SMAX - The Will to Power - mod

[bvanevery]

How do I find you cookbook among thousands of pages?

There are search functions.  I imagine you can narrow things down by my name.  You could simply read back a few years until you see it, as there aren't that many posts in Theory of Everything in recent years.  It's also possible to look at a user's profile and see all the stuff they ever posted.

You could also just hack it out with the 2 sentences I told you.  Open the Scenario Editor and...

[bvanevery]

Details not so mineralsy

not so mineralsy

MY 2184 as the Spartans.  Well the AI certainly builds Recycling Tanks now.  Somehow with Discover Conquer research foci, I didn't get Biogenetics and they did.  This year my allies the Gaians completed the Human Genome Project.  I went on a begging and trading spree with my 4 allies, which got me Biogenetics and a bunch of other things.  Definitely tech leakage, although one could argue they're merely rubber banding me.  I'm not awesome.  My start was minerals poor and a bit restricted on land spread, due to fungus and fungal towers in the way.  I have only 1 "good" city with 8 minerals, 2 with 5 minerals, and 7 other paltry cities.

I've got Rec Commons completed in most of them, but there's no way that starting a SP would have been viable.  The AIs totally got the jump on me this time.  I was building Foil Probe Teams to go steal from Zhakarov my one enemy, when the Gaian SP completion gave me an easier path.

I did have a good cash stockpile, so I'm rushing 2 Recycling Tanks this turn.  Won't have quite enough next turn for the 3rd, but it won't take long.  After that, I don't actually have any "minerals worthy" cities yet, so I'll be needing to grow.  I've started Children's Creches in the various puny places, and shortly I'll go Democratic as it's non-controversial.  Before that I've got an interloping spore launcher to kill though.

[bvanevery]

Details done with this

done with this

MY 2263.  This isn't any fun.  I've had no productivity the whole game.  I barely got my small nation garrisoned with Trance Plasma units, and finally Hologram Theaters and Genejack Factories built, which merely catches up to where AI factions have been at ages ago.  They seemed to have no problem making massive amounts of money to maintain their GFs, whereas I get no money and am seriously penalized by the high maintenance.  I could understand in the case of Lal because he was Governor and getting extra income, but why Deirdre should be making a lot of money, is a mystery to me.

Former allies mostly turned on me, due to me picking SE choices necessary for catching up.  The Cult turning on me is particularly irritating, because I am Green.  I had both the Cult and the Gaians as neighbors and as one might expect form the hard coding of the stock binary, they got into a war with each other.  Nice political choice wasn't enough to prevent the Cult from threatening me as I'm so far behind.  And unlike the Morganites, who really weren't in shape to send anything my way before my GFs and Tree Farms complete, the Cult has lots of powerful units ready to go.

This game felt like it was determined by 1) getting a mediocre start, with not many minerals, not much food, and not that much fungus free land, and 2) the accident of getting Biogenetics later than everyone else.  It seems unreasonably difficult to research it with a Discover Conquer research focus, and seems to say that picking Discover is necessary to prevent one summarily losing the game.

I think I'll actually try the Spartans again, because it is possible that they can't produce for crap.  Not just the INDUSTRY penalty but the new EFFIC penalty makes it difficult for them to afford anything.  I don't think I got in a fight with any other faction the whole game.  It was that painful just to garrison with basic units.  I'm thinking that going Fundamentalist early and stomping someone might be the only way to play the Spartans.  Well at least given a poverty start.

[bvanevery]

Details improved empire

improved empire

Owing to a much better minerals, food, and roomier start, I was able to build the Merchant Exchange and the Human Genome Project.  Now though I'm in a thing where all the AIs have beat every single SP I've been trying to build in my best city.  As of MY 2241 this is tiresome and I'm quitting.  3:45 AM has a lot to do with it too.  I had to beeline with Discover to get Recycling Tanks fairly early.  Even when I only had 3 cities, the Data Angels had already built 2 Recycling Tanks.  That's a bit ridiculous and feels like advantage goes to whoever gets a good start.

As the game progresses, the production of the AI factions feels pretty much impossible to compete with.  In this game, the Spartans also didn't have a good option for stomping anyone "early".  I also realized that Fundamentalist doesn't exactly come early anyways.

I really, really don't think a Minerals Multiplier should be so easily obtained by anyone at the beginning of the game.  It makes the game much more capricious as to who got early advantages.

Similarly, allowing the AI to almost immediately get ahold of Genejack Factories just after it got Recycling Tanks, is not exactly fair to the human player.  The AIs don't seem to suffer anything remotely like the maintenance and happiness penalties that I do.  I think these 2 facilities alone, turn the game into "watch AIs win and you can't catch up".

[Alpha Centauri Bear]

Juffos found an interesting thing in vanilla.
Conventional artillery bombarding mind worms initiates a conventional combat. This is an only exception from *all other* cases when psi unit is involved (direct attack, spore launcher bombardment, artillery duel with spore launcher on either side). As a consequence, conventional artillery with high enough weapon value delivers huge damage to mind worms as they got some virtual armor value = 1 assigned for such conventional combat.

