Apolyton University Mod (PTW version) - Page 13

joncnunn · January 15, 2003, 14:56

In that game: the enemy stacks totaling 20 Infenty were in my territory for about 10 turns. They were on improved tiles (roads) the entire time, about 60% of the time were inside a CR.

I showed no WW at the end of that timeframe.
Civ III Classic, 1.21f.

alexman · January 15, 2003, 15:08

Help!
We're being threadjacked!

Catt · January 15, 2003, 15:34

Quote:

Originally posted by alexman
Help!
We're being threadjacked!

No Longer! Click HERE for a revived WW thread and let's leave the AU Mod thread to its proper topic!

Catt

Theseus · January 16, 2003, 20:33

Bringing the discussion together:

The improvements to Communism make it FAN-BLOODY-TASTIC for the inner warmonger in me!!!

I'm haven't quite gotten to Motor Trans in AU 203, but I've been at war forever, having just gotten to max acceptable WW under Republic when I was able to convert to Communism.

WOO HOO!!

Dominae · January 17, 2003, 00:30

alexman, I finally got around to capturing that Monarchy shot you asked for (you'll need to go back a couple of pages to compare to Communism). Basically they're equivalent economy-wise, apart from the fact that I'll start incurring some military support costs under Monarchy (I had not started building Panzers yet at the time of the switch). My Palace is still in Berlin (starting location), and the FP is nicely placed a few tiles South of that.

So, is this how you think Communism and Monarchy should compare to one another?

Dominae

BRC · January 17, 2003, 00:33

Wow. Those were pretty close.

alexman · January 17, 2003, 01:11

Thanks Dominae!
It looks encouraging.
A better FP/capital placement would make Monarchy the government of choice, so the change did not overpower Communism.

For the AI (or the human with a bad FP placement), the two governments are about the same. Otherwise, Monarchy is better, just as it was before. It seems as though this is as much as we can improve Communism (and thus help the Communist-happy AI) without changing the feel of the game. This was our goal.

Too bad we can't keep the change...

Dominae · January 17, 2003, 01:45

Actually, given the restriction on not attacking anyone and the particular geography, I think the FP was in a pretty good position in my game. But the FP's proximity to my capital does reflect the AI's tendency, so your points all make sense.

I guess we (or you!) will have to keep working to try and find a Communism that resembles this, but without any bad side effects. There are rumblings of a new patch coming out: maybe all our wishes will come true then.

Maybe.

By the way, give yourself a nice pat on the back for taking Communism this far. You deserve it.

Dominae

=DrJambo= · January 17, 2003, 04:58

Have you considered increasing the number of shields (currently 20) per pop sacrifice OR decreasing the unhappiness (currently 20 turns) per pop/sacrifice as a way of boosting communism?

Or making Communist workers work at the same pace as Democratic ones?

I can't remember what this mod's done to Communism already!? Maybe someone can remind me. And why can't you keep the changes?

alexman · January 17, 2003, 11:33

Dominae,

This mod (including the Communism changes) is a result of the endless debate, testing, and ideas from all of us here. However, this is the closest thing to a pat on the back I could find in the Apolyton smileys...

Dr. Jambo, the readme is in the beginning of this thread.

Making pop-rushing more powerful is a possibility, but we need to be careful because Despotism would get the same boost as Communism. The latest Communism change that we made (and are not keeping) was to reduce flat the corruption of Communism by 1/3. The side-effect was that the distance corruption of Despotism was also reduced by 1/3, making it too close in economic performance to Monarchy.

Stuie · January 17, 2003, 11:39

What about changing Communism to rush with gold instead of population - or does this change the "flavor" too much? Would this encourage larger AI cities? Is the AI actually using the pop rush, or are its cities self-destructing merely from drafting them to death?

