As I have used 6 fully developped party palace to flip sectors, I know about the influence problem .

My point is: don't make a specific formula for the influence module. The current formula is:
the effect of all modules in a sector stack.
The problem is that for all non-influence module, stack means an addition (bonus applied on the same base) when for influence module stack means mutlplication (bonus applied on base integrating previous bonus).

So they shouldn't change the fact that effect of modules in the same sector stack. They must just fix that stack means multiply for influence module (current situation) into stack means add for influence module.

Hope this is clearer.

Thanks, Walldorf, for the clarification.



This seems like a really good idea to me ... the Law of Diminishing Returns.

Perhaps all starbases can have a base 10/turn maintenance cost but more bases in the same sector would have the diminishing returns you propose. In fact, I'm inclined to propose that all non-resource bonuses work this way with the starbases (in the same sector). I'd leave the resource bonuses alone, however, just to keep the importance of resources alive.

The biggest problem I see is this ... could the AI be coded to recognize these variables and adjust their strategies accordingly? :notsure:

I have a problem with the law of diminushing return: how do you fix which starbase will have the full effect of their bonus?
More, I am not sure if this a great idea to penalize a further people who will have UP setting limits on the number of modules.

And I will add that the 10/turn maintenance was removed since it wasn't preventing people who where using starbase to deal blows to AI (terrors star or party palace) and disbanding them after the sector is cleaned from AI system.

Starbases aren't an exploit... no more than ships are if you have more powerfull ships than the computer.. I mean to say.. just because its a way to beat the computer its not a @#$ dam exploit... there are many.. many.. many ways in which you can beat the computer.. the use of starbases is just one of them... sheesh.. it makes me so mad... I suppose you guys think that getting everyone allied to you to be an exploit too... or planting ships around AI's planets so that they can't get any trade or ships out an exploit too... well your wrong... In war there are many stratagies.. eash just as ligitamit as each other... why pick on starbases..

deathstars are another one... is it an exploit in the game that you can go around blowing the AI off the map.. no... its simply a way that the programmers have programmed into the game so that we can win..

If you want to play a game againsat the computer that you can't win.. then I'll program you a game thats impossible to win and simply not tell you that you can't win.. eg: tic-tac-toe ..

there are many programs that get that hard that you can't win.. i for one am glad that GalCiv isn't one of them..

and those winging that masso is so easy to play and arn't playing it.. then why not... i mean if its so dam easy then everyone would play it..

the ppl at stardock even added cheat features to GalCiv bucause ppl couldn't beat it.. so come on.. its either too easy or its not..

ALL GOMPUTER GAMES get EASY if you work out how to beat them..

Thanks StarDock this great game and excellent support and improvements to the game...

The exploit is using 6 fully developped cultural starbase (+343%) in the same sector.
If their influence add, you will have a bonus of +2058%
(influence multiplied by 21)
But currently, they multiply !!!
So you have a bonus of (1 + 3.43)^6 -1 = 668 100% which is insane. Here is the cheese.
[Message Edited]

in a game where you can create your own mods.. all you really need if you really wanted to cheat is one starbase with lots of culture mods... so no matter how one tried to stop it it wouldn't..

Stardock built it into the game so that you can win this way... why fight it... if you don't like it then write your own game...

as I said before.. if you get to know any game well enough it becomes easy and cheesy... I hate arcade games for this very reason.. If you don't want to use your or others knowlege then thats up to you.. only you can decide how you want to play.. but don't just label a tactic that works for someone as cheesy.. how about those who use starbases to hyperboost their trade.. (I once played a game like that where iI had every sector along the trade route full of starbases and simply won that way.. I could buy their planets ... is that cheese too...

J.T.Qwerk, it isn't annoying you that for influence module (and only influence), the bonus are multiplying instead of adding?

What is annoying is that for only 70 constructors you could have + 700 000% influence bonus. If this isn't a loophole, what it is?

I see no problem with putting a lot of time to create trade starbase in each sector along a trade route. BTW, have you tried the "vagabonding cash cow" : fully trade terrors escorting a pack of 10 mini freighter? For me , it is a perfect strategy. It isn't like you gain 1 million BC each turn.

