Probably wise - good to take a break, and to let people play through some games and see what happens. And I'm keen to see 3.1! Some of the stuff, like the Drengin/Korath economy lag, doesn't seem too much like it needs fixing, since it's a brake on their military ambitions.

There's no point aiming for perfection when there are fundamental problems with the AI (for example) that can't be fixed, but you have made a vast improvement on the GalCiv that I quit playing a couple of years ago - and the discussions on this thread have given me plenty to think about when screwing around with the game myself.

MK's tiered trees do some great things - in particular, breaking up the weapons and colonization techs and making it impossible not to research certain key techs. But that said, even playing with an earlier version of his mod (I haven't tried the latest yet), didn't prevent wars breaking out early - I'm really not sure how that could be prevented and just consider it an inevitable part of the game.

 

 

Release 3.1 and take a break. There is no point in continuing the work, if you have no idea what to do next. Play some games and enjoy what you have done. Maybe you will get some new ideas later, but until then, just relax. You have done a good job and deserve a break.


Another way would be to use several tech requirements per technology. You can have two or more requirements per tech. For example, I added Fusion Power Plants as a requirement to Xeno Factory Construction (in addition to Xeno Industrial Theory) and it worked. I then did the same with the other factory techs and it still worked. Even my change to Alliances (requiring Advanced Diplomacy, Trade and Interstellar Governments) worked flawlessly. The only downside to this, is that the tech tree looks horribly. The tech will be shown after all of the required techs. This is just a UI error, but is quite confusing.

I tested this a few months ago, just to see if it will work. How the AI would handle this change, or if there are any more problems with it, is still unknown. But at least a Civ-like tech tree seems to be possibly.

Wow, this works? I'll have to play with this and see if I can stand the look of it myself, because it definitely opens up some possibilities.

I have seen references to UI mods - would it be possible to change the look of the tech tree to accommodate multiple prerequisites?

Wars will inevitably break out, but if all the races have time to grow their economies and develop research capacity, they need not be one-sided conflicts.  Except in those cases where one race gets completely boxed in and gets very little share of the colonies.

Good work on the mod, I didn't really even try the base game after reading the changes here. As a new player I can't really help on any balance issues of yet, only thing I know is I'm getting kicked around as Arceans.

It definitely does.


I have no clue about UI-modding, so I can't answer that question. It certainly would be great, if it were possible.

Thanks for the pep-talk.  I think I was just stunned when I imagined the amount of work that would be necessary to avert early wars.  After some breathing I'm feeling much better.  It's funny, I said that the Drengin and Korath lag economically, which is true in general and during the first two years of my latest playtest.  Then, I play out a couple more years and the Drengin economy rockets up to 3rd place.  Unfortunately, the superpower Drath are right next door and came knocking with a massive fleet of transports and warships.  The Drengin lost a few planets, but have rebounded and are fighting for their lives.  It's way uphill, though.  They do have several very well fortified starbases, but the bases are far from the front and aren't drawing any fire.  Compounding their misfortune is that they are in the middle and the Drath are in a corner.  The Drath's other neighbors are the Altarians, who are unlikely help, and the Terrans, who have yet to militarize.  Their only real hope is the Drath start picking on others, especially the Terrans, who could probably arm up pretty fast.

Speaking of that, it's a bit weird.  Lots of civs in this game have all three defense theories, and a few have more than one actual defense tech - Titanium Armor, ECM, etc.  At first I thought the Drengin were using Starbase Defenses Mk II, but they're actually using all of the most basic specialized defense modules - Energy Fields, Starbase Armor, and Chaff.  Huh.

For military balance, now that I've significantly dropped the AIValue of Planetary Invasion and Planetary Defense, it seems that complete failure to militarize is less a problem.  Sure, some civs are still going to get steamrolled, but I don't think it's possible to prevent that.  I wonder what would happen if you removed the tech requirement for the most basic weapons - Space Cannon, Sparrow, and Particle Beam in my mod.  Year 0 wars?  At least nobody would be left defenseless.

Is early war aversion even necessary? The galaxy can be a dangerous place...

In my opinion, year three is not all that early a time to have conflict  - I will sometimes rush invasions + basic weapons to zerg an enemy who I feel has started too close as early as year 0-1. (unless the AI is doing this with no regard to economic sustainability, or it's always the same aggressors)

In related news, I have hashed out a weapons rebalance prototype - would you like me to post it here in its enormous table-riffic glory, or pm/email you the excel file?

