Brad Wardell's views about technology, politics, religion, world affairs, and all sorts of politically incorrect topics.
The galaxy is on fire
Published on April 23, 2006 By Draginol In GalCiv Journals

Prologue

What is a game of Galactic Civilizations II v1.1 like? What has really changed for the regular player over what was in the original retail box? And why did we make it?

v1.1 of Galactic Civilizations II isn't a "patch". It's not designed to "fix" things. In the course of making it, was certainly "fixed" bugs that we found and tweaked things that players didn't like, but the point of it was to listen to what players liked and didn't like and add those features to the game. 

Key to the success of this strategy is that the retail boxed version of the game has won Editor's Choice Awards already from Gamespot, GameSpy, UGO, IGN, Computer Games Magazine, Firing Squad and many other publications. G4's "XPlay" just gave it a 5 out of 5 (something they rarely do).  So players don't need to worry that the retail version of the game "wasn't finished". The point of 1.1 and subsequent updates is to keep the game new, fresh, and make it better.

The things that players will likely notice the most in v1.1 include (as a partial list):

  • A bunch of new game setup options such as random intelligences, random opponent selections, random galaxy selection. In essence, the game can now be set up so that the player really has no idea what the galaxy has in store for them. Previously players had total control (and still can have total control if they want) of the startup conditions but that could take away some of the mystery.
  • The ship designer, which has emerged as an incredibly popular feature, has been improved a great deal.
  • The computer opponents are far more intelligent than what initially came with the game. They're not "harder" as much as they're simply better. More clever. More intelligent.  The computer AI was considered one of the strengths of the retail box. I'm pleased to say that what's in 1.1 is at least twice as good as that. Having a month of playing the released game and listening to other people's strategies makes a big difference.
  • The economics of the game are much transparent. Player abilities, bonuses, and other things are just more balanced, more streamlined, and make more sense. It's never going to be a "simple" system. But it's more transparent now on how things work.
  • The User interface has been tweaked and touched up to make it easier to work with and manage units.
  • The costs of buildings, ships, components, etc. have been tweaked.

That's just a tiny part of what will be in 1.1.  The number of changes, big and small, amount to an overall improvement in the game.

And, 1.1 is a free download for all players. Soon we'll start on 1.2..

Note: I'm playing this at 1920x1200 (wide screen).

For this game-play example, I'm going to tell it as a story..

Introduction

The initial colony rush of 2178 (Earth time) has slowed down. But the recently discovered Epsilon quadrant has brought the major powers together again in the year 2225.  This takes place just prior to the "Dread Lord" incident...

For this game, I'm setting up a custom map called Drama (which I'll upload to the Library). I've turned on blind exploration and disabled minor races.

I'm playing as the Terran Alliance:

For my political party, I've chosen the Technologists as I want that 20% research bonus.

I'm going to play against them all but it's going to randomly pick the races and I'm going to randomize their intelligence. What this means is that I pick "Tough" as the difficulty and the game will tweak their intelligence up and down a bit to add some variety. These are new features in 1.1.

...The Gathering Begins...

The Terran Alliance has a colony called Promethion.  We have been assigned to build up this colony and take control of this quadrant (or at least make sure our friends along with us control the quadrant).  All the powers have similar missions no doubt.

I start out by quick building research centers to get a good start on tech.

I then set my spending to 100% and put it all on research:

Your factories and labs produce base on what % of your overall capacity you want to use and then what % of that capacity is going to go into military, social, or research. In this way, I have to make tough choices on what I want to do.  Right now, I'm putting it all in research.

My first colony ship, the TAS Lucky Star will make sure we claim some extra worlds. It has a base movement of 3 parsecs per turn. So it's faster than the standard colony ship but has less range.

For this game, I'm going to focus purely on propulsion techs. Go right to Warp Drive so that my ships are very very fast.

Unfortunately, I make my first mistake pretty early. I colony rushed. A lot of people think that the early game of GalCiv II is purely about cranking out colony ships. GalCiv 1 was that way. But in GalCiv II, that's not the case and 1.1 emphasizes that. My population doesn't increase very high and you are rewarded for having higher populations in the form of tax income.  So I've fallen a bit behind.

The Iconians have scouted far.  I need to get my act together soon.  The galaxy is already getting established.  So note to others, don't go right for warp drive.

The first thing I'm going to do is get some military power here. Not much, just enough to take the "conquer me!" sign off my back. The Short Lance is nothing impressive, but it should do the job.

Meanwhile, I still only know about the my tiny corner of the quadrant. The Altarians have claimed some of the area around my borders but other than that, the quadrant (the map) is a mystery.

