10 sectors away from nearest enemy sectors ;)(so no military at the start), 1 minor civ traded with me for money - so good starting economy, paid for wars between comps (it's game with 1.03.057)

I thought that's only way to survive. Maybe i was right, but i'll try closer respawn at this time. Most important thing is to stay far away from most powerful civs...

Good strategy gamer must go to fight only then he already won
or something like that (San Tzu's Art of War)

~SDC~

Have you read this thread, Brad?

http://www.galciv.com/forum.asp?BID=GS&id=50941

I have refined my strategy by taking the unpopular good alignment route. I don't think it is possible to beat masochist without buttering up with at least one of the AIs, and the good alignment helps tremendously in that department. Furthermore, my experience is that in a setting were AI planetary production is supercharged, the Torian/Altarian races generally perform better than the militaristic Yor/Drengin since they quickly multiply their production by developing their infrastructure and aggressively acquiring starbases. This makes it extremely important to maintain high diplomatic standings with the good races, and especially Torians since they rush research for capital ships unlike the Altarians who branch out to pick up Terra Computers and what not.

The Torians in my current game are churing out an absolutely obcene number of starships even though they only control 21 planets. The game time is September 2188, and I can count the following units through the diplomacy screen:

Battle Axes: 11
Battleships: 7
Combat Transports: 4
Corvettes: 10
Defenders: 6
Dreadnoughts: 49!!!
Freighter: 2
Phoenix: 14
Ranger: 21!!!
Starbases: 5 (2 economic, 2 military, 1 research)
Transports: 5
Sensor Drones: 1 ("Fortune" class asteroid find"

They have just completed Avatar research as well, and I will no doubt be seeing those in action soon. The game would be even more lopsided if I didn't donate stacks of AMM missiles next to 4 Torian starbses (2 military & 2 research) to the Drengin and Yor. The only fatal flaw in the Torian strategy is the lack of transports for ground invasion I believe. The sheer number of ships they are capable of producing is just jaw dropping however. I have my entire empire set on AMM production and I still feel there's a tough fight ahead of me.

In comparison, my current military is composed of:

Anti-Matter Missiles: 21
Battle Axes: 1
Battle Hammers: 2
Combat Transports: 5
Constructors: 34
Corvettes: 17
Dreadnoughts: 3
Freighters: 2
Rangers: 1
Sensor Drones: 91
Starbases: 27 (25 resources)

I have been able to avoid conflict by establishing trade routes with the Drengin early. War was unavoidable with the Yor due to their "military rush" doctrine, but the Torians quickly fixed that by pounding them back to the stone age. I was forced to make two Battle Hammers to destroy the first wave of transports from the Yor though (they share the same weaknes with Altarians: not rushing for capital ships from Dreadnought and up). Arceans and Altarians started in corners so they haven't had of an impact on the game yet.

p.s. Can we get the official word on infinite range AI colony ships?

I should add this to be clear: not possible to beat the AI without good diplomaitc maneuvering, unless you use cheese tactics. The worst thing I did was gifting AMMs to the Drengin and Yor next to Torian starbases and loading the game once to counter the infinite range AI colony ships.

To be more precise: 8,9,9,12,12 sectors away.

To Za H
25 resources???
Well i played too low number of games for now.

~SDC~

You must be playing on tight clusters with lots of Ctrl+N

Well i didn't saw more resources on tight clusters anyway. Ctrl-N... That Ctrl-N? I posted my 3nd and 5th game by the way (1st played to dreadnoughts on medium, 2nd - restart then i saw that i'm losing badly in colonization speed, 4th - i saw that i'm losing with 1.03 patch (enemy dreadnoughts, i like builder strategy too much - my common mistake)) so for now i feel free to use cheap tactics etc.

You was beta-tester, i think?

~SDC~

Daveman thinks back to one game of Civ (the original) where the French captured a pitiful new city of mine... next turn I get a message that the French have built the Pyramids there.

LOL... yeah, GalCiv cheats just like that.

Just to clarify:

1) Yea, the AI cheats big time at incredible. It cheats in the sense that it gets huge bonuses to its economy. On the other hand, the AI doesn't get to restart its game whenever it finds that it didn't start out in its own isolated corner of the galaxy.

