SMACX difficulty/balance txt mod

[dino]

@Moderators
Please don't make a news about these txt edits, it's work in progres.
If after some playtesting and eventual positive feedback, I'll ever find it ready and worthy advertising otside this subforum, I'll kindly ask for it.

I'm toying with txt mods to find the right amount of changes, that would allow AI to hold their ground against experienced SMACer.
The goal is to make a minimal amount of changes, that have the most impact, so the game is so similar, that does not require any learning from experienced player.
Original AI is so incompetent in base/terraforming development, that it was impossible without breaking the game into complete mess, if at all.
With Inductio's thinker mod it's worth trying again.

AI buffs:
++GROWTH     - ai will popboom with creeche and planned, or democracy and it helps with early game land grab,
+INDUSTRY    - just some extra more turn advantage,
NODRONES     - ai is not aware of that and still spent resources in psych and psych buidings, but is not thrown into disarray by pacifism drones, or when the bases grow too big,
INTEREST 3%  - player always invest the last credit, ai tends to accumulate many credits, now they work for them too,
TERRAFORMING -  double speed, 1 free former at the start helps with eventual bad initial placement,

Crawlers:
Supply Crawler module cost factor increased to 16 (5 rows), this changes ROI to respectable 4%/turn on forest and 8%/turn on rocky mine, which is more in line with what AI can do at this stage.
Originally crawler on forest had 6,5% ROI and 13% ROI on rocky mine, much better than anything else that can be done after you run out of land to put new bases on.
You can't make AI do the same and spam crawlers, because it's incapable of determinig when their crawler park is safe and defending it, or hiding it in a base.
It also leads to more fun early/mid game that doesn't consist mostly of producing and managing an army of crawlers.

Satellites:
Now we've reached a real elephant in the room, a completely broken SATELLITES, with 20 bases their ROI is:
8-16,5% raw,
20-25% with base facilities most factions will have, when they start pumping them,
30-40% with all base facilities
some more with The Space Elevator

It means that once a faction start pumping them, it'll double its total resource output every 3, or 2 turns depending on the amount of resource bonus facilities ( even faster with space elevator ).
Rapid exponential growth make it discover tech every turn. Human player can even disband most units and formers, to kickstart the acceleration even further.
If faction has more bases, ROI is even more absurd. The game ends right here and nothing else matters, there is basically no endgame stage, it ends right after it starts.

I've increased a cost of resource satellites from 12 to 30/30/40 rows for hydroponics/power/mining
With 20 bases they still give a really great ROI of 5-12,5% ( more with elevator ), usually over 8% at the stage they become available.
It's slightly worse ROI than crawler on a mine at this stage.

There is still an issue of providing more benefit to already leading factions and scalling poorly with map size.
Here comes the first feature request for Inductio: make each satellite produce just 10 resources that adds to a pool evenly distributed to bases.

I guess there is a list of faction bases, you could iterate on, to distribute all resources from the pool evenly, to the base size limit.
Or if it's too difficult for some reason, create a table with each base size, sort it from smallest and iterate on this table.
Find the base that got the most resources and use this value-1, but no less than 1, instead of satellites count, to increase bases output.
I know you are not interested in gameplay changes, but satellites in current form are seriously broken.

The second feature request is an option to read AI faction rules from a separate _ai.txt file.
I have a nice work around in my mod, to allow that without replacing txt files each time I want to play a different faction, but it limits amount of factions available on selection screen.
Third is an option to disable expansion features and use SMAC interface textures ( for that I replace files with renamed originals ), preferably separate setting for each feature.

Other changes:
Small rebalance of early weapon/armor values, 4 movement points copters, ---POLICE Free Market, --POLICE Democracy, worldbuilder Inductio posted.

