Apolyton University Mod (Thread II)

[nbarclay]

Okay, Dominae is already cooking up our next AU game, so let's go ahead and pin down what we want in the AU Mod this time around.

Restore original stats to French Musketeer. The two movement points, instead of the extra attack factor, has made France perhaps too powerful, as the commercial trait has been improved.

Definitely. Faster-movement-rate musketeers coupled with the Commercial trait would be too powerful on a huge map.

Remove the optional status from several Middle Age and Industrial Age technologies. Printing Press, Navigation, Nationalism, Sanitation, Advanced Flight, and Amphibious Wafare. The AI usually researches these optional technologies, even if they are not useful, thus allowing the human to get an advantage by speeding through the tech tree to the next age. On the other hand, removing the optional status also removes options to the human, which is against another of the mod's goals.

I think we have a consensus that making Printing Press mandatory would be okay. Otherwise, the topic is still too much under debate for it to be time to incorporate changes yet.

Move Pyramids back to Masonry. The change was made to help the AI choose Mathematics so it's not so easy for the human to get this technology and then get rich by trading it. But the change in gameplay, along with the fact that the AI gets started on the Pyramids later perhaps make the change a bad idea for this mod.

Considering how much I'd like to leave the Pyramids with Mathematics so I'll have a reasonable shot at building them, it would probably be good to move them back.

Remove zero-range bombard ability from archers. This ability has the undesirable effect of helping early human archer rushes. On the other hand, it also helps the AI who build Archers even when Swordsmen are available, and adds an element of combined-arms warfare to the ancient age.

I'm okay with leaving zero-range bombard for archers in for this game, unless Dominae is planning to crowd us.

Celts can no longer build Medieval Infantry. Gallic Swordsmen should not "upgrade" to Medieval Infantry, as the latter are cheaper. The would upgrade directly to Guerillas.

I don't like the idea of making the Celts the only civ with a swordsman-replacement UU that can keep building (and, perhaps more importantly, upgrading to) its UU after feudalism.

Fix catapult upgrade bug. They should upgrade to cannon.

Did you even need to ask? Definitely fix the bug!

[alexman]

Originally posted by nbarclay
I don't like the idea of making the Celts the only civ with a swordsman-replacement UU that can keep building (and, perhaps more importantly, upgrading to) its UU after feudalism.

I can accept the agrument of not making the change because it doesn't help any of the mod's goals, but I'm not sure I understand your concern.

In the middle Ages, do 50-cost Gallic Swordsmen give the Celts an advantage over 40-cost Medieval Infantry? Without crunching any numbers, I would say that these two units are more or less of the same effectiveness against pikemen and musketeers. Remember, a warrior upgrade to Med. Infantry is cheaper (60) than to GS (80).

So if we accept that the Gallic Swordsman is no better than the Med. Infantry, if the Celts have already had their GA, then they get no benefit by being able to build their UU in the middle ages. On the other hand, if they have not yet had their GA, then they can build Gallic Swordsmen anyway.

Again, I realize that this does not directly help any of the mod's goals, so technically we should not make the change. It just seems awkward to actually lose value towards the eventual guerilla upgrade when you upgrade a GS to a Med. Infantry unit.

[Dominae]

I'm with Nathan on all the proposed changes (for now!). alexman, let me know when you're done, so that I may finish the scenarios for AU207.


Dominae

[nbarclay]

Originally posted by alexman

Again, I realize that this does not directly help any of the mod's goals, so technically we should not make the change. It just seems awkward to actually lose value towards the eventual guerilla upgrade when you upgrade a GS to a Med. Infantry unit.

So what? The fact that you can upgrade GS's to medieval infantry in no way implies that you have to. Thus, the only real effects of the change are (1) to let Celts keep building and upgrading warriors to Gallic Swordsmen longer and (2) to take away the option of upgrading GS's to medieval infantry if someone does want to.

Edit: At least that's true from a human perspective. There might conceivably be a case for the change if AIs tend to throw away shields upgrading to medieval infantry, depending on which type of unit one regards the AI as better off with.

[alexman]

I don't think the AI is smart enough to know when to upgrade. My guess is that it upgrades anything as long as it has enough cash and the unit happens to be in a city with a barracks. In this case it would probably upgrade any GS in such a city, since the upgrade cost is zero. I will test this later.

Also, would it be a huge no-no if we reduced the cost of Mathematics? That way it would be less tempting for the human to research it at a 40 turn pace, and more appealing to the AI (about as appealing as Polytheism).

We could increase the cost of Map Making (which is very high on the AI priorities anyway) to keep the total beakers of the Ancient Age constant.