I liked the sensible damage progression you had, so I kept that while hopefully creating enough variance in the weapon types to be interesting. The main advantages of the various types are size for beams, sizemod for missiles, and cost for guns. I did my best to balance them via derived statistics like damage/size (at each hull size) and damage/cost, but I'm going to need to test in order to find out how much damage/cost is worth vis a vis damage/size.

I've combed through the early portions of this thread and MarvinKosh's space weapons thread to find how you and others test, and I'll be running through some test games to see how the AI does with the altered weapons. As is, I suspect that defenses may need tweaking to fit with the changed weapon balance, but I want to see what the problems are before I go trying to fix anything.

 

As a side note, once I get a general idea how relatively valuable weapon cost reductions are, I could work up a set of neutral weapons/defenses with half the 'value' of the good defenses and evil weapons through cheaper component cost.

So I'm 3 years into a playtest where I dropped the requirements for all the basic weapons and defenses - the stuff you would get with the corresponding theory tech.  I also dropped the tech requirement for Battle Stations I, Starbase Defenses I, and the attack and defense modules you would get with the theory techs.

Turns out there's another chokepoint for warship production - Interstellar Construction.  For whatever reason, the AIs didn't have any fighters in the build list.  It wasn't until they started getting IC that heavy fighters became available.  From what I can tell, you can fit a basic weapon onto a tiny hull, but the designer wasn't making them available.  Not that they need to be cranking out fighters a couple months into the game.

It was weird at first, because the military graph was going up and down.  I thought they were maybe building fighters and scrapping them.  Turns out there were some 1 beam colony ships floating around.  Build one, military goes up.  Colonize a planet, military goes down.

Other than that, things are pretty typical for three years in.  The Korath, Yor, and Drengin are doing very well with no changes.  The Yor always do quite well, but it is refreshing to see the Korath and Drengin running fast for once.  I'm thinking that I've been subtly biased against them in choosing a galaxy.

I have been plagued by all the AIs choosing the same weapon, though.  I used to get a little variation, but my last three or four tests have been 100% for one weapon.  It's pretty lame.  I'm pretty sure it's determined during galaxy setup, or something.  Even all the colony ships had 1 beam.

Yeah I think that in the absence of a technology requirement, the AI will happily spam the smallest weapon it can find onto a ship design.  That's usually going to be lasers.

In my mod, all weapon types have identical characteristics.  The drive to beams seems to be due to "just because."  Those who have gotten around to weapons research have all gone for beams, so it's just the flavor of the game, I guess.  It must be predetermined and global.  Even though they have access to all types of weapons, their build list only has ships with beam weapons. 

I don't know what triggers the game to design more types of ships, but it's also an issue if you trade weapons techs.  I've given guns to an AI, who remained defenseless until they got around to researching missiles and made missile-boats.  I've also given lasers to an AI, who remained defenseless until I got curious, took them over, and gave the Beam Weapon Theory for free.  The next turn they were building ships with the best lasers I gave them.  Was it the weapon theory that triggered the design?  Was it researching it themselves that did it?  I'd guess it's one of those or both.  Just like my current playtest didn't generate warship designs for the AI until they actually researched a tech that would allow for new ships.  Even then, before they had researched any weapons techs, they could only make heavy fighters with beams.

I allowed myself a slim hope that giving them immediate access to the weapons would spur variation, but no luck.

It wouldn't surprise me if Weapon Theory techs have a polarising effect on the AI's research and ship design efforts.  That could explain why the Thalans by default are so intent on developing beam weapons - they lack any sort of cue to develop the other weapon types.

The reason for this is obvious - if the AI pursues multiple weapons types, it will make less progress in terms of damage output than if it dedicated itself to one type of weapon.  So having techs categorised as beam, missile or mass driver helps it to avoid doing that, possibly by giving a negative modifier in the research decision tree.

If the AI is coded correctly then this modifier (if it exists) would disappear when enemy ships become practically impervious to that weapon type, and it would become attractive for the AI to pursue a different weapons tech line.  But I've never seen them do this except when the tech is categorised as something like Trade.

While playing out my Yor game, I've come up with a couple tech names for Yor governments if you're looking for them.

Distributed Decision Network (I believe you mentioned this one earlier, and it fits) > Enhanced Individuality Protocols > Cybernetic Hegemony

 

Also, I think the Precursor Ranger event may be a little broken - the Drath found one recently, and it's the most pathetic precursor ship I've seen. I suspect it may have to do with the weapon/defense component changes, unless others have come across equally weak Rangers in unmodded games...