2) The lone AI colony ship out of range thing is a bug. We've seen it, only have theories what causes it. But I think most people who have played know that the AI plays by the same range rules the player does. On those large galaxies, youd on't see the AI quickly colonizing those isolated planets in the middle of nowhere for instance.

~SDC~

As Za H says, the key to winning against Incredible opponents is through Diplomacy.

Starting off with +70% Diplomacy, Populists, and rush-buy the Diplomatic translators.

Then the key is how you USE your Diplomacy. They only way to keep up in the tech race is through trading. With the AI's huge bonuses to everything, in a short time you will find that they have all these technologies and you're in the stone age.

At Incredible level, the AI won't buy anything for cash (except the Alexians--is that intentional or a bug?), so your only source of money is from the Alexians, and what you can earn on your own.

In addition to trading technologies, you also need to get the AI to fight each other. Only if they are fighting each other will they slow down enough to give you a chance to stay even.

And it's about staying even, there's no way you can ever overtake them (until the very end of the game).

But I did manage to beat them, and the game involved a lot of fighting (it was painful having to give away all my wonderful trade goods to the Altarians to get Dreadnaughts from them, but it was the only way I could stay in the game). It wasn't one of those pacifict games where you never fire a shot.

But when the end finally came, it was through cultural conquest, because at the end of the game, enough Constructors to build a Party Palace comes a lot faster than enough population to invade the enemy planets.

The Leader: Sounds to me that you're just a very good player then.

It's the age old dilema of AI design - ultimatley the computer AI is only going to be as good as its designers. In this case, me. There will be a certain percentage of people who are simply outstanding players. And as long as the AI isn't flagrantly cheating, that means some top gamers will be able to find a way to win.

All we AI writers can do is learn the tactics via the Metaverse or from talking to people. Because the GalCiv AI, even when it's "cheating" at the highest levels is still playing by the same rules as the human (just gets more money basically) then we still have to figure out a good strategy to make it better.

That said, the single biggest advantage the human player has, based on what I've seen in the metaverse submissions, is the Ctrl-N feature. Players can just keep hitting that until they get a really good start. And let's face it, with a good enough of a start condition, even if it were multiplayer, a great player will have a hard time winning out if the starting conditions are stacked against it enough.

and in GalCiv, the starting conditions are intentionally high in variety so that players get more replayability out of it. I.e. you're supposed to use Ctrl-N.



~SDC~

The AI cheats are well documented. They aren't really cheats in that sense.


These were some of the cheats in the civ series.

Wonders poofed into existence randomly (civ1)

Heavy production, research, and tax revenue cheats at high levels

At higher levels the base combat values for units were no longer valid (all 3 civs) That's right, they changed the base combat attack and defense values. That's how their destroyers (4-4) on civ2 could match up with your cruisers (6-6) on civ 2. On 3, the computer gets an experience cheat at higher levels to get the same effect.
Not to mention the infamous French make peace with Germans, French receive tech a tech b. French attack Germans.........French make peace with germans French receive tech c and d.....


Germans attack french.... germans make peace with frence, receive e and f.

The war and peace declarations and tech trading just wore on and on. And in 3, they just trade techs all over the place to totally outpace you.

Not to mention the infamous no fog of war and perfect knowledge of location for even your stealth units (Kiss those subs goodbye baby) I'd say Galciv cheats less than any game AI I have seen.

And It should be very hard to win a game on the top levels, Not easy. Ai's just aren't quite there yet, but this one is close.

You can win on Moo3 on impossible right now if that's all you desire in a game. The ai won't ever attack you either (and it uses cheats!!!) You can just ram away at it untill it submits. Isn't that cool? Seeing as how QS is getting hammerred mainly over that issue, I think not.

That said, the single biggest advantage the human player has, based on what I've seen in the metaverse submissions, is the Ctrl-N feature. Players can just keep hitting that until they get a really good start. And let's face it, with a good enough of a start condition, even if it were multiplayer, a great player will have a hard time winning out if the starting conditions are stacked against it enough.


