v1.1 BETA 2

[Draginol]

We put up BETA 2 of v1.1 tonight.

Here's the list of changes:

http://www.galciv2.com/Journals.aspx?AID=110651

There's a couple changes in particular that I thought you guys might find worthwhile:

1) If you play as a good civilization, other races are even more likely to like you. So there's a more substantial benefit to good races.

2) AI trading techs with players and each other tweaked such that it won't tend to trade weapons anymore.

Here's why:

So last night, I'm playing the game and fighting a long war with the Arceans. The Arceans had been blackmailing other races and building up quite a treasury. I managed to build up a fleet of ships that have defenses set to fight against the Arceans when POOF, the Arceans upgrade their main ships of the line to a whole new set of weapons to counter mine.

Sure, that's a smart thing for the AI to do but it's incredibly NOT fun and totally unrealistic. Did the Arceans researchthose techs? No. They traded them with another player.

So in general, more work has gone into having the AI not tech trade weapons techs with each other or players. They will still do it but it should be much less likely.

You want players and AI to have to make their bed (pick their weapons and defenses) and have to really make a bunch of tough choices on what to research. If everyone just has every weapons tech, it makes that aspect of the game meaningless IMO.

3) We're still working on the escorting algorithms or more to the point, the AI's simulated intuition. The AI will not escort trnasports if it thinks the coast is clear. The problem is that (without cheating) it's hard to simulate a good guess. As we play the game more and listen to others, we get better at putting together better guesses.

4) We coded in a new API called EvasiveAction. Right now, only some AIs with transports use it. But essentially it'll try to run away from danger. We'll see how that works.

5) There's been a bunch of nice new tweaks over v1.1b1 with regards to the ship design screen.

But there's tons of usability stuff in this build too.

Still left to do:

a) Get those neutrality learning centers in!!
b) Get the mods UI in (they're almost done)
c) Play balance the new changes more

After that we'll call it 1.1 and then new stuff will go into 1.2.

 

[Sahkuhnder]

Sounds good.


Really looking forward to the neutrality learning centers. Neutrality is IMHO the most powerful of the alignments and can be fun to play.

Glad to see you guys are staying on top of the patches and upgrades.

 

[Nath]

Don't forget to eat/sleep/breathe.

 

[DaviddesJ]

Sure, that's a smart thing for the AI to do but it's incredibly NOT fun and totally unrealistic. Did the Arceans researchthose techs? No. They traded them with another player.

You want players and AI to have to make their bed (pick their weapons and defenses) and have to really make a bunch of tough choices on what to research. If everyone just has every weapons tech, it makes that aspect of the game meaningless IMO.

The problem is that trading all of their techs is realistic. If the rules of the game are that trading one tech for another gets each of the parties to the trade an advantage (a new tech) at no real cost to themselves, then rational actors will trade. It's irrational to refrain from trading. If you code the AIs to behave irrationally, then the human player will find ways to take advantage of their irrationality.

A different way to prevent everyone from acquiring all of the different technologies is to have a cost associated with trading. I.e., you can trade for technology, but you still have to pay something to incorporate that technology. Then players will still have to make a calculation about whether it's worth acquiring a particular tech, and it won't be rational to do everything.

Or you could have a limit on the ability to incorporate new technologies. I.e., you can trade for several technologies, but they are only added to your capabilities over time---the more you trade for, the longer they take to incorporate.

There are other variations too. I'm sure you could come up with some of your own.

The problem, in my view, stems from the unrealistic model that one race can instantly, with no cost, give another race the ability to do everything that they can do. Not from the AI decisions under that model. Just my opinion.

 

[Barbazoul]

It is not unrealistic to lower the weapon tech trading of the AIs ! On the contrary ! Why would a civilization research all the laser technologies, specialize in it, then give it away to everyone for a few pennies? This is what happens now, and THIS is unrealistic. AIs shouldnt give their weapon advantages so easily to each other, or even to the human player.

Limiting warfare tech trading can only give good results. The players and AIs will have to choose a weapon/defense tech line and stick to it. We will finally see more than one type of ship in the universe. I think it's a great idea to make the AI value its warfaring techs more : civilizations that are behind military will have to invest research points into military techs, it wont be possible to just pay a few credits to get all the weapons tech available anymore... More realistic, i would say.

 

[woodelf]

I think weapon trading shouldn't be done either. I can get Stinger I, II, II, IV ect from someone for a non-military tech. Makes no sense.

 

[DaviddesJ]

It is not unrealistic to lower the weapon tech trading of the AIs ! On the contrary ! Why would a civilization research all the laser technologies, specialize in it, then give it away to everyone for a few pennies?

You're asking the economic reason? Imagine a reverse auction. You and Fred both have Super Disruptor Tech. I want to buy it from you. I offer you a bazillion dollars, basically my whole empire for this one tech. At some point, sure, that's worth it to you, so you say yes. Then I go to Fred, and I say, I'm about to buy SDT from Barb for a bazillion dollars, what do you think of that? He thinks, wait a minute, if he buys it from Barb I get nothing, where I could sell it to him for a bit less and do very well. So he makes me a somewhat lower, but still high, offer. Then I go back to Barb and point out that he would do better to get something than nothing. This bargaining process converges to the unique economic equilibrium where I do get the tech for just pennies. This is the true "economic value" of the tech.

