Aw, that sucks. I hope it won't be too hard to recreate.

I think I'd probably just be re-inventing the wheel.  The only thing I can think to do at this point would be to alter the weapons tech tree and add some new components so that the AI gets some decent defences without having to plod through the Defence tree.  It doesn't make sense to me for there to be so little gained with the early Defence techs which actually slow down the AI's development of weapons in the event that they're facing an opponent with better weapons.

The way I have it set up, the AI sticks to a pretty good balance of defense to offense.  This is true both with research and with their ship design.  My work is, again, constrained by using only AI 11.  After I polish my 7 main tech trees, I plan to run a test with a couple AI 8s tossed into the mix so I can see how the AI 11 balanced strategy works against the AI 8 blitz.

I so hope Frogboy fixes AI 7 and 10.  This is breathing new life into the game.  I actually PMed him about our work here.  Seemed important enough to mention.  Now, whether he has the time to check it out or ends up acting on any of it is anybody's guess.

Funny thing is, I haven't actually gotten a chance to play a game in this environment.  The Kushana AAR is going slow enough, but maybe by the time I'm done with this it will be time to start up another . . .

Why is this thread suddenly so quiet. It's creepy, like a haunted house.

Anyway, I noticed that I had set the AI for the Arcean's Unique techs too high, and they weren't researching space weapons any more. A simple AI value change fixed the problem I suggests that you do not set the AI of these techs to be much higher than that of space weapons, if you do not wish to reach the same fate.

For my part, I've more or less finished up work on the tech trees, and am giving modding a rest to actually play the game. So I'm pretty much done here. Can't speak for anyone else though.

Me, I've just been bizzy bizzy bizzy.

Can't say I'm quite near the end yet, though I have gotten over the hump.  I added the Terran tree to my mix, since I was running 2 Korx trees.  They're a pretty easy fix.  I overdid it slightly on the ethics techs..  It's a fine line between less than half researching by year 6 and a few starting to research them in year 1 and 2.  Pulled them back just a bit and it's time for the next test.  Maybe tonight.  I think I'm going to go all out for fun and toss in a couple AI 8s then turn on tech trading and super abilities.

I wonder mostly how the slow-to-militarize 11s will cope with the hyper-militant 8s.  Should be fun to watch.

Also, did anyone else notice that most weapons techs have no Willingness to Trade entry at all?  Think that's the reason the AI pretty much never trades weapons techs?

I ran my test through year 4 last night and it's going very well.  Tech trading and Super Abilities on, eight identical custom races with paired tech trees and Super Abilities.  I set my Iconian, Drengin, and Terran to AI 8, and the Korx, Drath, Torian, Yor, and Arcean to 11.

I set it up like that for a few reasons:

• AI 8 doesn't seem to like Extreme Colonization.  Since the Iconians start with Aquatic and Toxic, they can get those worlds and research the Advanced techs early.  It worked.  I considered setting the Torians as 8 because they can get Aquatic without Extreme Col, but I just didn't want militaristic Torians. See next.
• AI 8 is militaristic.  I have the Drengin set to 100 aggression, so they should build up a military early and go at it to get the most from Super Dominator.  It worked.  They're currently at war with the Arceans, Yor, Drath, and Torians.  They actually share borders with 3 of them.  Early wars mean the Dominator Corvettes are a big bonus.  They're a military powerhouse and are gaining ground . . . for now.
• I just wanted to see how the Terrans do.  They're supposed to be AI 8 anyway.  They're doing fairly well overall.  Not pushing wars of expansion yet.
• AI 8 doesn't seem to like the Precursor History line of the Drath tech tree, at least not the way I have it set up.  The Drath consistently thrive on AI 11, so I intend to leave them there.  They don't need a huge military anyway, it's not their thing.

Additional observations:

• Super Manipulator rocks when the Drath actually research War Profiteering.  They were able to take advantage of the AI 8 militarization to start some early wars.  The economic boost is huge.  With all the fighting by year 4 end, they were getting 942bc from war profit alone - that's just over half their economy!
• The Korx always seem to under-perform, but rarely get steamrolled.  Their Mercenary Academy provides a modest 2% war profit, but it's a good boost.  Also, war profiteering doesn't seem to work for wars that you're involved in.  Guess your mercenaries aren't fighting on the other side!  Cut-Throat Competition helps them keep up, and their Freighter Command ensures no loss of trade.
• The 11s seem to be able to cope with the 8s militarism.  After the AI 8 bullies started throwing their weight around, it didn't take long for the 11s to ramp up weapons research and shipbuilding.  As a result, there is little to no defense research, but that's ok.
• The Drengin have unique starbase mining techs. I forget the names right now.  The first one allows the Mining Matrix, which gives +10 resource extraction.  For some reason they won't use the Matrix.  I've seen constructors go to starbases that can only build the Matrix, then turn around and go away.  For now I'm intervening to make them upgrade, but I have to see if anything can be done to fix this.
• AI 11 is way too lax about claiming resources and upgrading its starbases.  The 8s consistently swoop in and steal resources right next to an 11's home system. Nothing to be done, I suppose.
• The Yor have the unique techs Firewalls and Anti-Virus Software, which are useless without Mega Events.  Since I didn't give them to my Yor clones as starting techs, they waste a significant amount of time researching them - ~950tp and ~1800tp.  It sets them back a bit, but they're always pretty strong.
• I haven't played any of my tests out to the bitter end, but I've gone through year 6 and 7 for a couple.  Overall, with the tech trees fixed and equal bonuses, it seems that no tech tree is terrible.  In that time it was quite rare for any 1 race either to dominate or get wiped out - unless they had a really bad start position.  I'm going to play this one out as far as I'm able to see if a tipping point can actually be reached.  Can the AI actually push any of the races to victory under these conditions?

I think that's the major things for now.  I'll post a follow up after the next 4 years or so, which I'll probably do tonight.

"The Yor have the unique techs Firewalls and Anti-Virus Software, which are useless without Mega Events.  Since I didn't give them to my Yor clones as starting techs, they waste a significant amount of time researching them - ~950tp and ~1800tp.  It sets them back a bit, but they're always pretty strong."

I am fairly sure those techs are supposed to remain as starting techs. That way, I assume they will not be effected by those mega events if they should come up.

Just wanted to say that I wholeheartedly concur with the suggestion that putting the Drengin on AI 8 will help them out. I tried switching them from AI 11 today, and they're a monster - ranked #2 in the game, but really #1 in just about everything but economy. They actually haven't really started a major war - just bullied minor races and joined the swarm of opportunists descending on the Korx after the Yor beat them up. Recently someone paid off most of the galaxy to attack them, so we'll see how that goes.

Also agree that AI 8 is militaristic in general - there's three others and they're all fighting at least two wars. Then again, the galaxy is a violent place right now - my Altarians are lucky to be fighting just one major race as of the moment. At any rate, things look good in terms of balance, and it's rather satisfying to see that the Thalans are losing - they've dominated the last few games. Right now it looks like a toss-up between the Drengin, Yor, and my alliance with the Iconians for the lead. The Torians are big right now, but they've picked on too many people and it's come back to bite them.

So I played out the next four years, which turned out to be the last four years.  With the advanced races getting to the end of their tech trees, the last few months of year eight turned into a race for the tech victory.  The Drath won, beating out the Yor by 1 week, with the Iconians not far behind.

• AI 8 really doesn't like Extreme Colonization (Colonization category techs).  By game end, the Terrans still had not researched it.  The AI 8 strategy really seems to be conquer, not settle.  It works, to a certain extent.  The Terrans were a major power, the Iconians were leading for most of the game, and the Drengin started out strong.
  • Not sure what I want to do with this.  Do I try to force it, or just accept it as the AI 8 way?

• AI 8 also doesn't like Manufacturing techs a whole lot.  The Drengin started out strong, but bit off more than they could chew.  Over the long haul they declined while their enemies grew.  Turns out they never researched Slave Canyons - not until I made them do it near the end.  That means they were stuck with Slave Pits, with a lousy 5mp each.  They simply got out-produced, even when they were down to only 1 enemy.
  • I had to dial down the AIValue because AI 11 would research it way too early.  I'm going to make a separate tech tree for Drengin at AI 8 and tweak it.
• Korx got crushed.  I don't know exactly what it is, but they never keep up.  My guess is that they are researching their unique techs too soon.  They spend the first year or so on that and don't get a chance to upgrade their infrastructure before the arms race.  Being stuck with Traditional Factories and Labs in year 4 and 5 is a death sentence.
• Looked at the Mining Matrix in the xml.  Didn't see anything overtly wrong.  There is a lot of junk code throughout the xml, like ModuleRequirement and TechRequirement tags and half-tags with no entries.  I cleaned it up and will manufacture a test to see if it helps at all.

I'm satisfied so far.  Next time I'm going to drop the four trees I'm happiest with and include the four I haven't looked at yet.  I think the Drath, Iconian, Terran, and Yor will go to make room for the Altarian, Korath, Krynn, and Thalan.

