliminating leaders from civ but I've got no problem with a mod that does this.