[nbarclay]

Since Dominae already has a map for AU 207, I don't want to even think about shuffling around tech values for this game. There simply isn't time to give the issue the kind of thought and discussion it deserves. And besides, AU 207 AARs might provde additional data to feed our discussion.

For the longer term, what evidence is there that human players' researching Mathematics on a 40-turn basis provides an unbalancing advantage compared with other strategies human players can choose from? The fact that a strategy is viable and has certain advantages, or even that it might be best in certain situations (e.g. a starting position with very little commerce), does not by itself make it enough of a problem to justify rules changes. Indeed, if the option is viable but not seriously unbalanced, we would actually be taking away strategic choices in direct violation of the AU Mod's objectives.

Nathan

[alexman]

It's not that the strategy is unbalancing, it's that it is a strategy that is not available to the AI. The fact that the human player knows that the AI doesn't research Mathemetics until late, gives him an extra advantage over the AI.

One of the goals of the mod is to improve the AI. Reducing the cost of Mathematics would make the AI more likely to research Mathematics earlier, so the window of opportunity for the human to take advantage of selling it to the AI is smaller.

[alexman]

After a quick test, it seems that the AI instantly upgrades all GS after they have triggered their GA, but keeps them around if they have not.

[player1]

Originally posted by alexman
Restore original stats to French Musketeer.[/B] The two movement points, instead of the extra attack factor, has made France perhaps too powerful, as the commercial trait has been improved.

With not try them with attack of 4 instead (like I did in my owm mod).
It surely won't be unbalanced (Samurai is 4/4/2, cost 70).

Also, it would benefit the AI, since he could use them offensively.


P.S.
Of course, it would be even better if all musket would have cost of 50 shields instead, but that's another story.

[nbarclay]

Originally posted by alexman
After a quick test, it seems that the AI instantly upgrades all GS after they have triggered their GA, but keeps them around if they have not.

Which raises the question of whether doing so is good or bad for the AI. If it's bad, removing the upgrade path could be justified under the goal of improving the AI. But the lost value toward upgrading to guerillas is irrelevant if poor use of GS's during the medieval and early industrial eras results in their not surviving to be upgraded.

Now that we're on the subject, we might consider how Rome is affected as an opponent by having it upgrade legionaries to medieval infantry.

In any case, these are best kept as topics for the AU 208 version of the mod.

[nbarclay]

Originally posted by alexman
It's not that the strategy is unbalancing, it's that it is a strategy that is not available to the AI. The fact that the human player knows that the AI doesn't research Mathemetics until late, gives him an extra advantage over the AI.

One of the goals of the mod is to improve the AI. Reducing the cost of Mathematics would make the AI more likely to research Mathematics earlier, so the window of opportunity for the human to take advantage of selling it to the AI is smaller.

Unless the advantage provided to the human is unbalancing, leaving the strategy available to humans does not seriously undercut AI competitiveness. Therefore, taking the option away from human players reduces the range of strategic choices on the human end (contrary to the purposes of the AU Mod) with no significant benefit in terms of enhancing AI competitiveness. In my view, "The AI doesn't know how to do such-and-such," in and of itself, is nowhere near enough to justify changing the game.

Nathan

[alexman]

I updated the initial posts with version 1.17

I sneaked in the Math/Map Making cost change because I think it's a good one. We can always change it back after the next AU game.

[nbarclay]

Alex, I very much appreciate your work as coordinator and keeper of the AU Mod, but when you propose something, the only post by anyone else regarding it is opposed (at least for the time being), and you add it anyhow the same day it was first proposed, there is a very real danger that you are creating the Alexman Mod instead of the AU Mod. It seems to me that the AU Mod should be basically conservative, making changes from the stock game only when a fairly strong consensus supports the changes.

Nathan

[Nor Me]

The potential problem of altering Rome's upgrade path is whether Longbowmen are built instead of Legionaries in significant numbers.
The AI might then build few Legionaries for offence while building them instead of Pikemen for defence.

As for the Celts, it seems unlikely that many Med Inf would survive to be upgraded unless there is peace. Nor would I expect many GS to be built after Feudalism if possible unless the AI had no horses. As far as I can tell, the major source of Med Inf for AI civs is upgrading.
So it would make sense to leave it as it is only if the Med Inf were a better unit than the GS. Or if the Celts end up building more longbowmen.

[Dominae]

For the sake of getting AU207 up and running, I'm going to avoid any discussion of the latest changes in 1.17, and just use this latest version in the modded scenario. Since AU207 will take some time, perhaps the concurrent course could be "Seminar - The Philosophy of the AU mod", or somesuch.




Dominae