Bluefrog · January 17, 2003, 12:28

Quote:

Originally posted by Stuie
What about changing Communism to rush with gold instead of population - or does this change the "flavor" too much? Would this encourage larger AI cities? Is the AI actually using the pop rush, or are its cities self-destructing merely from drafting them to death?

In my AU203 game, which I'll eventually post a full AAR, I ended with a huge treasury, which was useful to a degree, and little means of rushing the later improvements, due to only being able to sacrifice up to half a city's pop for any pop rush.

But...I think making Communism a goldrusher is a bad move. With all my extra cash, I was able to use the gold on things I rarely used it for in the other governments. Namely, espionage. I stole probably 10 cities from enemy civx through propaganda, stole tech, stole troop locations, etc.

I also had two instances where I highly suspect the AI used propaganda on me, but there are things to make me doubt this. I'll post more later...time to go do the "hunter/gatherer" schtick at Chick-fil-A.

Fosse · January 17, 2003, 12:56

All of the extra gold for espionage surely makes Communism more fun for the player (I wouldn't know, I lost my 203 game during the early middle ages). But it really doesn't help the AI (maybe they flipped a few cities... but at a huge cost, and if they successfully flipped two of yours, imagine how many times the failed).

I would still argue for making Communism use gold rushing, for the AI's sake. First off, when being invaded a communist AI will draft and rush their population out of existence, creating small cities with HUGE unhappiness. This means very unproductive cities, especially since it costs so many pop points to rush anything in the industrial or modern era.

If the AI could spend their gold instead they could keep those pop points and have happier and more productive cities.

If we're worried that AI commies would be too cash poor to rush their units that way, then maybe as a balance you could up communism's draft rate by one, since drafting is a more efficient use of converting population to units than pop rushing is.

=DrJambo= · January 17, 2003, 13:42

not if you increase the amount of shields you get awarded per pop-rushed citizen? To say 25 or 30?

Catt · January 22, 2003, 18:19

I'm curious about everyone's thoughts with artillery having a defense of 1 (and therefore being destroyed instead of captured). My only experience has been in AU 203, but I had a few take-aways from that experience. IMHO, the change doesn't appreciably help the AI, may in fact hurt the AI, doesn't appreciably diminsh the inherent human advantage around artillery, and adds some very small minor annoyances. Did others take away a more positive view of the change?

My conclusions were shaped by the following observations. I didn't have any sense that the AI was building any more or less artillery and I didn't have any sense that the AI was using artillery any differently. On the relatively few occasions where I fought against AI artillery, the only change in behavior was for me to try and find an old cavalry unit (or any old unit with sufficient movement to get to the arty) to kill it. Based purely on my well-known faulty memory, I'm guessing I might have captured a total of 10 units had the change not been made -- hardly, IMHO, a boon to the human. On the other hand, on the few occasions where the AI was going after my poorly defended (or even adequately defended) artillery, it would divert some of its best attackers (tanks) when such forces would have been much better used on alternative targets (this behavior is probably not affected by the arty change) -- but the downside to the change became apparent when China actually threw an unexpected force against a small arty stack covered by one infantry -- the infantry fell, and then so did my 4-arty stack. I lost 4 arties, but was rewarded with 4 Chinese tanks stuck without movement points on my rail network without cover (instead of one Chinese tank and 4 disbanded or captured-back arties). Those 4 defenseless tanks actually represented a sizeable percentage of offensive Chinese forces at the time. While this could be exploitable (that's within our control, obviously) I raise the issue more as an inadvertant "exploit" -- another example of perhaps less than stellar AI "though processes." (it is so rare, at least for me, to lose an arty stack of any size, perhaps this isn't that big an issue).