Since I'm relatively far behind, I decide to get Xeno Ethics. Why? Because then I can choose my alignment:

By choosing good, the other civilizations, particularly the powerful Altarians, are much more likely to have good relations with me which I'm going to need.

I'm going to go for a political victory this time around. So what I'm going to do is try to stay out of war and build up my infrastructure and rely on the powerul Altarians to protect me. Because by the end of 2227, the Drath are already gone (wiped out by the Drengin) which means there's bad things out there.

And the Torians are now out too. They're dropping like flies. The Drengin are a real menace.  And in 1.0, the Drengin would be free to waste everyone.

But this isn't 1.0..

The Arceans, who are powerful, have decided to stand up against the Drengin Empire.  My job then will be to help bankroll these guys.

I am in an interesting position. The Drengin Empire do not know about me yet.  If I venture out into the galaxy, they will become aware of me.  So I am going to stick in my corner for now and build up.

Speaking of another reason not to be evil -- even though it's easy to choose evil because of all the little benefits, nobody is going to help you out.

The titans are now at each other's throats. The Drengin, who in 1.0 would have just wiped everyone out, are now in a position where people are ganging up on them to avoid being wiped out in turn.  Perhaps the Drengin should have researched more diplomacy techs. But if they had, they wouldn't be Drengin would they?

I set rally points by pressing the button on the lower left of my screen. Then I can send ships to that rally point. I'm going to build up an impressive starbase network to make me a powerhouse in the long run.

Two words: Screw you.

Ut-oh.  The Korx are mad at me. Time to look at updating my ships.

To help combat predators, I need to build up my defenses. One way to do that is in the form of military starbases that assist my ships. 

In time the Korx not only make peace with me, but the the conquering Drengin force them to give me their worlds.

The problem is, now the Drengin are aware..

One feature of GalCiv II that doesn't get discussed but I think is a really important feature is that the game remembers your ship designs between games. That means, the longer you play the game, the more ships you have to choose from.  

The F10 Hornet was a radical departure for the Terran Alliance and designed specifically for the Epsilon quadrant. The "stinger" in front is the designed to provide superior guidance for its Singularity canons and help ensure that it gets the first killer shot in a battle.
Name: F10 Hornet
Length:
32.5 meters
Span: 45.5 meters
Mass: 72 metric tons
Weapons: 3 Westinghouse G1 Singularity Cannons.

The Drengin have pretty much everyone at war with them, and  yet are holding their own. My military isn't even worth speaking of at this point. But I suspect I'm going to have to do something about that soon.

The Drengin just want to conquer everyone apparently. They're in for a surprise...

The Avalon class attack cruiser is probably one of the most powerful  ships of its class in the galaxy -- not just this quadrant.  With an attack rating of 12 (though no defenses), it packs a mighty punch. Its advanced sensor array also would allow it to spot enemies from a great distance.
Name: Avalon class Attack Cruiser
Length: 1
35 meters
Span: 140 meters
Mass: 605 metric tons
Weapons: 6 Westinghouse G5 Singularity Cannons.

As proud as I am of the Avalon. The Drengin have something monstrous on their ships...Psyonic Shredders.

A Drengin fleet of 5 ships has over 200 attack power.  All on mass-drivers.  So it was time to research armor to counter it.  Defenses on a ship use less space but cost more. I have lots of money though.

The text says it all.

A tweak my Avalons to make use of my new armor technologies and to use the Nano-Ripper technology my scientists have discovered.

The Drengin Dreadnoughts however are now out.

My ships each die with 1 hit. ONE hit!

A mop up fleet comes into the sector where the Korx once lived and I send in my flag fleet. 9 of my Avalon class ships. The fruits of many months  of labor.

 

That's right. It's MY turn. My flag fleet takes out several of his fleets. With allied help, we pave the way to the Drengin core worlds.

The Drengin capital world ("Torment") in this quadrant is now open for invasion.

The Drengin do have some very good ships. The mark 6 Drengin battle ship is bristling with weapons.

The most exciting battle so far has been this one where my fleet took on 2 Dreadnoughts.

Only 3 of my ships survived. By the backbone of the Drengin navy was snapped.

But the galaxy was about to get a lot more interesting...

The Galaxy is on fire...

The Altarian Republic and the Iconian Refuge are the two powerhouses of the galaxy. Their alliance, and I mean a real treaty alliance, helped make the galaxy safe from the Drengin who would have otherwise conquered all. 

