Apolyton University Mod (PTW version)

[alexman]

Originally posted by nbarclay
Weakening the Great Lighthouse would have two side effects, one positive and one negative. [...] On the negative side, civs caught on the wrong side of, for example, a 5/3 split would be caught outside the main tech trading loop longer.

I don't think this is a problem. Overcoming a bad start is very satisfying IMHO. Civ3 is designed so that even if you are way behind in tech, you can catch up. (Heck, you can even win wars using a sufficient number of outdated units). If we were woried about bad starts, we should have also given a boost to jungle and desert tiles, for example.

In regard to the specific proposals, I think they greatly devalue legitimate use of caravels for short transoceanic hops.

But still, Caravels would be much more valuable than Galleys. With the current rules, Caravels are essentially Galleys that don't sink in Sea tiles. The change I proposed gives Galleys an extra movement point, and cuts the movement cost for Sea tiles in half, in addition to giving safe passage in Sea tiles. Caravels become obsolete very fast anyway. Their job would be to expore the ocean tiles that were inaccessible before Astronomy, not to be used after Navigation.

Nor do I view suicide caravels as a problem. Consider the attrition rate in voyages like Columbus's and Magellan's and there's plenty of precedent for risking the loss of caravels in order to explore. But the cost of suicide caravels isn't so low as to make sending them out obviously the right choice unless a civ is pretty desperate.

The problem is not realism or gameplay. It's that the AI never uses suicide ships of any kind, so it's a choice available only to the human.

last I saw, AIs were perfectly willing to escort galleys and caravels with frigates if they didn't have an ironclad handy.

Yes, but they also slow down the Galleon if they are escorting it with a slower Ironclad. AFAIK the AI doesn't know to escort Galleons with something that has equal movement.

A more minimalist approach to reducing early contact would be simply to take the "Safe Sea Travel" flag away from the Great Lighthouse (and probably cut its cost by 100 shelds since that's a huge chunk of its value).

Originally posted by Dominae
How about leaving the "safe sea travel" power of the Great Lighthouse intact and removing the bonus movement?

These are both definitely worth considering.

[punkbass2000]

Re: New Naval Units Proposal

Originally posted by alexman [*] Increase movement cost of Sea by 1, and movement cost of Ocean by 2.

Oddly enough, I'd recently been thinking about modding movement points for units generally, so that old units become exceedingly slow as well as weak, while modern units have significantly more MP, but terrain values are higher. I think this could work if I made terrains and units for groups of civs on different continents who wouldn't be able to reach each other until later in the game than normal. If each group's units used different MP on terrain on continents that aren't their own, then the old units might be able to do something defensively on their home continent, but that's about it.

[alexman]

Alternative proposal [changes relative to AU mod]:

• Increase movement cost of Ocean by 1.
• Give all ships except Galleys the ability to ignore movement in Ocean.

Galleys without the Lighthouse are not really affected, except that suicide runs in the Ocean (not Sea) are made more difficult.

Galleys with the Lighthouse could get two tiles deep in Ocean and back safely before. Now they can get only one tile deep and back safely.

Caravels and everything after that are unaffected.

[Dominae]

Sounds good. I like the minimalist approach, and this one also has the effect we're looking for (reducing the potency of Lighthouse Galleys).


Dominae

[nbarclay]

Originally posted by Dominae
How about leaving the "safe sea travel" power of the Great Lighthouse intact and removing the bonus movement? It's the combination that really makes the Wonder a no-brainer for human players. This would have the nice side-effect of making Magellan's unique with respect to its ability.

Depends on how far you want to go. In terms of being able to safely explore long distances with galleys, the safe sea movement is actually the more powerful of the two effects, so removing it would do more to limit early contact. Another thing worth noting is that if the Great Lighthouse only provided safe sea movement for galleys, it would effectively expire with Astronomy, at which time galleys get safe sea travel anyhow.

[nbarclay]

Originally posted by alexman

I don't think this is a problem. Overcoming a bad start is very satisfying IMHO. Civ3 is designed so that even if you are way behind in tech, you can catch up. (Heck, you can even win wars using a sufficient number of outdated units). If we were woried about bad starts, we should have also given a boost to jungle and desert tiles, for example.