On the "minor annoyance" piece of the argument (and I stress minor), I got sick of moving a small stack of arty along rails, only to have an enemy blitz unit take potshots at the stack and slow things down considerably. Since the game defaults arty movement to the beginning of the turn (when perhaps many of us use it?), it's easy to bump into blitz units. An obvious solution to this annoyance is to watch for blitz units along the path

but, unlike with regular attackers or defenders, the blitz pot-shots meant absolutely nothing to the effectiveness of the arty because it wasn't engaging in combat where an extra HP might be important -- my arty would just stay injured, perhaps red-lined, until a turn came along when it wasn't needed - instant heal and then restart the process. So - I could manually check my intended path before every arty move and avoid pot shots; or I could go ahead and move stacks and da*n the pot shots as meaningless (except of course as they tended to slow the game down a bit) - either way added one little bit of micromanagement or delay.

Other thoughts?

Catt

alexman · January 22, 2003, 18:28

Another side-effect that I saw was that AI artillery becomes an excellent leader-generator for humans. An elite victory with minimal risk to your elite.

Good points, Catt. I agree that this change needs to go. Even if it weren't just a "minor annoyance", it is at least an unnecessary change.

player1 · January 22, 2003, 18:38

I told you so!

Dominae · January 22, 2003, 19:13

The part about stacks of Artillery being able to defend (although poorly) is right on. It takes quite a few AI Cavalry to capture a city with 20 Artillery and 2 Riflemen as defenders. Usually there is time to bring in reinforcements.

I could be wrong, but I think the new Artillery also can count as MP. This is another big change from stock Civ3 (one which the AI does not take advantage of).

But I disagree that captured Artillery is not a problem. Catt, you said you would have capture 10 of them in AU203? My figure is more around 30. Given that you captured bombard units cost no Shields (+ time) to produce, and upgrade at minimal cost, this is definitely a weakness in the game.

Still, the change has to go.

Dominae

Theseus · January 22, 2003, 21:05

Agreed, across the board.

I had a 3xMA Persian Army take out a 4xInfantry Army protecting about 20 Arty... no follow up despite the fact that all I had left for protection was 1-2 vet Infantry... Persia just didn't have enough units!! I actually would have LOVED to have seen the AI accomplish such a capture.

Catt · January 23, 2003, 10:57

Quote:

Originally posted by Dominae
But I disagree that captured Artillery is not a problem. Catt, you said you would have capture 10 of them in AU203? My figure is more around 30. Given that you captured bombard units cost no Shields (+ time) to produce, and upgrade at minimal cost, this is definitely a weakness in the game.

Still, the change has to go.

I completely forgot the MP use - great for both contentedness and flip suppression. Also, I'm glad it seems I wasn't wandering in the darkness, and that others had the same reaction, so the change should go.

I don't want to debate a moot point, but I think a similar argument wrt other game issues has come up before and will probably come up again and so will indulge in at least one response

(though for the life of me I can't remember the exact circumstances that rest on the same line of thought). I agree that 30 artillery represents a lot of shields and a fair amount of production time when looked at in isolation -- but in my experience, the AI rarely has more than 2 arties in any one city, and usually has 1 or none. If you can capture 30 arties, I'd speculate that you're also capturing 30+ AI cities, at which point the value of captured arty relative to your power and standing in the world is pretty small, regardless of the shields represented. I just don't think that AI arty captured by the human is the big problem with arty we should try to solve for (though I think I also believe that only Firaxis can help with the inherent imbalance between human and AI use of artillery).

Catt

Dominae · January 23, 2003, 11:29

Catt, point taken about the number of cities taken if a large force of Artillery are taken. But the situation is not as drastic as you indicate. Starting with Catapults up to Artillery, and across many civs, it is not uncommon to get around 20 bombard-type units with minimal city-capturing (or, at least, minimal gain in relative strenght, especially if the AI is powerful). A stack of 20 Artillery is "just right" to do some serious damage throughout the game. The fact that it comes free is unfortunate. This is only one of the facets of the "AI and Artillery" problem, which is tough to fix.

Dominae

ducki · January 23, 2003, 11:56

I don't think the free Artillery is any worse than the free workforce that many people figure into their gameplan.