 

Whoops.  I just went and took a look at the ships and I removed all the weapons and defenses except those few that remain.  I guess it falls to me to design the new ranger.  That thing is ridiculous, though.  Three Doom Rays, three Nightmare Torpedoes, and four Black Hole Guns.  Mix those up with two each of Aeron Missile Defense, Zero Point Armor, and Ultimate Force Fields along with a splash of other weapons and modules and you get insanity squared!

Nearing completion here.  I need a little feedback, though.

I settled on 10b population for each planet, along with 1pp, +% farms.  10b makes it so you don't have to deal with fractions if the farms have 10% increments.  I'm testing with 20/40/60 farms for 12b/14b/16b populations.  Is 16b a good cap, or should I go 30/60/90 to top out at 19b?

I'm happy to say that the Drengin are doing fine.  I gave them a logistics type GA that also has 50% planetary defense and all the morale and resistance that goes with it.  I also gave them a Superior Warships tech to boost stats and allows for a +% Starship Quality shipyard - kind of like the Stellar Forge, but not as awesome.  The Stellar Forge itself got a nice boost to make it pretty awesome on its own.  The Korath should be fine with their unique techs and a much improved, GA Aul Incinerator.  The Doomsday Generator boosts overall HP and functions as another +% shipyard to retain its goodness.

One of the last things I'm looking at is breaking up the unique Drath/Altarian techs and spreading them around the TechTree, instead of having one big line for all of them.

Anyway, so how about those farms?

Personally, I'm not a big fan of the % solution to farms.  For one, we've already got a % improvement.  Secondly, this makes food bonus tiles not only redundant, but completely and utterly useless.  I do however like the 1pp limitation.

I haven't tried this-farms is not something I've gotten to-but have you considered having farming tech unlock upgrades to the initial colony and/or civilization capital?  Coupled with a small standalone 1pp farm (say 1 or 2) at some point along the way, this would allow you to 1) have starting pops differ between colonies and the HW 2) have ending pops be (almost) the same for all planets and 3) still use a 1pp to solve the food issue while not making food tiles completely worthless.

Let me know if that didn't make enough sense and I'll try to clarify.  Also, granted, this makes the % food improvement that we do have even more worthless, but I'd prefer that to having even more reason to hate food bonus tiles.

20B is the absolute max sustainable, but I've not seen an AI able to handle that level.  12B would allow them to keep roughly 2/3 (67%) of their morale bonus, 14B 60%, and 16B just over half (52%).  16B is probably the best bet, but if you had to turn it down to 12B I wouldn't be horribly surprised.

Ideally, food production bonus tiles should come way down in terms of the bonus they offer, so that it's still worth building a farm on them, but they won't create as much approval woes for the AI and human player.  Unfortunately those aren't coded in XML or someone would have tried this already.

 

Tolmekian I want to congratulate you for your mod. I have played tens of GC2 games, and now I have tried your mod, specifically v3.0 with smart minors.

I like the revised techtree, especially the removal of the ultimate weapons/defenses, they seemed a little too powerful, even though the AI would research them too towards the end of the game. I also enjoy starbases that can hold their ground.

What I found odd was the minors building many influence starbases... why? Can they influence planets into rebellion? Also the AI seems to enjoy getting massacred by starbases. I witnessed entire well-armed starfleets getting wasted on trying to destroy a useless economy starbase instead of destroying enemy ships. This also happens with fleets, as one of my enemies had tens of "private help" ships, those little inoffensive Vengeance class, and he kept throwing them at my fleets getting slaughtered. I can't remember what the AI did in vanilla about this. I know they attacked starbases but in vanilla they could actually destroy them easily.

Regarding farms/population, I think it's fine as it is. It's fairly balanced with regard to morale and to the income population generates.

Looking forward to more features!

No, although I'm pretty sure they don't know this. It's also possible that they're building them as range-extenders.

Is the AI actually building farms with a starting limit of 10b in your tests? As I mentioned earlier, the AI totally ignored farms with a starting limit of 12b. I also tested it with 10b, but it seemed to be same there, too.

A cap of 16b seems fitting for the Torians, but the others could do with a little less (14b or 15b). The Yor, on the other hand, really need the population to keep up economy-wise, so giving them the max cap of 20b would be best.