The Drengin are defeated but not destroyed and the Iconians and Altarians are now at each other's throats.

It was now time for me to bring out my own Dreadnought.

The Sentinel Class dreadnought is a monster. Well, at least by Terran standards. It is huge.  Armed with nearly a dozen guns and covered in armor, it is meant to survive the kinds of battles that would destroy entire navies. It is a one-ship navy with a cost of 1.3 TRILLION credits.
Name: Sentinel capital ship
Length: 950
meters
Span: 220 meters
Mass: 20.2 million metric tons
Weapons: 38 attack (Durion Corp. Mark II Graviton drivers) and 10 defense (Fivel LTD. Kanvium Armor)

Just to give you an idea of scale, here are some ships (courtesy of Starship dimensions) to compare the Sentinel class dreadnought to.

With my monitor class scouts, I keep an eye on the Iconian/Altarian war.

The Arceans launch a sneak attack on me!

It's insane. The instability caused by the Iconian/Altarian war has made everyone reasses their relative positions.

With some negotiation, I am able to call the Arceans off so that we can finish off the Drengin. But I won't forget their sneak attack.

Meanwhile, the fat lady is warming up to sing about the end of the Drengin Empire in this quadrant.

That's right. All mine. Goodbye Kona.  And as the Drengin fade, the Iconian Altarian war comes to an end as well. Now it is time to wrap up my strategy and win a diplomatic victory.

I am now in an alliance with both the Altarians and the Iconians.  Time to wipe out the Yor.  No one will miss them.

Ah, the Yor's rnadom intelligence set them too dumb to do anything about this.  Heh.

Because...you know, they were right.

Unfortunately, by a cosmic coincidence, my favorite weapon type is the same the Drengin were focusing on. So everyone else has really good defenses against mine. So I lose the first battle.

Dreadnought vs. Dreadnought. I outnumber them. But they have a lot of fleets and I'm at the heart of their territory.

A Sentinel falls.

And the Arceans are now back in the war. 

Everyone is at war now!

End Game

But the Yor are no match for me now. Soon everyone is after the Yor.

With the Arceans, they recognize my military might.

The Drengin did well but they got in over their heads.

And so the quadrant's destiny is settled by political treaty saving billions in lives!

Lessons learned

  1. The alien opponents don't build enough transports. (tweaking)

  2. The aliens need to decide what to build more based on how much their colony is producing. Dreadnoughts are great but if you can't build them very fast, better to stick to an easier ship (they already do this somewhat but apparently need to do it more).

  3. Aggressive civs need to be more aware of the potential consequences of aggressive behavior.


Comments (Page 3)
4 Pages1 2 3 4 
on Apr 24, 2006
Brad should help the A.I by purging the ability to make sensor ships, which are a human exploit.I say only one sensor per ship allowed.


This wouldn't necessarilly fix the problem though, it would just mean more tedium in having your sensor net be 10 ships instead of 1.

I don't mind so much how sensors are implemented right now, I'm just not convinced that the AI understands their importance, or uses them correctly. It would especially seem as though the AI doesn't credit other players enough when they have sensor techs (or sensor ships). I mean once someone builds the Eyes the AI should assume that they basically see everything.
on Apr 24, 2006

Building sensor ships is a viable strategy. I build AWACS ships.  The AI simply needs to do it too. But it'll take time and I think there are other incremental ways that are easier to keep making it better.

But you can see one of the reasons I get irritated when someone screams "The AI cheats". You can see how much easier it would be if the AI could simply see all your ships.  Just a big FOR loop through the ships to see if it's safe.  5 lines of code.

on Apr 24, 2006
That is exactly how the AI is supposed to work -- it's supposed to wait until it gets space superiority before sending in transports. The problem is the FOW

Are the build transport waiting in orbit or in open space? Because there is a big difference if the world have reached their maximum allowed population. Being able to fill regularly tarnsport while keeping planet at max pop can be very important when trying to invade a planet.