Frogboy, I haven't tried the upper levels yet as I was told there is no advantage to the metaverse in doing so. I'll move up as I go. But I clobberred most Civ games even on impossible. You have to use diplomacy, trade alot and still half the time, it wont work. You forgot another advantage the human player has. The save. When you make a mistake (perhaps I shouldn't have sneak attacked that guy at the moment, you just go back to a save where you know you're in it and move on. I only do that in games where you get "the invincible phalanx" like in civ. LMAO yup one phalanx took out 10O ships (cruisers, battleships, I pounded everything at it and it still wouldn't die. The AI at that point obviously decided "You're going to lose." I reloaded hehe. IMO it's cheese unless you have a good reason like the above. As for cntrl-n, that's not cheese. It simply saves time, since everyone put in a bad spot would simply go through the rigors of restarting.

I had to when I started in a 4 star cluster (1 habitable besides earth, and the nearest star was over 7 full sectors away!!! (gigantic) That's not just wimping out to hit cntrl-n. That just wasn't a viable start position.

My problem is not so much worrying about the economy of the AI, because I play at Intelligent level so far, and we are even up on that.

My problem is at least one of the AI's gets some super system, i.e. 1 system with 3 or 4 habitable planets (which when I found them were all 20+ in PQ) in it. (and I'm playing occasional setting as far as PQ, medium galaxy, scattered) Under those settings a 3 or 4 habitable planet system is quite a find. I have noticed this in two long term games, so I can't truly say it's a pattern or problem yet.

After testing Cntrl+N's several times, I could never do better than 2 habitable planets in one system, and that was rare, and usually 19, 15 PQ.

I'm not complaining, not yet anyway , just making an observation.

~SDC~

To clarify, I mean one STAR as a system, not the sector. In one case, the sector also contained another star with multiple inhabitable planets, giving the AI something like 6 habitable planets in 1 sector. That seems like a lot to me for Occasional setting.

~SDC~

Starting position presents a tremendous advantage to the human player and I wholeheartedly support the decision to give the AIs big bonuses at the higher difficulty levels just to give a good challenge to gamers.

For my high scoring Incredible game that got deleted, I managed to colonise 102 colonies through an excellent start, poorly placed AIs and speed+2. Probably anyone could have won that game with their eyes closed and I did so in 20 years through some deft diplomatic maneuvering.

With a healthy dose of luck, skill and good starting position, the Incredible level is beatable. With an average start, it there is still a good fighting chance through going for the tech victory. In my last 1.035 game, I managed the tech victory just when my fleet got decimated. I believed I would still have won if I were 1-2 years away via tech as the AI still needs to conquer my 70 undefended colonies.

I am currently winning my new Incredible Gigantic (at 50% military victory now) game through a good starting position (68 colonies) and good luck as the powerful AIs started some distance away and decided to beat each other up instead of picking on me. Of course, I worked hard to ensure that they are frequently occupied by more pressing matters than myself. However, I used the exploit of ring-fencing found resources and I believe that also made a difference.

The difficulty level at Incredible seems about right to me now. Maybe there is even room to make it harder once the overall skill level of players increase over time.

Frogboy,
I also used Control-N quite a bit until I got comfortable with the tech tree and my strategy.

However, it is so easy to win rushing to trade goods that I now actually use Control-N to give myself a worse position just to see if the strategy still works (it does).

Regarding the 0%/100% strategy for sliders, I think it does make the game too easy and I wouldn't mind seeing some limits on these based on your government type (a la Civ).

Your max for any slider could start at 100% (Imperial) and drop 10% for each successive government type since you have less and less control over your systems.

You'd still want the economic bonus for advancing to the next government, but there would be a trade-off.

But i still don't think that difficulty must be so high that it forces players to use Diplomacy (i hate diplomacy, i like to win on my own, also comp can't use diplomacy that good, so it's cheese tactics too IMHO), good starting position (cheese, i know) and military way (well if you can win vs 2x+ number of forces, it's cheese not because of implementation, but because of result) etc. to win a game. I like to win with better economy!

Also, comp already using one 'advanced' uber strategy - it uses it's economic advantage on Incredible to build military fleet, and forces you to build military fleet too. But, because you have worse economy, you spend more in % than comp, and you have even less resources (in % compared to comp) for driving up your economy than if noone builded a military fleet. So, AI advantage is even more. In other games, comp gets bonuses, but plays in inefficient way. Here, it uses it's advantages to get even more advantage. So, overall, it has best economic bonuses (on Incredible difficulty). (Respawning wonders in Civ2 is not economic advatage - it's cheating).