I am not sure what to do with that. Any suggestions? Here is what pops up to my mind.

• Leave it as is and use as feature.
• Force it to be a psi combat thus reducing bombardment damage to some small number as if in equal weapon/armor conventional bombardment.

[bvanevery]

Spore launchers inflict psi damage because they are psi things.  Perhaps the spores do something horrible in the air just before they land.  You can head canon that any way you want.

Mindworms are not artillery pieces.  They are targets.  Mindworms can't do anything to you at a distance.  It would be smart to fight them at a distance.

One could reasonably question why an air attack on a ground unit wouldn't also be "fighting at a distance" though.  Do mindworms suddenly become scary when you're strafing them?  If you could drop bombs on them, why would they be able to do anything back?  The game doesn't quite implement bombing proper.  I think I've tried making a flying artillery piece, but I can't remember if it worked or was actually useful for anything.

The thing about your mod is, do you actually have any real global warming?  Or real punishing behavior for transgressing with illegal chemical attacks?  I've done After Action Reports of that sort of thing.  No Survivors, No Secrets is one in which I prevailed.  Artillery is useful when facing many stacks of 32 or 64 mindworms or Isles.  Need the Air Superiority version to shoot at similarly huge stacks of Locusts.

I've got the original game's death scenario, so there's no way I'd consider getting rid of this.

[Alpha Centauri Bear]

I didn't change global warming formula. The only related change is raising initial clean minerals from 16 to 24 saving faction 8 pops.

As for psi/conventional attack we can talk it all over comparing to real life. Although, I am skeptical anyone encountered mind worms in real life. That is largely irrelevant as whole concept of psi combat is quite controversial. It is introduced there as some other combat mechanics alternative to spice the game and not just letting strongest weapon prevail. That is fine, but having a single and not symmetrical exception from the whole case is bad play wise. Same way you can argue that spore launcher in a duel may inflict psi damage but should receive conventional damage as it is just bombarded with mindless shells. There is no point in psi damage for ranged attack at all. Yet it is there.

I am inclined to remove this exception to make rules more straightforward and simple. Indeed, nobody ever reported that problem before. People just largely don't care.

[bvanevery]

I didn't change global warming formula.

Then you have inherited Thinker mod's nerfing of it.

Although, I am skeptical anyone encountered mind worms in real life.

We have a secret project video and extensive verbal description of mindworm combat.  What these things do, is no kind of mystery in the game's lore at all.

having a single and not symmetrical exception from the whole case is bad play wise.

Taking issue with psi combat as a whole is one thing.  Taking issue with whether conventional artillery attacks make sense against mindworms is another.  I don't see what's hard to understand about the latter, lore-wise.  Have you not watched secret project videos in a long time?

Same way you can argue that spore launcher in a duel may inflict psi damage but should receive conventional damage as it is just bombarded with mindless shells.

And armor and AAA shouldn't magically work against Needlejets.  But it does.  At some point you have to accept the game mechanics on their own terms, unless you're prepared to implement a completely new combat system.

Indeed, nobody ever reported that problem before. People just largely don't care.

I think you should add real global warming back into your mod first.

[bvanevery]

Ok I had more time to think about this while walking the dog.  There are a few separate issues.

A mindworm can't behave like an artillery piece on defense from an artillery attack.  The idea is ridiculous.  If you were shelling a base, for instance, and they had 1 mindworm as a defender amongst a dozen other conventional defenders, then suddenly you'd be in an artillery duel.  Players can't have artillery defense like that.  You have to either build some kind of artillery piece to fend off artillery attacks, or you have to take the damage.  Anything else is just getting silly with the rules.

So, given that a mindworm must take damage from an artillery attack, the 2 issues are:

• How much damage will the mindworm take.
• What sound will play during the combat.

That's it.  Well you could quibble about text descriptions and animations to go along with these 2 facts, but these are the essential facts of the problem.

In the original game, mindworms are "glass cannons".  They do a lot of offensive damage and have very little defense.  The bigger the caliber of artillery used against them, the more damage is done.  I can't remember if a big artillery piece can actually destroy mindworms in a stack outright?  There will always be 1 defender left alive but I can't remember if many of the rest can be shelled to oblivion.  It can definitely destroy Isles or Locusts, presuming you made it have Air Superiority for the latter.

In your mod you'd want to do something more consistent with your concept of mindworm offense and defense.  But... artillery is supposed to damage them, one way.  That's not negotiable, per the rules everyone expects of the game.

[Alpha Centauri Bear]

I never said mind worm should behave like artillery. I meant to calculate their damage based on psi combat not conventional one.

[bvanevery]

From an audio standpoint it's a bad idea.  Just because there's a mindworm in a stack of otherwise conventional units, doesn't mean you should be playing mindworm sounds and animations.  Those conventional units should be getting hit by the sound of the artillery barrage.

I think you should be readjusting how a mindworm takes conventional damage.

Bear in mind that they already tend to have a "superweapon" problem.  If you make them basically have this great armor, you could be throwing the game balance off in the name of "rules consistency".