Peace Phoenix, The AI use this to exploit the humans. I just finished a game in which the Alterians (?) we so far advaned in technology that they were two shiptypes ahead - so no chance of winning militarily. They flooded the galaxy with starbases and constructors, literally 6 or 7 starbases per sector that they owned that were developed all the way to terrorstar status. Almost all used the influence modules. The only way to win this game was to terrorstar their systems in oblivion. Which I did with 10 simultaneous terrostars built or moved up to their planets. I could do this because I was allied with them and they did not object.

~SDC~

nope... it makes no difference... multiplying or adding... its the same for human and AI players... the AI players just need smarter AI to utalize them better...

It would be a problem if the influence was calculated differently between players in the game... but as long as they are calculated the same way thers no problem..

others out there may say that your "vagabonding cash cow" tactic was cheesy... to me things are only cheesy when the AI players Can't do it as well.. (not simply that they havn't been programed to do so) can you imagine just how hard it would be playing the computer if they were programmed to use any of the stratagies that we use.... Somehow I don't think that stardock is going to go down that track because it'll make ppl angry.. (not me though thats what the difficulty levels are for)

So at the moment I can honestly say that I'm happy with the game.. sure there can be improvements... and yes there are discrepencies in scoring the difficulty modes between AP and GC... (AP is easier to achieve the same score than it is in GC.. and I think thats wrong.. but like i've said I think Stardock is looking into balancing that..unless they want everyone to switch over to AP.. and this is the cheese to get everone accross)

Again, have to ask: Any time-frame on the 1.21 patch? Will it address the very annoying save-game problems? (Like ending a game and showing only 1 tech researched when I had a "Beyond Human" win)

SLW, last we heard, the word is soon. (How soon? Er... we don't know. Could be a week. Could be 6 weeks. Or ???)
[Message Edited]

J.T.Qwerk... ROTFLMAO!

Now, you forget to mention that being able to move your ships is cheesy! Hows the poor AI supposed to be able to figure out where you are really sending them?

My brother always said that in any computer game he played and lost .. that the computer cheated..

his reasoning....
The computer knows where your moving to because you've selected a destination.. knows what youve selected to develope.. what weapons you have.. and therefor the program knows what your doing.. and can prepaire and take advantage of it..

Note: he never said it when he beat the computer

Sirian,

I really was laughing. The number of people that complain because a game, on it's top level, cannot "beat" them, makes me laugh. Why do people expect a few thousand lines of code be able to beat them? It doesn't learn... It cannot apply the lessons the player has learned. Only code updates will make it better. And for any "custom" game (non-traditional), this means that the game coders need to already be super masters... that they need to understand not only the principles, but also the fine balance of their game and every strategy possible in the game. And that they have to rebalance their game's AI strategy with every patch or enhancement... or even in their finely balanced code, the AI has a blind spot or does not properly know the advantage of the new values.

Why do you play games? To have fun. What is required of a game for you to have fun? According to you, it's a straight gamble... and nothing more. Perhaps you'd enjoy Slots or Craps or Hi-Lo more?

There are 2 kinds of games... those you know you have already won before you start (such as SMAC), and those you don't (Slots, Craps, Yahtzee, etc). Personally, I sometimes am in a builder mood (ie, games I know I've won before I even start the game), and others in a gambler (random) mood.

How do game designers make a game less predictable (and therefore, hopefully, prevent people from knowing they won because they've decided to play a game)? They add in randomness. Via resource locations, etc. But in "strategic" games, such as Civ, you know that you've won with a decent starting placement. The rest of the time is merely your own ritual of passing time until the game announces that you've won.

Now, for the builders, they like to know that they will win if they "build/strategize well". Sirian, you may not like that, but there are plenty of people that do. Picking on SD for making a game that qualifies (if only on the lower levels of play for some), I think is a strawman. It's the "When will they make a few thousand lines of code that can beat me?" scarecrow, all over again.

Yes, the fact that the balance point (human and AI have the same in game advantages) is high in GC shows that StarDock is after the non-grog (eitherwise known as casual) player. Because when you have a family and a job, you, as a consumer of strategy games, simply don't have the free time to become a master grog of the game. Not the average consumer of their product.

You are not an average player of these types of games. It shouldn't take you long to find all the strategies that are "unparalleled" versus the AI. Especially if you are an experienced MoO1 player, as most of the basic MoO1 strategies are equally valid in GalCiv... and that is for good reason. They are, at their core, identical games. 4x.