I'm not sure what the big problem with the Korx is, but they've been steady at the bottom of the pack. The idea that their unique techs are behind it might have some basis; they're the only race that's doing worse now then they did before I starting modding things. I have a couple of custom races with the Korx tech tree and they've done okay, though, so maybe it's something else.

Haven't really run into this behavior before. My Drengin on AI 8 tend to be pretty quick about researching Slave Canyons - sometimes, they get it before the arms race even starts.

Personally, I'd just go with it - it's good that the AIs act in visibly different ways. Then again, it may depend on circumstances - you seem to use higher extreme planet counts then I usually do.

In other news, I've finally gotten the Arceans to pull themselves together. They'd been having issues for sometime, but they look to have been partially setup based - right now, they're middle of the road. The first run of the new Korath tree is looking good too.

I'm not sure what the big problem with the Korx is, but they've been steady at the bottom of the pack. The idea that their unique techs are behind it might have some basis; they're the only race that's doing worse now then they did before I starting modding things. I have a couple of custom races with the Korx tech tree and they've done okay, though, so maybe it's something else.

I'm thinking it's a combination of their unique techs and the structure of their tech tree.  Even after their unique techs are done, they seem to get distracted by various techs and never really hit the Artificial Gravity->Planetary Improvements->Research/Manufacturing/Economic Improvements line.  I think they may need a little race-specific TechTree tweaking to get them researching those techs faster.

Personally, I'd just go with it - it's good that the AIs act in visibly different ways. Then again, it may depend on circumstances - you seem to use higher extreme planet counts then I usually do.

Think I agree on this one.  Just let it play out.

Couple more interesting notes:

• Anyone else think it's kinda odd that the Drengin have unique research structures and common economic structures while the Korath have unique economic structures and common research structures?  Think I'm gonna combine the two so they both share the unique techs.  Why the difference?
• There is a tech in the TechTree xml with an interesting category - MilManufacturingitary.  I think you can guess what happened.  I forget which one it is, but you could search for it.  Maybe it's in the Drengin tree, but I don't think so.

Edit to add - In a first test it looks to me that a big factor in Thalan dominance is their tech tree.  With my identical custom races, the one with the Thalan tech tree is just exploding.  I remember now that I forgot to set them as Super Hive, though.  I think they may be Super Spy, or maybe Organizer.

There is a tech in the TechTree xml with an interesting category - MilManufacturingitary.  I think you can guess what happened.  I forget which one it is, but you could search for it.  Maybe it's in the Drengin tree, but I don't think so.

 

It is the Drengin/Korath tech tree, specifically Artificial Slavelings.

Thanks for that.  I forgot to follow up on it.  Since I just fixed the Korath TechTree last night, I guess I should go back and fix that before I test it.

I took care of the Thalans, and I also cleaned up the Krynn tree, which was full of errors.  The change log is almost as long as the Iconians.  A little more work on the Altarians and Korx and it's on to round 2.  Of this set of rounds.

Let's see . . . 5 Iconian tests, 6 diplomacy tests, 7 AI tests, and 1 tech test so far.  19 playtests and counting.

Tolmekian, what galaxy settings do you play on? That could be useful to know if the effectiveness of different AI strategies interacts with things like galaxy size, which I suspect it does. There seems to be a consensus that AI 7 and 10 are not even competitive. But I've gotten both the Drengin and Arceans to win games via military conquest of the entire galaxy using AI 7 combined with a high aggression settings, and I've also seen the Torians and Iconians do very well using the supposedly broken AI 10 on occasion (following tech tree fixes and some aivalue rebalancing of my own). However, all these games I've played were on Large maps. Maybe it's different on Huge or Gigantic maps; I don't know because I've never had the patience to play on maps that size.

Has anyone else tried re-balancing the ai values in the planetaryimprovements.xml? There's plenty there to question if you start thinking about it. For starters, I changed the aivalue of the embassy from 1 to 5. I don't know if the ai does better now, but they do build more embassies on those influence tiles.

I, for one, have balanced the hell out of almost every single able file that I dare to edit. Although, I did miss out on the embassies. Thanks for the heads up!

I've done a little work in the PlanetImprovements.xml, but nothing major.  It looks to me that the AI does a fairly decent job as it is.  One big change was the AI for the Biosphere Modulator.  It was 0, so it was never built.  I changed it to 155 - the same as Weather Control Zeniths, since they do the same thing. Now they build one on every planet, as they should.  I also changed the Iconian Molecular Fabricator from 0 to 5.  Since almost all of their improvements are 5, they build it along with everything else.  The only other thing I can think of off hand might be the Krynn Consulate.  They should probably put one of those on every planet.