[Alpha Centauri Bear]

Responding to Juffos feedback

AI is doing much better now, I didn't see unprotected bases even with (rare) supply pods enabled.

Neat.

Formers are making roads and getting more things done than earlier, but I can notice them being unable to decide what improvements to keep - mines to solars and back to mines. Miriam has a former planting fungus on a rocky mine near New Jerusalem while the square is being worked by the base.

As with any other AI in the game they don't have long term plan. Instead they just do what they think is needed right now and that may change turn over turn. Occasionally, they may find replacing improvement is better than building new one. I didn't specifically tune them not to. Algorithm may be not that deeply sophisticated. I am not building another Deep Blue here. As long as they do right things 90% of the time they are still ahead of human player with their 30% production-growth bonus.

I am playing with 1.25 mineral weight in thinker.ini to see what effects it would have.

You mean this: ai_terraforming_mineralWeight=1.50? Sure try it out different combinations. Keep in mind, though, that these weights are not same as resource values! These are just priorities. There are multiple factors at play one of which is terraforming time. Mineral improvements being longest to do make them less desirable to be built. That's why in my latest versions I set mineral weight twice bigger than nutrient to encourage formers to build more mine improvements.

Keep also in mind that AI does not think in patterns like farm-solar-road. It just build what brings more value right now. So it may be that placing farms first is more effective strategy. I don't explicitly stop them from doing anything unless they do obviously stupid things all the time. Occasional stupid things is fine.

I find it odd that AI terraforming performance took such a huge nosedive in WtP. If memory serves, it was much stronger in Thinker. Maybe making the AI use the farm+solar+road and farm+mine+road orders instead of single improvements could help while forbidding the replacement of existing improvements? Exceptions for replacing forest spreads or drilling boreholes included, of course.

I beg you pardon. How did you come to this conclusion?  😁

Induktio in Thinker directs his formers to do what he wanted them to do. For example, plant more forests as he believed forest to be quite productive and underappreciated by vanilla AI. That do make sense. However, this approach is somewhat brittle when game rules change.

WTP uses regular industry artificial intelligence approach. I.e. for each action it calculates the result, evaluate this result according to corresponding resource weights, adds all up and prioritizes action based on these results. It takes everything into account: different resource weights, current base needs, terraforming time, travel time, and many others. This approach is absolutely flexible and rule agnostic. One can change game rules as I did quite often in the past (like farm output, borehole output, terraforming times, etc.) but the algorithms continues to work without special tweaking for these specific rules. Which is a huge time saver and element of overall mod stability.

Now. I cannot say with utmost assurance whether Thinker or WTP does better in terraforming (under WTP rules!). I didn't compare them too deeply. However, this can be done easily. I can add a parameter to switch off WTP terraforming for say last three factions and see whether they win/lose/wash on regular basis. Would require some 10-20 test game. If you or anyone else is up to it - I'll introduce it.

Another thing I would love to see would be the turn resolution order change in Yitzi's patch, where bases grow before the governor adjusts PSYCH specialists for drones. Losing a production turn to base growth is incredibly frustrating and manually checking and adjusting the cities is tedious.
I know about how Yitzi disappeared and left no documentation, but hey, a man can dream 

What is that? Is that AI thing or overall game mechanics applicable to all?

[Alpha Centauri Bear]

Here you go, actually.
https://github.com/tnevolin/thinker-doer/releases/tag/276

Knock yourself out and test compare them.

[Alpha Centauri Bear]

I have tested it myself a little and perceive Thinker and WTP terraforming algorithms comparable. Neither one do a lot of stupid things. Some action are questionable but cannot be definitely proven wrong. In the end bases receives more or less optimal yield.

[MercantileInterest]

I've been tinkering with the Cyborgs: 

 TECH, InfNet, IMPUNITY, Cybernetic, PSI, -15, TECHSHARE, 1, MINDCONTROL, 1, TECHSTEAL, 1, SOCIAL, --PROBE, UNIT, 2, UNIT, 5, SHARETECH, 1,
  Society, Cybernetic, RESEARCH
  Politics, Fundamentalist, nil

The goal, of course, is to give them a more distinct playstyle than vanilla, since University is already the RESEARCH faction.

In this iteration, their main strategy is to infiltrate everyone, as any discoveries by infiltrated factions automatically get copied by the Cyborgs. This is very powerful, so I've tempered it by reducing their PROBE value, which means their infiltration devices will be more rapidly discovered. They're also immune to bribery.

Along with negative PROBE, they also suffer from -15% PSI combat. With WtP's intense mindworms, this is a substantial obstacle.

With this setup, the Cyborgs benefit immensely from the Empath Guild. Without that, size of the map makes a big difference to playstyle, since you have to get your probes to rival bases.

This sort of probe heavy playstyle infringes a little on the Data Angels  so my old post on how they might work applies:

Faction Stats:

• Founding base is not HQ, may not build HQ, does not acquire HQ upon conquest
• Faction does not experience drone riots, instead drones may not work tiles
• Enemy factions may never in any way infiltrate the Data Angels, other probe operations will function as normal
• Rival bases within 4 tiles of Data Angel territory gain +1 drone
• -1 Support