It's a matter of taste. The reason I don't play Deity is that I don't like having to come back from way behind.

But still, Caravels would be much more valuable than Galleys. With the current rules, Caravels are essentially Galleys that don't sink in Sea tiles. The change I proposed gives Galleys an extra movement point, and cuts the movement cost for Sea tiles in half, in addition to giving safe passage in Sea tiles. Caravels become obsolete very fast anyway. Their job would be to expore the ocean tiles that were inaccessible before Astronomy, not to be used after Navigation.

If ships could be upgraded on the fly no matter where in the world they are, I would consider it good to give caravels a huge advantage over galleys. But because they can't, I think a system that gives galleys most of the abilities of caravels when Astronomy comes makes a huge amount of sense. (They still don't get the extra transport capacity or the extra defensive hit point.) Keep in mind that turns during that period are typically ten years, maybe as little as five, so it's only a quirk of the movement system that leaves galleys many turns away from a harbor where they can be upgraded. (And the same goes for safe ocean travel for galleys after Navigation/Magnetism.)

Also note that the replacement for caravels comes with Magnetism, not Navigation. A beeline to Navigation can give caravels safe ocean travel long before Galleons are available to take their place, so the "shelf life" of caravels isn't always all that short.

The problem is not realism or gameplay. It's that the AI never uses suicide ships of any kind, so it's a choice available only to the human.

On the other hand, from what I've read, the great deity Soren gives AIs knowledge of the safe crossing points without having to engage in the kind of long, complex search a human player has to. I'm inclined to view those two factors as more or less canceling each other out. In some situatins, the AI has the advantage, while in others, the human has the advantage.

[nbarclay]

Duplicate post deleted.

[nbarclay]

Originally posted by alexman
Alternative proposal [changes relative to AU mod]:

• Increase movement cost of Ocean by 1.
• Give all ships except Galleys the ability to ignore movement in Ocean.

Galleys without the Lighthouse are not really affected, except that suicide runs in the Ocean (not Sea) are made more difficult.

Galleys with the Lighthouse could get two tiles deep in Ocean and back safely before. Now they can get only one tile deep and back safely.

Caravels and everything after that are unaffected.

Better, but it still messes up the continued use of galleys for exploration after Astronomy and especially after Navigation/Magnetism. Granted, it's unrealistic to have galleys around in the industrial era. But in a game where it can take many decades for ships to get back to port for upgrade, I think having an automatic upgrade to galley abilities makes a huge amount of sense, and I hate to see that undercut.

[alexman]

How long can it take to get your galleys back to port? Make them take a Coastal/Sea route, or lose a MP in Ocean until they get there.

After Astronomy Galleys are identical to Caravels in Sea and Coast, and have one less movement in Ocean.

[nbarclay]

Originally posted by alexman
How long can it take to get your galleys back to port? Make them take a Coastal/Sea route, or lose a MP in Ocean until they get there.

After Astronomy Galleys are identical to Caravels in Sea and Coast, and have one less movement in Ocean.

Getting galleys back to port can take quite a while if you have the Great Lighthouse and have had good success finding mostly sea routes between land masses, especially on maps larger than standard. (For that matter, it's not completely trivial if they're on the far side of a continent from your civ.) And then after the upgrade, they have to go back to where they were to pick up their exploration where they left off.

[Fosse]

I support alexman's proposal of 2 cost sea and 3 cost ocean.
a change that makes for a more historically accurate effect of not traipsing across the globe in the ancient age and the excitment of maybe discovering a new world, while also balancing out a human/AI difference? I'm for it. (the idea that AI's knowledge of short routes offsets our suicide gallies doesn't fly with me... because if they don't have a short route they don't try... but we keep trying until we make contact).

This also prevents the "sea-hopping" ability that we have... since the map maker places sea tiles in the middle of ocean sometimes, humans can navigate vast expanses of water far soon than they should be allowed to.

Getting stuck on the short side of a 3/2 split? Or on an island in an ocean? So be it. Every game is different... restart if you have to (many of us do anyhow).