This doesn't happen in the real world because (1) it's not the case that two different people have exactly the same technology that does exactly the same thing, and (2) even if they did, they couldn't transfer it to a 3rd party for zero cost.

 

[Barbazoul]

I understand your point. But if the AI doesnt trade the weapons tech as easily as before (as promised by Draginol), the problem will be solved nonetheless : neither Fred, nor another AI, will agree to sell the warfare tech to you. Because the warfare techs can actually give you or the AI an enormous advantage, they shouldnt be so easily tradable. For the moment, at one point in the game everyone has the same weapons techs at the same time... no one benefits from it. If AIs stop trading their warfare tech, the game will become richer and more interesting ; and the weapons/defenses system will have a true role in galciv 2. I think it's a good decision. Weapons and warfaring techs should not be treated like other, less dangerous, techs.
In the real world, NO state would trade the A Bomb technology to another... But if one state discovers the cure for cancer, this will be traded for sure. (It wont be a fair trade, but it will be traded).

 

[Draginol]

In the real world, countries, other than allies, don't trade their latest weapons technologies.

It's not that it's impossible to get weapons tech, it's just much MUCH harder to get it. There's always espionage.

 

[Masquerouge]

Thanks a lot for not forgetting about the Neutrality Centers

 

[DaviddesJ]

In the real world, NO state would trade the A Bomb technology to another...

Nonsense. In the real world, we already have several examples of nations trading their nuclear weapons technology to other nations. The US passed on nuclear weapons technology to the UK. The USSR passed on nuclear weapons technology to India. China passed on nuclear weapons technology to Pakistan. These are all good examples of exactly what Brad calls "unrealistic": nation A sharing weapons technology to strengthen nation B against nation C. We also have good real-world examples of nations exporting nuclear weapons technology for economic, as well as political reasons. These support the proposition that if you insist that people behave in ways counter to their economic interests, on "principle", you'll often be disappointed or surprised by the results.

 

[DaviddesJ]

I understand your point. But if the AI doesnt trade the weapons tech as easily as before (as promised by Draginol), the problem will be solved nonetheless : neither Fred, nor another AI, will agree to sell the warfare tech to you. Because the warfare techs can actually give you or the AI an enormous advantage, they shouldnt be so easily tradable. For the moment, at one point in the game everyone has the same weapons techs at the same time... no one benefits from it. If AIs stop trading their warfare tech, the game will become richer and more interesting ; and the weapons/defenses system will have a true role in galciv 2. I think it's a good decision. Weapons and warfaring techs should not be treated like other, less dangerous, techs.

It's possible, but I'll believe it when I see it. There are lots of ways it might not work out as you say. For example, if it's still possible to acquire military techs by invasion or espionage, the result of this sort of change is likely to be a large bias towards those tactics.

 

[Nath]

Constructivity first!

Thank you for making things a bit more diverse. The homogenization was really getting on my nerves.

A "we'll trade you this tech, but you can't trade it to anyone else" option to diplomacy would be verrah nice. Far too often I keep techs to myself purely because I don't want certain civs getting them.
Only being able to trade techs that you've personally researched would be nice, also. But I assume you've decided not to do that.


Non-Constructivity second!

Well. This reality tangent is fun and all, but I'm ignoring the fluff and wondering why in the world triple D seems to be in favor of everyone in the game having the same weapons.


Oh look, another post. I type too slow. Or I get too distracted and leave a half-done reply open for half an hour.
So in place of this possibly imperfect solution, you suggest......... hm. I think I missed your suggestion.

 

[malekithe]

Subjective valuation of techs for trading purposes is a difficult line to walk, it would seem. On the one hand, you don't want a heretofore pacifist to be able to trade himself into military parity in the space of a few turns. On the other, you don't want the AI to overvalue weapons techs to the point where they're willing to offer their entire kingdom for a high tech missile.

It's definitely not realistic or, in my opinion, good for the balance of the game if the AI simply refuses to trade weapons technologies, especially if those technologies can still be acquired through other more nefarious means. There has to be some price at which the AI is willing to part with its weapons technologies. Otherwise, there is too much emphasis placed on the research of these technologies. This, in turn, overemphasizes research centers and research spending for the player or AI engaged in warfare. If a single dominating power has an abundance of research centers and has used those to create a large advantage in military technology, the other players' only choice is to combine their efforts into collectively improving their weapons technologies by trading amongst one-another. Any other course leads to at least one of those other players being wiped from existence before the larger power loses momentum. Restricting trade of weapons technologies is tantamount to eliminating the need for maintenence of high diplomatic relations.