Alien tech:
I've moved Aquafarm to Doctrine Flexibility, Spore launcher to Centauri Meditation, sporific gas pods to Centauri Genetics.
Made disabled User tech a prerequisite to Progenitor Psych, Centauri Ecology prerequisite to Adaptive Economics and Secrets of Creation a prereqisite for Transcendence like in SMAC.
Result ? If there is no aliens in your game, or you have not met them yet, the Progenitor/Resonance techs will not show up.
To start this optional tech tree branch you have to trade their tech, or steal it with probe team.
It's even lore fitting, that humans can't start talking to aliens on their own, without an extra effort ( probe team action to study them and discover Progenitor Psych )

Faction selection instructions:
Click Original Seven button to reset, pick a faction, expand a list with custom factions and pick a version of selected faction with a mark.
There are only original seven and aliens available, no mark mean a buffed faction for AI to play with, marked faction is a normal faction for player to choose.

The ideal aim of this is to make an experienced SMACer loose most of the time on TRANSCEND, but to be winable with decent start, by employing extreme autism.
I'll be balancing it according to my playtests, thinker mod progress and eventual feedback.

I recommend to disable economy victory, use tech stagnation and no spoils of war. The latest development version of thinker mod recommended.

I intent it more as a discussion, and ideas to pick from, than anything more, but if anyone wants to try my SMAC config here it is:
http://www.mediafire.com/file/vclvbw29pcztzdj/SMAC_difficulty_balance_v1.zip/file

[dino]

Some more theorycrafting about huge changes in the mineral cost of overpowered features.
Even after the changes they're still the best stuff to build once they're available, just not grow 10x in power in 10 turns best.
So if AI is coded to use them it'll still be a good choice.

I've read some interviews and postmortems by Sid and other Firaxis designers.
You see, 99% of actual customers never reach the understanding of the game systems we have, we are not a target audience.
These games are, in Sid's words, designed to be played by teenagers with their dads over the shoulder.
Their philosophy is to not be concerned by balance, or versimilitude at all, but provide a simple wow factor: +50% this, +100% that.
It was super cool to reach satellites and understand a degree to which they can be abused... like the first three times.
Later I've got bored with buiding a base facility each turn in each base, that also grew in size each turn and develop a tech each turn.
I've started to just quit a game once I've reached them. So, I don't believe we should hold all original design choices as sacred.
This change will extend late game in time greatly and give you some time, to actually play with all the toys you are discovering along the way.

I also disagree with moving overpowered stuff upward tech tree. It'll still be overpowered and it's way better to make "super cool" toys just "cool",
than to deprive a player cool toys he is used to, until a very late into the game.

Humans will also always be better at micro, so if we want to be challenged AIs have to enter midgame with more bases.
If it's taught to abuse satellites the way we do, the game will always be lost the moment it reach them with any headstart, same way it's currently won by getting them.
This is why I am strongly convinced they have to be changed.

[Induktio]

Crawlers:
Supply Crawler module cost factor increased to 8, this changes ROI to respectable 2,5%/turn on forest and 5%/turn on rocky mine, which is more in line with what AI can do at this stage.

Hmm, I thought 8 was the default cost for that module. If you don't want to deal with them in the early game, why not just edit the tech requirements?

I guess there is a list of faction bases, you could iterate on, to distribute all resources from the pool evenly, to the base size limit.
Or if it's too difficult for some reason, create a table with each base size, sort it from smallest and iterate on this table.
Find the base that got the most resources and use this value-1, but no less than 1, instead of satellites count, to increase bases output.
I know you are not interested in gameplay changes, but satellites in current form are seriously broken.

This change could potentially bring glitches if it starts modifying the base production code to a large degree (which it probably would have to do to accomplish that.) I'd suggest try to find a way with alphax editing first.

The second feature request is an option to read AI faction rules from a separate _ai.txt file.
I have a nice work around in my mod, to allow that without replacing txt files each time I want to play a different faction, but it limits amount of factions available on selection screen.

In principle it would be good to have an option to set these factions bonuses specifically for the AI factions. Keeping separate txt files for AI factions is pretty annoying. Maybe these bonuses could be just set from thinker.ini. Looks like doable.