I don't mind disbanding my exploring gallies to replace them with fresh caravels when the time comes, instad of upgrading (more realistic anyway... and you can set the old ships on GoTo city and forget about them until they are back anyhow).

Another great element of making seas cost 2 and oceans 3 that has been overlooked so far.... the map size stays the same while the world gets bigger! More realistic maps...

Okay... it requires some imagination, but... ocean tiles can now be thought of as just being BIGGER instead of more difficult to traverse (like mountains or jungle), which means that all of those world maps where the oceans are reduced to make the continents fun to play on would be closer representations of Earth's land/water ratio.

This also makes random maps with lots of land more Earth like in terms of water amount.


Also... dropping the movement increase in the GL sounds good to me... Lighthouses make naval travel safer, not faster. The more unique wonders the better.

[Catt]

I was having trouble articulating something I hope grows out of the AU Mod until an exchange with alexman got me thinking about it . . .

One thing I hope to see out of the AU Mod is somehow influencing the AI civs to exploit any inherent advantages available to them either through their start position, map type, civ traits, combinations of the foregoing, etc. -- I would love to see one or two "Killer AI's**" show up in most games rather than seeing four or five "decent and challenging" AIs. In other words, in the effort to make the late game more engaging (read: not a tedious cake-walk to victory), one of the surer ways to do so seems to me to do what we can to stimulate very large AI empires in the Middle Ages / early Industrial Ages that can counter a large-ish human empire. Larger empires means that someone has to be made smaller -- hence the "Killer AI" concept. An AI that can aggressively exploit a relative advantage over its neighbors early and often will more likely be a "Killer AI" in the later game.

The big question is: can we do anything through the editor options to stimulate such behavior? Sure we could "handicap" certain civs and "strengthen" others - but that's a bit silly. Instead, I think we've taken the wiser choice of trying to improve the AI's performance with all civs in all instances. The question remains that remains for me, however, is whether there is something present in the AU Mod or something not present but available that we could do in order to stimulate the creation of Killer AI civs? Is there something we can do with other game parameters (in conjunction with the AU Mod or otherwise) that helps stimulate Killer AIs? Fortunately, alexman, who drives AU ever forward, knows as much or more about levers available to influence AI behavior than anyone I've seen post here or at CFC.

So, as we all play AU Mod games, I'd love to see feedback and hear anecdotal evidence on how often one or two dominant AIs (at least compared to other AIs) emerge compared with one's sense of how often they emerge under stock rules.

Catt


** The term "Killer AI" came from Theseus who hypothesized that 3 Billion year maps were more likely to create a Killer AI because of the more unbalanced start positions inherent in such a map - a hypothesis which I think may be true.

[Dominae]

With the tweaks done to the City Governors, we actually took a step back from creating Killer AIs; since all civs are now optimized by imitating human-like behaviour (Production at Build-Often, Wealth at Build-Never), the variety which could create Killer AIs is gone. This is not a bad thing, since I'm sure we would get bored of seeing that same old civs on top all the time (like Yang in SMAC).

In my games, Killer AI civs come about naturally. If they're not on my continent, they have time to wipe their neighbors out (which they invariably do, eventually). Thus, it usually becomes a battle of my continent against theirs. It appeas that the relative aggressiveness of each civ (high for Germany, low for India) is the prime factor in determining who comes out on top. If a civ is aggressive, it usually has Offensive Military at Build-Often, and is not afraid to use the units produced in consequence. This goes to show that Civ3 is, above all, a war game.

Short of playing with individual civs, I cannot see a method of achieving the Killer AI effect you're speaking of, Catt, at least not regularly. Geography is probably the most influential, non civ-specifc factor in creating Killer AIs. This is a good thing, IMO, because different geographies means variety in games. I personally wouldn't want a Killer AI civ to appear each and every game I play. What's wrong with having many "contenders" in the Modern age?


Dominae

[Catt]

Originally posted by Dominae
With the tweaks done to the City Governors, we actually took a step back from creating Killer AIs; since all civs are now optimized by imitating human-like behaviour (Production at Build-Often, Wealth at Build-Never), the variety which could create Killer AIs is gone. This is not a bad thing, since I'm sure we would get bored of seeing that same old civs on top all the time (like Yang in SMAC).