I would venture to guess, according to some reports in this game, that workers are actually more of a "problem", but at least now you can't buy them for 30 gold.

Also, correct me if I'm wrong, but in your example of minimal city-capture spread across AIs(oscillation), you actually are gaining a lot in relative strength due the the pruning effect. The fact that the AI "wasted" time on Catapults would be more the problem, not that a series of pruning wars netted you a nice stack of bombardiers.

Again, correct me if I'm wrong or miss the point, but I see the free artillery problem as more a symptom of poor build-queueing(sp? ) by the AI as well as yet another nod in the general direction of warmongering as the "easiest" way to a win, and no worse or more unbalancing than free workers.

Maybe I misunderstood your point about minimal city-capture wrt relative strength, but in other threads here it's intimated that you actually gain more strength by spreading your wars around than by concentrating on one foe.

Dominae · January 23, 2003, 12:07

ducki, I'll address your points in another thread, because alexman will crack the whip if he even smells a threadjack.

Let me just state the obvious: bombard units are a problem in Civ3, and we should be looking for a fix that helps the AI in this regard. Obviously this problem is a symptom of other things, but that's out of our hands.

Dominae

nbarclay · January 23, 2003, 12:30

Quote:

Originally posted by Dominae
Catt, point taken about the number of cities taken if a large force of Artillery are taken. But the situation is not as drastic as you indicate. Starting with Catapults up to Artillery, and across many civs, it is not uncommon to get around 20 bombard-type units with minimal city-capturing (or, at least, minimal gain in relative strenght, especially if the AI is powerful). A stack of 20 Artillery is "just right" to do some serious damage throughout the game. The fact that it comes free is unfortunate. This is only one of the facets of the "AI and Artillery" problem, which is tough to fix.

Of course those "free" artillery that start their lives as catapults come with upgrade costs and centuries of maintenance costs (unless you stay in Monarchy and under the unit limit). Those costs reduce the imbalance somewhat.

alexman · January 23, 2003, 13:50

Speaking of artillery, the Koreans must be bummed. Since they can't build cannons, they can't capture them either! I believe defeated AI cannons get destroyed instead, just as scouts get destroyed if attacked by non-expansionist civs.

Should we change this? If we make Cannons upgrade to Hwach'a (right now it's the other way around), and give the Koreans the ability to build Cannons, any such units captured in the field will not be destroyed. Instead, they can be upgraded to Hwach'as for zero gold. Koreans will still not be able to build Cannons, since their UU has the same tech and resource requirements, but it higher in the upgrade chain.

Another thing about the poor Koreans: their UU cannot trigger a GA because you can't actually kill anything with a bombard unit. Should we add lethal bombard to the Hwach'a? Would that be exploitable?

player1 · January 23, 2003, 14:26

Leathal?

No way!

player1 · January 23, 2003, 14:28

Anybody tested that thing about Koreans unablity to capture Cannons?

Dominae · January 23, 2003, 14:31

Against the AI lethal bombard would be overpowering. Against humans it would create very interesting strategic situations. I vote for no change with respect to this mod, and lethal bombard (for the Korean UU) in the MP version.

Dominae

ducki · January 23, 2003, 14:47

What about making bombard units like workers...
...in that they retain their nationality, with the inherent (slight) reputation hit, and operate at half effectiveness?

Is that something that can be done in the editor?
Is that still within the spirit of AU?

P.S. to Dom - I realize that the bombardiers are a problem, I just - personally - haven't experienced it to be any more of a human advantage than slaves. I'm sure it's just due to relative inexperience, but it's still my personal opinion.

alexman · January 23, 2003, 14:51

OK, no lethal land bombard for Hwach'a. Although I'm not convinced that it would be more overpowering than, say, the Ottoman UU.

How about lethal sea bombard? Too unrealistic?

In any case, let's change the upgrade path so that cannons can be captured by the Koreans.