And how does exactly battle when many fleets are sitting on the same square? Which one is targeted first? If it is a defensive fleet, then it would be nice thing for the AI to stack transport fleet and attack fleet on the same square when trying to invade a world

on Apr 24, 2006
I agree with that that the AI really should value sensors more and especially the eyes of the univers wonder which is in the hand of a human player seriously overpowered (imho). Sensor techs are very cheap and give you an awesome advantage against the AI. Imho the AI value of 1 they currently have is ridiculous (sp?). I mean even ultra life support gets an AI value of 5, and I can't think of something which is more useless then that... the only other techs with an AI value of 1 are those that lead to the tech victory. Imho that's a major disadvantage for the AI, but I can also be completely wrong. Correct me if that's the case.
on Apr 24, 2006
And when exactly is 1.1 going to be out??? Because I hate waiting.....
on Apr 25, 2006
Aggressive civs need to be more aware of the potential consequences of aggressive behavior.


i've noticed this lately (great post btw). i tend not to go for the jugular, but play games until tech levels are high all around. in the game before last, the drengin managed to wipe everyone else out, but they got a significant upper hand in terms of planets. i've noticed they often hold back producing colony ships in favor of research more than other races.

anyway, last game the drengin and the korx both keeled over fairly early under the altarians and thalans. then the thalans found an artifcat that turned all the zeros to tens. i'd seen this event once before, but i was all ready 5 turns to an influence victory. this time, i all ready had a major lead in every way. i need to start playing at a higher difficulty leevel. i managed to snatch up most of the free planets, mostly in thalan space, but a fair number in altaria and a couple in iconian territory. ones colonized by other races started rebelling very quickly. i had rushed custom colony ships that could land 1 billion but move like 41 parsecs in a turn. gotta love yor miniaturization (i paid to be good). i rushed a fleet of sctructors and gave them all influence bubbles, which held them over. but something caused a war, and the altarians were allied with the torians, iconians, and drath (who were pretty weak). well, now everyone's at war with at least someone else, except me. the game's started crashing every few turns. le'sigh. maybe i should have thought about more ram.
on Apr 25, 2006
In my latest game (beta4 .18) I've noticed that the Ai has been building lots of transports, grouping them with escorts, and parking them outside of potential enemys' territory.

I think evil civilizations are still viable as long as you don't go for the mass invasion strategy. I use my free starbase upgrades to pump out influence starbases all over my borders. I wait for the good guys to be distracted (or pay them off) and then take out one of my border civs who's fallen mostly under my sway as well.

Are the social production bonuses from the various sources going to be fixed for the full beta?


on Apr 25, 2006
Building sensor ships is a viable strategy. I build AWACS ships. The AI simply needs to do it too. But it'll take time and I think there are other incremental ways that are easier to keep making it better.


Sounds good to me. I've noticed the AIs parking (what I assume are obselete) scouts in various places, but once you pick them off they don't replace them. I wonder if it would make any sense to even put the sensors right on the transport, even if it slows it down a little. That way it might be easier to keep the transports with covering fleets.

It also seems that invasions are a bit disjointed, in my aforementioned game I had one planet (from a minor) quite far from my base systems, but close to the Drengin. When they declared on me their mega fleets (5 fleets with 100+ missle rating) started moving (speed6) through that area, but since I moved out my two defenders to pick off transports, their fleets seemed to just ignore that system and continue on the long slog towards my base systems. Had they 'parked' around that system and waited for their transports (though they weren't covered anyway...) they could have taken it, or forced me to try and send a relief force sooner than I would have been ready to. Anyway, its not like they were passing up some <10PQ planet it was up to 18 by this point. The Yor were doing the same thing constantly with their fleets as well, rather than parking in the sector with the system and waiting for their transports to arrive they kept on going, ultimately yo-yoing as I was able to move faster fleets into the border area.

I'm not knocking the AIs tactics so much actually, they do a fairly decent job, but they still do seem to become too fixated on some fairly remote target, and passby more obvious (or at least closer) targets which they could probably take out with reletive ease. Are they for some reason always looking for a 'knock out blow'? I think they would do better in wars of attrition, especially when they hold the edge right off the bat. If they don't do any significant early damage the human player (who usually seems to have better production, or at least better focus for their production) will eventually come out on top. If however, the AI takes down the border worlds first, it puts an extra damper on the humans production, and if the AI were feeling cruel, just have them invade with mass drivers and leave the worlds semi-torched

But you can see one of the reasons I get irritated when someone screams "The AI cheats". You can see how much easier it would be if the AI could simply see all your ships. Just a big FOR loop through the ships to see if it's safe. 5 lines of code.


Heh, well I'm not one of those screamers, I appreciate what you are trying to do, and am hopefully doing my small part to keep you going with it

Though here's a wrinkle... add an option to turn off FOW for AI or everyone. Granted that kinda makes sensor techs meaningless, but it seems the AI already doesn't think much of them
on Apr 25, 2006
Attack value (weapons and logistics), speed, and sensors are the key to victory! Speed and sensors allow you to fight on your own terms (out flanking slower superior fleets) and get in that important first shot! A survey ship can score a lot of cash exploring anomalies required keep your spending at 100%. The Eyes of the Universe is the most valuable super project! All throughout the long beta period of v1.1 I have been asking if it now values sensors in addition to it's new focus on speed. Well I have my answer now! NO!
on Apr 25, 2006
As I have mentioned elsewhere before, If you up the AI's priority on sensor tech (to about 15), and then make all sensor tech give +2 sensor range, the game is significantly more enjoyable. An AI that is able to detect your ships is a much more worthwhile opponent, because they don't make so many mistakes.