P.S. Are you ready to 30.000 bc cost for declaration of war between comps in next patch?

~SDC~

The difficulty is so high on Incredible to challenge cheesy players (which in your definition includes me). If you don't want such high difficulty, there is SURPRISE! lower difficulty levels in this game

I do think that computer AIs should be aware that warring is extremely costly and they should see some clear advantage to go into it. What I do not want is making it literally impossible to bribe them or making them declaring peace soon after accepting the bribe (I remembered it used to happen in Civ).

Ugh! Ellestar, it's not called *Incredible* for nothing. What you seem to want is to play on Intelligent, but your ego won't let you step down to that "lower" difficulty. Might I suggest you go into the data files and edit Intelligent to call itself "Incredible" and be done with it.

The whole point of Genius and Incredible is the odds really are stacked against you. The AI is more efficient in terms of managing its econ. Then, it gets bonuses to that econ. On top of that, Brad's been playing around with activating AI routines that only the most anal retentive of humans could match (so far just the Arceans). The result is that you cannot beat them in a face to face military confrontation - you *must* use diplomacy or "cheese". That's the way it's designed: let the wily human beat the uber computer through tricks the computer *can't* do. If you want to win in a "go it alone" military confrontation or even cultural war you're going to need to step down a few difficulties and stop the whining because it's never going to be the Incredible difficulty you're asking for.

There are still a few issues where the ai needs some bonuses.
I finished a game on Genius, where the ai used an AMM to destroy one of my mightier ships.
Namely, a scout. :/

Personnally, I think Ctr-N is cheesy, because "As for cntrl-n, that's not cheese. It simply saves time, since everyone put in a bad spot would simply go through the rigors of starting."
is wrong. I don't restart. That may sound silly, but at least you learn a lot from bad starts (even though you eventually lose).

Well if you think that you are so good rushing on Gigantic Incredible AI just consider these things:

1) What is the best starting bonus computer may take then on Incredible it have other bonuses? Answer is simple - +2 to speed. How good will be your rush with just this one setting? And why developers don't give to comp this bonus?
Other bonuses just give another 20% to something like 1.4, so it will be 1.6... Who cares? Just another thought - why it has other bonuses, but not speed?

2) I think that during colonization comp uses some algorythm with yellow stars as points of a net. So, it's just a variant of 'find the path' algorythm. Why comp can't use same algorythm to predict virtual borders of the empire (don't forget that everyone knows each other starting locations and comp knows about yellow stars)? It will work fairly well on Scattered. It can note colonization progression over time (again Influence). It can calculate your speed. So, it can predict meeting with a player. And forget about processor time - things like this are 20+ years old.

3) Comp is constantly buying buildings at the start. Who don't want to have Gravity Accelerators on a Gigantic galaxy? If comps will try to get gravity accelerators... Of course comp with best respawn will get it. So, colony ships of a most powerful comp will have a speed of 6. And it will be 7 just several turns later.

4) So another thought - why comp not buys wonders/trade goods, but buys buildings for a same cost as wonders? Seems not that efficient, isn't it? So comps can easily get all Wonders/Trade Goods. Exchange trade goods for techs please.

5) It's clear that comp can get decent speed and can predict future borders of it's empire. Also, it can calculate range of your ships (precisely, if it has contact with you and Diplomatic Translators). So it can move fleet from the center of it's empire to it's borders. If it loses planet to an invasion it can just set social spending to 0% and military to ~70% and stop buying buildings. Also it can pay to a nearest to the player AI for a declaration of war.

6) Why comp just waits something? Demanding tribute? It can just build several fleets (with transports of course). Then he got some planets, it can demand tribute for a peace treaty. Rinse wash repeat several turns later.

So little changes, so much impact. I bet that (1),(3), (4) and second part of (5) can be implemented in just an hour or so. Developers just don't want to close all ways to victory. (Dungeon Masters? And we are playing a quest, or at least it seems so from this point of view.) And try to say that this will be unfair.

I just want to get back 4*PQ taxes cap for Incredible AI. I said why.

~SDC~