Complaining that GalCiv is a war game is another scarecrow. Its a game about the control of production centers, just as ANY game with production is. And in any production game that allows fighting, war is always the heart of the game. You cannot win in any such game (Civ, MoO, you pick what you want) without already being able to win militarily.

MoO1 is a fantastic game, a true master-piece of TBS in my opinion. But it isn't the perfect game you keep talking about. Your memory and love for the game seems to be whitewashing it. If you do not bribe the AI occasionally, you are guaranteed to be at war with it (because the game is hard coded to expect diplo gifts). You could prevent the AI from winning a peaceful victory, just by going "Loki". You could use the Diplo to get whatever you wanted out of the AI. Bribe it occasionally, and it wouldn't pick on you until there was noone else left. But eventually, it would be war. Unless you got bored and started a new game, or won by council first.

Complaining that a war game is a war game is just silly. Now, complaining that the AI is always "reasonable" is equally silly in the case of GC, but only because SD has decided they wanted the AI to be "reasonable". Reasonable means that it will leave you be under reasonable "in game" circumstances. So, you think it should be different, but the designers don't agree. That doesn't mean they will not or cannot change their minds. It means that that in SD's view, the AI being "reasonable" is more fun for them and the majority of their customers then having "MoO1" behavior. There is some randomness in them in GC, you know. Random events that can greatly shift diplomatic relationships. They don't occur often, but they are in the code.

I agree with your ascertation that the AI doesn't pursue a better path for a Cultural win. That could do with some refinement. It is something that SD has been working on. However, that seems to be much more complicated to compute then basic ship and invasion tactics. I suspect it is the starbase aspects... the AI could use with some improvement in its starbase development and enhancement. Just a guess of mine, of course.

Allied victory: An Allied Victory *is* a win for the AI, so long as the Allied Victory condition is set. So the AI "accepting" a shared Allied Victory condition is just fine. However, it allowing itself to be culture flipped by allies does seem to be a questionable behavior to me. Is the reasoning for that because SD just hasn't decided that changing that behavior is of high enough priority to change, or is it a more fun game that way? I don't recall seeing any explanation from SD on the matter, leaving us to make our own presumptions (bug versus planned behavior).

As for Tech Victories... I think the AI should be allowed to win via Tech victory myself. I believe that if they get the Tech victory currently, it just disables the Human from being able to win in that fashion. And that is, again, a decision by Frogboy and his team. To prevent a long game from suddenly being "lost", which is a common complaint with games that feature peaceful means of winning. Such as can been seen with MoO1 and SMAC.

So what's that leave? Oh yeah... the Diplomacy angle. You do realize that from the very start of the current GC, people have complained about being able to Loki via Diplomacy in GC? And that Frogboy has always said: That's by design. In the developer's own words, "you are the human, and the computer isn't. The computer doesn't get fun or satisfaction from playing the game. You do." Diplomacy is there for you to be able to get whatever kind of edge you want/need to enjoy the game. So, it's literally a "helper" for players. If you think it's too powerful, then please inform SD of how it is too powerful (such as the Diplo Baiting), and why.

I welcome your commenting and thoughts on GC and how you believe it might be improved. But I do wonder... if you have MoO1 and DosBox to run it on your computer, what are you doing here on the forum? Giving your processor a rest?

Nice to know that they can try to win with cultural means. I wasn't aware it was possible. Maybe because I play on large map and accept alliance win

I think that there is an option at the creation screen that said: disable alien tech victory. So AI could, if YOU allow them, to achieve a tech victory

All right, time for my two bits. I agree that there are weaknesses in the AI. It does some pretty predictable things and also some non-sensical things. But you know what? So do I. I disagree with the assertion that the AI only goes for military wins. I have had several games recently where the computer was determined to flip my planets (yes, I just said "determined" not "programmed").

J.T. I concur with you. Just because you can consistently beat the computer in a certain way doesn't make it an exploit. It's a skill that you have developed. When you get tired of winning that way, play a different way or play a different game. If you keep fighting the "exploits" you pigeon-hole the player into acting like how Sirian said the AI acts--you have to win by military victory. I don't like that idea. One of the beauties of GalCiv is that you can win in several different ways (and there are several ways to achieve each type of victory).

Anyway, I hope that was worth any where near 25 cents.