FYI: In that same mod, I also changed Eyes of the Universe into an espionage bonus (it's too powerful before, and useless if everyone gets +8 anyways), and combined all the starbase sensor modules into one that adds +5. (I also took out galactic survey wonder, which is just useless all around.)

If you're like me and don't care about the metaverse, I suggest you try those modifications out. Maybe this won't be the "official" fix to the AI's lack of sensors, but it's a lot better than leaving them blind.


Personally I think that this method (or perhaps something a little less, like 6 sensor tech that +1 each) and a new rule that allows ships in orbit to retain their sensor range is a lot easier to implement than programming the AI to use AWACs ships properly. I mean seriously, setting up a sensor net might seem like a chore for us, but for an AI, it's a lot more work (Things like where to prioritize the sensor net, how to fill the gaps and maintain them when they are taken out, etc... things that we do without a second thought and take for granted).
on Apr 25, 2006

One of the things I would like to do in 1.2 is have an option:

[] Allow AI to use extra CPU power for more advanced strategies.

If that is checked, then I can do more advanced algorithms.  A lot of the suggestions we get sound easy to do but they would bring the game to a crawl to do it.

Consider the sensor-suggestions.  Why aren't we putting more emphasis on that?

Because the more sensors the AI has, the slower the game will get.  The AI can't see.  It can simply look at tiles on the map one at a time and query what's on that tile.

Imagine on a gigantic galaxy if the AI was looking at every single tile over and over how slow that would get?   The AI has to do a lot of simulated intuition instead. I.e. "guess" whether there's threats in a given area because looking at each and every tile in a path or near a destination would become very problematic.

on Apr 26, 2006
Imagine on a gigantic galaxy if the AI was looking at every single tile over and over how slow that would get? The AI has to do a lot of simulated intuition instead. I.e. "guess" whether there's threats in a given area because looking at each and every tile in a path or near a destination would become very problematic.


I hear that and I basically agree.

However, ... (always a however ehh? )

It still seems that even if the AI is going to be 'blind' it should do a better job of keeping its transports protected, and not by escorts alone. I finally finished up the game I was talking about, and the AI was still sending lone transports out when I *know* it could see my fleets (since they were right next to his systems), and since I had the eyes it should have known that I could see everything around that area. At that point it really didn't matter what the AI did with its transports, but it shouldn't matter. The Drengin even had a few decently appointed dreadnaughts I had to slog through, so it could have protected the transports with them, or just run them to a 'safe' place, heck, keeping them on planets is probably smarter at that point, unless it was just suiciding population
on Apr 26, 2006
Kalin,

Could you upload your mod to the library? Sounds very interesting. I could edit the XML files myself, but I'm lazy.

Regarding the sensor thing...sounds like Frogboy has two contradictory priorities:

1) The AI should not 'cheat' and therefore should not be able to see into the FOW.

2) The game should have good performance, so the AI should essentially ignore sensor research.

So we end up with a non-cheating but tactically disadvantaged AI.

Combine this with the Eyes of the Universe giving the player an unbelievable advantage sensor-wise...and you have a less than challenging scenario.

Cheers

h
on Apr 27, 2006
[qoute]Name: Sentinel capital ship
Length: 950 meters
Span: 220 meters
Mass: 20.2 million metric tons
Weapons: 38 attack (Durion Corp. Mark II Graviton drivers) and 10 defense (Fivel LTD. Kanvium Armor)

I wish there was a way the game would generate those 'flavor' summaries for our ships, I'm not a programmer and there might be easier ways or maybe it's not even possible with how it works.. But if upon confirming a design if it'd make a quick 'volume' check to determine mass, then determine the length and breadth of the ship to determine length and span, that ought to be a nice universal way to determine the flavor stats.

It'd just make it easier to figure out how how our ships are becoming more and more powerful. By time the dreadnaughts come out, most of the small and medium ships have probably yielded to electric death, so you don't get much opportunity to compare them.
on May 04, 2006
Nice read but is there any reason why I dont see an end game video when I win?
4 Pages1 2 3 4