Another possible problem is that, if the AI hasn't had access to farms early on, it won't/can't build them on all of it's planets. The Iconians and Yor were doing fine. The Torians were building their Harvesters, but when it came to farms, they, along with the other races, only build them irregularly. The only solution I found, was to add a new Basic Farm improvement and to give all races, which don't start out with a farming improvement, access to it. The only exception to this are the Thalans, because, IMHO, it doesn't makes sense lore-wise.


I'm not very fond of it either, but it is the best solution, as of yet, for the AI's overpopulation problem.

If an improvement provides a food bonus of even 1mt, you can be sure that the AI will build it on a bonus tile. However, it will never build enough morale improvements to compensate for the constant decrease in approval due to the increase in population. The only exception is AIP 7, but that one seems to be overcompensating by building one to four morale improvements per planet, while only rarely building farms. Changing the morale improvements to 1pp and increasing their AI value, so the AI will build them on all planets, handles one part of the problem. But you need to know what the max population for a planet will be, for a given race, to be able to decide how big the morale bonus needs to be. Food bonus tiles make that impossible.

I already removed the Food Distribution Center in my mod, because, as you said, it is superfluous now.

When it comes to the Yor, the %-bonus actually makes sense. Food bonus tiles are areas with extremely fertile soil. Why would that increase the electrical output of Charging Stalks?


Interesting idea. I need to test, if it actually works.

Thanks for the feedback, bbr91.  We, as modders, can't change the basic behavior of the AI.  Minors building influence bases and the AI obsession with starbases are things we have no control over.  It actually works pretty well for the AI.  Those who are slower to build warships will build tons of constructors.  Those constructors will fortify the starbases, which serve to defend the empire by attracting enemy warships, which allows the weaker AI time to build up a defense.  I can see how a player could abuse that by plunking a super-starbase down to devour the incoming enemy, but just like I can't mod the AI, I can't change what a player will do with my mod.

Not likely.  They usually build them very near their home system.  Although minors are exempt from the influence game, they do generate influence and seem to take it into account.  For example, when I force minors to surrender, the influence lines of the nearby majors expand.  Even though the minors don't control any space, their influence does seem to push back that of the nearby majors.  So, I'm guessing that the minor AI "feels" the press of major influence and tries to shore up their own influence with starbases.  That would be perfectly acceptable when having the minors use AIP 7,8, or 11, but when using AIP 5 - the minor personality - they simply shouldn't ever build them.

They build farms just fine.  I have the AI value for farms set at 10.  Factories and morale structures are 10, labs are 12 or 14, TGs/SPs/GAs are 10.  With those settings the AI seems to build a pretty good mix of improvements.

Most civs research farms early on with my settings.  They don't build them uniformly, but for the most part they use them well.  Some civs seem to like them while others don't build a single farm, even though they have the techs.  I nixed the Robot Farm and Harvester for v3.1, and it's testing well.  Oh, I changed the Hyperion Matrix to +20%, and any GA/SP that boosts pop is also 20%. So everyone is capped at 16b, while the Thalans could have 2 planets go to 18b with the HM and Gaia Vortex.  I think the only other one is the Orbital Terraformer, which allows another 2b.

The Yor don't seem to have any economic woes in my testing.  They are usually pretty powerful, though in my latest test poor starting position has left them in a rough spot.

The AI just can't be trusted to deal with the fallout from high pop planets. I'm okay with farm bonus tiles being worthless.  Having zero value is a step up from having a negative value.  Just like you say, it's impossible to adequately solve the population vs morale issue without strict limitations.  Speaking of AIP 7, when I was tinkering with it for v2.0 I got it to research farm techs by changing the Category to Biology.  Whoops.  They build virtually nothing but morale structures and farm on all their planets, like 2,3, or 4 of each.  They also really like Planetary Defense, which I assume is because it has a morale bonus in my mod.  You'd end up with planets having nothing but farms, arenas, and PD.  Kinda silly.  That's when I gave up on optimizing for multiple AIs and went back to AIP 11 exclusively.

I agree.  Very interesting.  You could do a lot by upgrading the IC/CC.  Imagine not only upgrading the IC/CC for more food, but higher morale.  Forget forcing population and morale limitations with 1pp structures, and do away with farms and morale structures entirely. I'm not necessarily advocating it, but you could hit a lot of things very elegantly by simply upgrading the IC/CC.

As a last thought, while typing this I wondered if colonizing minors could handle AIP 10.  The don't usually get a whole lot of planets, so maybe the planet improvement bug wouldn't hit them.