What might be a better system (but still likely exploitable in some fashion) is a differing subjective valuation of techs depending upon a number of factors. I'm sure this is already implemented to some degree in the current diplomacy algorithms, but it may benefit from expansion. The factors I would consider are: base technology cost in RP, type of technology (military, research, diplomacy, etc...), diplomatic relations, relative galactic power, and general state of the empires (this one is what creates some randomness in trading). The AI would need to perform this valuation for both the technology it would be recieving (from the context of it's own empire) and the technology it would be giving away (from the context of the opposing empire, or what they know about the opposition).

The values would interact as follows. The valuation would start with the base cost in RP. Then, two modifiers would be determined. The first is the diplomacy modifier. The diplomacy modifier would be some combination of diplomacy level and technology type. Weapons techs might generally have larger diplomacy modifiers, while diplomacy or domestic techs may be much smaller. On the average, the lower the diplomacy level, the higher this modifier would be. The second additional number is the empire state modifier. This looks to compare the technology type versus how the empire stacks up in a number of areas (or how the AI thinks they stack up). The net result is weapons techs mean more to an empire with a high industrial base but little weapons technology. On the other hand, military techs mean next to nothing to an empire with a high technology output, but a poor industrial base (they could research the tech easily themselves and there's not a lot they can do with it anyway).

Once you have the base value and the two modifiers, they are combined in some way (multiplied, most-likely). The AI would then compare the value of the offered technology with the received technology. There would be some difference between these values. To determine whether the trade is a go or no-go, the AI would compare the difference in valuations with the difference in power between the two empires. The further ahead an AI is, the more it may be willing to "take a loss" on the trade. This would have the added bonus of making it unlikely that a backwater empire would trade away an expensive new technology you just gave them. The other AIs would have to offer the backwater empire a sweetheart deal to get the expensive technology from them, as the backwater empire would have no tolerance for "losing" trades due to its position in the game.

The whole idea behind this is that, if two empires are on equal footing and have modest diplomatic relations, neither should really shy-away from a mutually beneficial trade. They may even engage in a trade if they see the other side benefitting a little more. The reason you have to allow this is because the diplomacy modifier will sometime "skew" the results of their evaluations into making an AI think they are getting a worse deal than they are. For instance, it may be possible for both AI trading partners to decide the other would benefit more from a particular trade. The reality is, it's a relatively fair trade and, unless one AI is significantly lagging the other in power, the AIs should probably choose to go through with it. As such, there must be some tolerance for perceived "losing" deals to enable AIs trading with one-another.

The net affect is that it easier to trade: for domestic technologies, with people you're weaker than, with your good friends, for techs you won't benefit from, techs your partner desperately needs.

It's harder to trade: for miliary techs, with people you're stronger than, with your enemies, for techs that benefit you greatly, techs your partner has little use for.

Additionally, because each trade improves your power in the computer' eyes, it becomes progressively harder to broker the same tech to multiple empires. To further aid this, it may be beneficial to add another component to the mix that will decrease a tech's trade value as more poeple discover it. That would be relatively simple.

Hmm... Too long a post, but I figure I'll ramble on a bit about my impressions of the game. I find that galciv2 is a wonderful thought exercise. I'm having trouble enjoying the game itself, in all honesty, but I love crunching the numbers to figure out all the little things. For instance, if a planet will take 8 turns to reach by colony ship, how many people should I pack on that ship for maximum benefit? I've got spreadsheets galore devoted to various aspects of the game like this. I rarely come to any worthwhile conclusions, though, that are practicable in a real-game scenario. I don't know if that's a good thing or a bad thing...

 

[DaviddesJ]

So in place of this possibly imperfect solution, you suggest......... hm. I think I missed your suggestion.

Message #4 in this thread has some suggestions.

If you're seriously interested (not just whining), I can give you others. E.g., you could have a model where having a higher level of beam weapon technology actually reduces your capability to make missile weapons. That is logical because your shipyards and manufacturing capability will naturally be specialized to your "primary" weapons technologies, and they won't be as good at making "secondary" weapons technologies as would a different race that specializes in those weapons.

All involve breaking the model that you can give away a tech and it costs you nothing and gives the recipient the same capability that you paid to research, for free. Because that model fundamentally leads to "everyone has the same techs". This isn't so much of a problem in Civ4, because everyone is supposed to eventually get the same techs. It's a bigger problem in a game where different races are supposed to proceed down different tech paths.

 

[Nath]

Wow. I think Malekithe just said "All these external 'solutions' are silly band-aid fixes and it's really just a matter of tweaking the value algorithms."
When you put it that way, I'm inclined to agree >.>

DaviddesJ: Well damn. I really DID miss your suggestion Err... forgot about it as only I could. Close enough.
My apologies

I like the idea of specializing in one weapon/defense type causing the others to be more expensive. Makes sense if you stretch it a bit, would be pretty easy to codify and it's not really a silly band-aid fix implemented to compensate for whacky underlying algorithms
The only problem would be "Damnit! My espionage/invasion/whatever gave me mass drivers and now my Missiles of Doom cost more "