I fear that I haven't express correctly my point of view:
I think (it is my personnal opinion) that the fact that bonus from different cultural stabases in the same sector multply instead of add is a broken thing: it is like transforming 6 machine guns into one atomic bomb. It screws some balance. It should be put in the same category that terrors alpha strike from the day where you could build lots of terrors stars without having the corresponding money.

I don't think that the developpers have the multiply effect in mind when they have developped the cultural module for starbase. I could be wrong.

Star Pilot: Thank You!
I was beginning to think no one would read a post unless it contained an equation and a complaint about the AI.
If anyone knows more specifics...

what i'm getting at is that no matter how the Culture bases work if an AI player places the same number of modules on a starbase then they counter each other... so it doesn't matter how they get calculated.. adding , multiplyine squaring,trippling,quadding what ever... its the same for all players.. so its NOT an exploit

Just like playing someone in a game of chess.. the idea is to hide what your doing till its too late for your opponent to see what your doing till its to late... This is the same in any Tactic you use to win... be it military or culture or any other way you can think of

The AI's use of starbases needs to see that you've built a culture base and then try (if they can) to neutralize its effects..

Also the AI's never seem to break alliances... or break trade routes...

So what needs to be done to future releases of the game isn't necesseraly changing the math as its the same on both sides... they need to develope better AI... and this only comes with time... AI is the Hardest thing to program... I know... I can't do it.... I've tried... I think Stardock is doing an excelent job in trying to make it better... and i'm sure that they'll succeed with all the practice that they are getting.. Don't forget we as humans see things and can calculate and ignore usleess things a lot faster than any computer... the computers have to calculate everything .. so the more complex the AI routines the longer it'll take... Sure they can program it to look at every contigancy but then the game would take about a week to play 1 turn... like some of the better chess computer programs do at high levels of AI... This would bore me to tears... and then the game wouldn't sell.. so they have to comprimise. and cater for the most used and the best exonomical AI routines that they can..

I am under the impression that we haven't the same definition for exploit. Maybe because I am french. Let me see if I could sumarize.
If human and the AI play under the same rules, there isn't any exploit (your position)
If there is a hole in the rule (for me multiplying instead of adding) that human can abuse and that the AI won't intentionnally use or against which the AI won't correctly defend, it is an exploit (my position).

I would add that the AI see the cultural attack only if the starbase are in the same sector as his planets. it doesn't seems to check the influence in adjacent sector to see if there isn't cultural starbase that could threaten him.

Hi J.T.Qwerk, I apologize if I did not find the right words. Stardock does a great job and I love the game. I know how hard it is to program an AI and I agree that Brad does it very well.

And I don't mind when someone uses culture bombing or alpha strikes. I even do it sometimes by myself

I don't mind to beat the computer, too. In fact I would hate it to loose

The problems is that it should be avoided that you are doomed to these tactics when you want to submit to METAVERSE. I don't know if Stardock realized beforehand how much work it is to keep something like METAVERSE balanced

And no, I still don't agree. Multiplying the influence of star bases of one sector is far too powerful. It's not realistic and it's not what you would expect. And I agree with Peace Phoenix: I don't think that it was done intentionally.

I would not expect that it would be complicated to fix that. Influence calculations are complicated in this game, but the effect of multiple star bases of one sector shouldn't.

I don't know if Stardock realized this issue. When they did and don't want to change it, it's fine with me.

The only thing I do expect is that they fix bugs and I'm very confident they will.

You shouldn't be so sure: if a calcul is complicated and an undesired effect appears, you can safely bet that it is the complexity that creates this effect. So even if the fix seems simple on the paper , it could be awful to implement :notsure:

Woao I agree with Sirian in something (I have to mark my calendar)

Yes I agree with you that the AI's all go after the same type of strategy, I think that by tweaking this a little bit could make the game even more interesting, like Altarians going after economics and peace, drengin after military superiority and war, Yor going after tech development and medium relationship with others, it coulkd make the game really interesting, but I am not a game developer so I don't know if Frogboy could improve this aspect.

Not true. There are occasionally gems of games that come along and pose a real threat. Take a look at this game of mine - bear in mind that we were playing under varient rules (no ships ever) to make the game more challenging and fun.

http://www.network54.com/Hide/Forum/message?forumid=189557&messageid=1027379412
[Message Edited]