I can't properly test my changes, due to my hardware problems, so I'm not really surprised that the AI does build farms with a starting limit of 10b. Thanks for confirming it, though.

I will set the starting back to 10b, but keep the max limits as they are for now (14b/15b Arcean/16b Torian/20b Yor).

I haven't yet optimized the research for the AI since my last re-write of the tech tree files. No point in doing that, if the game crashes constantly. The AI is literally on it's own currently. Every tech, except for fleet modules and Terror Stars, have an AIValue of 10. This is the most likely reason for the differences between our mods. The AI in your mod is able to research farms much earlier and has, therefore, still space left to build them. In my case, the AI often had it's planets fully developed before it even began researching farms.


In the case of the Torians, I'm currently thinking of replacing farms with upgrades for the Harvester. This should solve the AI's potential problem of not having any space left to build farms on. I might give the Thalans a unique farming improvement, too, for the same reason. The Thalans are currently the only races that doesn't start with any regular farming improvement. Shrine of Tandis is the only GA/SP I gave a food bonus, because it already gave a 25% bonus to everything else. I'm not sure yet, if I give a food bonus to any other GA/SP, too.


My main concern was that, if all races were going for max population on all planets, the Yor couldn't keep up economically using the same limit, because the other races have a much higher potential income per citizen ratio. Even if their economy is strong, the Yor may not be able to maintain a fleet as big as the other races could, and every lost population hurts the Yor much more than the others. Still, it remains to be seen if my fears are really justified, or simply based on my own playstyle. Wouldn't be the first time that something isn't anywhere as bad as I first believed.


There is something odd about it. I gave the upgrades a cost of zero, just like the base improvements, but they took several weeks to build. Maybe it happened because both, the base and the upgrade, had a cost of zero, but it's still strange. Well, at least I can confirm that it does work.


I don't see why they couldn't. It would certainly be interesting to see, if they will start colonizing like the Torians.

Hi there,

First, I just wanted to say thanks for trying to fix all the problems with ToA. You've inspired me to try this game again after giving up on it years ago due to all the bugs, balance problems and shoddy AI decision making.

I've just started experimenting with your mod and I'm still trying to get things set up the way I like it for a challenging game. I'm a bit confused about the different AI personalities and just wanted to clarify whether I can just set all the AIs to generic and still get a decent challenge? Or are there specific AIs that really ought to be set to something else to perform well?

For context, I tend to play huge or gigantic maps, scattered distribution, lots of stars and planets, occasional habitable planets, rare super planets, 9 opponents, 8 minors, the difficulty set to tough, very slow tech, no surrenders, no tech brokering.

In my first try with your mod version 3, I found that the AIs didn't expand very well in the colony rush phase which resulted in me gobbling up enough planets that I was probably going to win no matter what. To be fair though a lot of the AIs started in weird spots in that particular game, I had a quarter of the galaxy to myself for REXing. I gave up when the Drengin surrendered to me as it felt pointless to continue with all their planets added to my already large empire (I forgot to turn surrenders off, things like this are why I hate the surrender option).

Anyway, long story short, I'm hoping for some insight and tips on what game settings to use to get the most out of your mod for a balanced and challenging experience where all the majors have a reasonable chance of being competitive.

Thanks again for all your work on this.

Thanks to you for trying my mod.

I have all the AIs set to Generic in the mod.  Well, to be precise I have them set to AIPersonality 11.  With those settings, the Altarians, Arceans, and Korx will default to their unique personalities, which are based on AIP 11.  Those settings should make for a good game without you needing to change anything.  While it is true that starting position determines a lot, the AIs should be able to put up a fight if you're playing on tough difficulty.  I'm trying to think of something that could cause your problem, but it's not coming to me.  In fact, your settings are pretty close to what I use for playtesting, so the mod is practically designed for your games.  I'd suggest giving it another shot and seeing how the AI stacks up on a second run.

I'll see if I can think of anything in the mean time.

 

By the way, what minors are you using?  Dumb, smart, colonizing?

 

Thanks for the reply. I'll give it a go again, I think maybe I just got a really lucky start (and the AIs mostly got really unlucky start positions. To be precise, only the Korx, Thalans and Torians were doing well, the rest were anemic.)

I was using the smart minors, but I might also switch that to the MOO2 colonizing version for some fun.

If all else fails, I'll turn the difficulty up another notch to boost the AI economies.