I never played SMAC, but I think I understand what Yang represents and (if I do) I agree wholeheartedly -- I don't want to give a specific civ a leg-up over other civs when played by the AI. But I do hope to stimulate explotation of relative advantage -- something which I'm not sure we can do with the editor, but which is worth trying.

As an example, I recently had cause to post a link to AU 102 which was a game in which the human could build no military units whatsoever. In linking to that thread, I re-read some of the game reports. In my own game, China became a legitimate superpower or "Killer AI." This was true in several other games as well. But China started next to Germany, and in several games Germany became the Killer AI and China was eliminated, literally or for all practical purposes. Germany has an aggression rating of 5; China, 2. Was it pure luck that determined whether Germany or China rose to the top? I don't know -- but I hope with a decent sample size to start to get a feeling for it (i.e., AU Mod versus standard rules).

In my games, Killer AI civs come about naturally. If they're not on my continent, they have time to wipe their neighbors out (which they invariably do, eventually). Thus, it usually becomes a battle of my continent against theirs. It appeas that the relative aggressiveness of each civ (high for Germany, low for India) is the prime factor in determining who comes out on top. If a civ is aggressive, it usually has Offensive Military at Build-Often, and is not afraid to use the units produced in consequence. This goes to show that Civ3 is, above all, a war game.

I'm not sure I agree with your conclusions (although my data may be different) -- but don't get me wrong -- this is the sort of feedback I like to see.

In my games it seems that an isolated "AI continent" is dominated by one civ as often as it is shared by 3 or 4 civs. Perhaps more importantly, I'm not convinced that the aggressiveness rating is as critical to success as you see it. If anything, my anecdotal observations are that hyper-aggressive civs tend to falter in the late game, even when successful at making gains in the early game, and furthermore that the aggressiveness and build-often-offense doesn't confer a significant warfare advantage in the early game. Hit or miss, Bismarck may be an Industrial Age monster or a Middle Age OCC.

Short of playing with individual civs, I cannot see a method of achieving the Killer AI effect you're speaking of, Catt, at least not regularly.

Yeah - I'm not sure we have the tools at our disposal. Soren's advice was to play with fewer civs than standard for map size and therefore give the AI's time to expand and develop infrastructure unthreatened by a human rush. But that still strikes me as leading to a group of challenging but relatively "equal" AIs.

Geography is probably the most influential, non civ-specifc factor in creating Killer AIs. This is a good thing, IMO, because different geographies means variety in games. I personally wouldn't want a Killer AI civ to appear each and every game I play. What's wrong with having many "contenders" in the Modern age?

Just to be clear -- I'd love to have several contenders (and definitely want variety in who the specific contenders will be), but would like to see 2 legitimate contenders rather than 4 "dominant AI's," none of which present much of a challenge to the human that has made it through the "Industrial Corridor" with a tech lead or tech parity. I want to worry about the game outcome while I'm researching Combustion instead of (1) worrying about my pathetic laptop's performance as I approach the Modern Age; (2) worrying about how to end the game quickly; or (3) not worrying because I lost before Combustion in any event (IOW, I'm not trying to eliminate losses -- the game would be very boring if I always won -- I'd just like the possibility of a loss to remain post-Steam Power for me, which it rarely seems to do - I'm either out of action by then or on my way to a win (with a few enjoyable exceptions, of course)).

Catt

[Dominae]

Ah good, I see what you're wishing for in Killer AIs: a civ that will put up a fight during and maybe after the Industrial age. In recent memory, Theseus had to handle a pretty scary Germany in 'Son of SVC' (he handled it nicely, and in typical Theseus fashion!).

I do stand by my observations that the aggressive civs often come out on top, but with an addition: aggressive civs that are also either Industrious or Scientific seem to come out on top. Persia, Germany and others want to kick some butt, and have the economy to support their desire; the Zulus have nothing past the Ancient era. I think this is the kind of information you're looking for, but unfortunately it's somewhat civ-specific. And we can't just make all civs Industrious and Scientific, now can we?

I'm with you one this one, Catt. I'll keep looking.


Dominae