As for galaxy settings, I always play on larger galaxies.  My tests have all been using the same settings.  Probably easiest just to show you:

  The only exception would be that for the majority of my testing I turned off Tech Trading, Tech Stealing, and Super Abilities.  I turned them back on for my final testing and polishing.

AI 7 won't colonize anything outside of its influence.  Under my conditions that means they end up with a terminally small empire.  With abundant planets and habitables along with tight clusters, you could probably get the planets close enough together that they could establish a sufficient empire to begin conquest.

AI 10 has a problem with leaving planets undeveloped.  I don't know what conditions could allow them to thrive, but I can't say that it would never happen.  They expand fine and research well, as far as I can tell, but without any improvements on their planets they totally lack staying power in the mid-to-late game.

And, as a final note, After my last test I'm pretty much ready to declare my work done.  For real this time.  I'd really like to actually play a game now.

Edit to add: After the last test I've changed my opinion on the Drengin and Korath trees.  It's not just odd about their econ and research techs/structures. It's bad/broken.  Since they both start with A Ruthless History and Specialized Research, it ends up that the Drengin can't upgrade their Black Markets and the Korath can't upgrade their Slaveling Labs.  TOV and I went back and forth over how to get Advanced Market Centers to upgrade the Black Market.  Now I just say fix the tech trees and be done with it.  It works nicely.

I laughed at this because I always play the Yor, and I usually take a long pause at CM III.  Seems like that's the point where my economy can just barely handle the production, and plenty of ships are rolling out, no need to go further. 

Another post earlier mentioned Korath doing just fine at AI 7, not sure what that is, but at "obscene" level, they do just fine indeed.  I'm ready to wipe them out, of course, but I also had the fortune to have Creativity kick in to make Black Hole Guns a one-turn research project.  Straying off topic, sorry...

Wow, just found this thread; lots of good info here!

It seems to me that at least some of the issues discussed here clearly qualify as bugs, such as places where techs were obviously miscategorized, or where there were missing or mis-spelled values in the xml files.

I think it'd be very handy to get a full list of such issues (as opposed to things like research priority tweaks, which are a bit more subjective).  Would one of you be willing to write up such a list and post it to my bugs thread?  With luck, those fixes, at least, could be easily incorporated into a patch.

Does this mod(patch really from the read) address the issue of the Torrians simply not builing anything at all on any of their worlds?

thnx

Unfortunately, no.  The Iconians have the same problem when set to their default AI, which is AIPersonality 10.  As I mentioned earlier, AIP 10 has a bug that prevents development.  It's a problem with the AI and nothing we can fix.  You can, however, change the personality that the Torians and Iconians use.  Generic would work fine for both.  Try Thalan if you want either to be more militaristic.

That's no small task, though I'll see if I can get some time.  I've finished with the TechTrees and moved on to the PlanetImprovements xml.  Did you know that there are three Galactic Achievements that aren't in the game because the required tech is misspelled?  I vaguely remember them from DA many years ago, but they got lost in the update.

The Galactic Monument is StarFederation - should be StarFed

Galactic Stock Exchange is AdvancedTrade - should be Advanced Trade

Hyper Distribution Center is CulturalConquest - should be Cultural Conquest

On top of that, the Temple of Neutrality has the same alignment numbers as the Temple of Righteousness.  It's pretty much a cut and paste with only the name changed.

Anyway, I'm on the verge of releasing my work as a mod.  I figure all the fixes with and without the colonizing minors.  Now I just need to figure out how to package up a mod.

I think that the ideal thing would be to have two mods:  One to fix things that are clearly bugs like misspellings and miscategorizations, and another to fix gameplay adjustments like AI priorities.  The reason I say this is that Stardock could conceivably take a bugfix mod and just patch the base files with it without needing to do a lot of testing.  But anything that messes with the AI configuration is probably going to need a lot more review.  Making it two mods makes it more likely that at least the bugfixes will get merged into the base.

I agree that a bugfix-only mod is a good idea.  But the other mod should be the full works, otherwise your average player has to sit and merge XML files by eye.

I wasn't sure if the mod system allowed chaining mods/having multiple mods active.  If it doesn't, then I agree; best to have the second be the whole thing.

 You guys all rock...looking forward with great anticipation to playing the tech tree mod once released.  I stopped playing the game several years ago - (only in small part due to perceived shortcomings in ToA which always seemed IMO to be unfinished - more to do with unrelated life stresses), and just on a serendipitous whim decided to vist the forums again after a long absense and lo and behold found this effort!         This mod (edit: and I didn't mean to forget the original topic Weapons fix mod too!) when released looks like it may breathe some new life into an old friend - many thanks for the time and effort you all have spent!