Third is an option to disable expansion features and use SMAC interface textures ( for that I replace files with renamed originals ), preferably separate setting for each feature.

Well, this is a feature that can be implemented very easily. Earlier I actually noticed the engine has a _single flag_ to set whether all the expansion related items are present in the game. Just by disabling this flag at runtime, all the expansion related features are gone. Even interface textures will revert to vanilla versions, all the expansion techs/units will not be available etc. Looks like only original 7 factions can be selected though. This should be implemented in the next testing build.

[dino]

Hmm, I thought 8 was the default cost for that module. If you don't want to deal with them in the early game, why not just edit the tech requirements?

I like them, I want them early, I don't want them so good that rushing a swarm of them to put on forest is better than most of other early development strategies.

I haven't played in ages and was recreating my mod from memory, so details like default crawler cost escaped me when I was writing the post.
Yes, I have cost factor at 16, also I was looking at unit formula when calculating the roi, but there is some modifier to crawler cost that I haven't figured out.
Original is actually 3 rows and modded one is 5 rows, so the mod is a bit more tame than I've described, and unmodded crawler is even stronger, I've updated description.

[bvanevery]

I'm confused by this claim:

Result ? If there is no aliens in your game, or you have not met them yet, the Progenitor/Resonance techs will not show up.

In SMACX, I've never noticed any phenomenon of Progenitor Psych, Field Modulation, Bioadaptive Resonance, Sentient Resonance, or Secrets of the Manifolds disappearing from the tech tree.  Are you talking about some phenomenon in straight SMAC terran.exe?  Are you talking about some Induktio specific feature?  I really don't get what you're saying here.

Actually, I think I may understand now.  You're saying in your mod, nobody can research Progenitor Psych and Field Modulation.  They can only be obtained by trading with the Aliens.  So if there are no Aliens, then anything that has PrPsych or FldMod as prereqs, cannot be researched at all.  You have decided to make a branch of the tech tree, that is inaccessible except by trading with the Aliens.  Intersting concept, ok whatever.  I suppose Factions could be given exclusive starting techs in the same way, if one was so inclined.

You see, 99% of actual customers never reach the understanding of the game systems we have, we are not a target audience.
These games are, in Sid's words, designed to be played by teenagers with their dads over the shoulder.

When was the article written?  I really hope there's a larger number of "hardcore" masters of 4X TBS intricacies nowadays, or my own indie developer plans are going to be rougher than I thought.

I also disagree with moving overpowered stuff upward tech tree. It'll still be overpowered

That depends on how overpowered something is.  For instance, I don't think it's unacceptable to own the Planetary Energy Grid under any circumstances, but IMO it's clearly too powerful for the point at which it appears in the game, and its cost.   So in my mod I moved it later and made it cost more.  AFAIAC that solved the problem.  I feel I solved all the overpowered Secret Project problems this way.  Of course, the cost in terms of modder time was enormous!  Totally uprooted the tech tree to get it done.  Months of development.

Satellites could be solved by simply taking them out of the game.  I didn't want to do that in my mod, because I really didn't have removing parts of the game as a goal.  I have thought about making a "stripped down" tech tree mod, reducing SMAC to a more "playable essence".  The problem with such a mod is twofold: 1) after 4 months of person hours spread over 6 months of mental monopolization, I am burned out with working on SMAC mods.  I'm definitely in maintenance / serious fix only mode at this point.  2) making a 2nd mod would compete with the prolific mod I just spent all those months making.  It's not in my best interest to spread the player base even thinner than it already is.

If playtesters come up with something that makes satellites no longer a game breaker, I'd be open to putting that in my mod.  But for me it would have to be a .txt only mod, such as increasing their cost.  I already pushed them later in the tech tree.  It's the sort of issue where I have to hear that playtesters want it, or even care.  I'm ok with some things in the late tech tree being goofy, because I personally don't use them much.  I think putting up a bunch of satellites is far more of a chore than a thrill.  Same as micromanaging citizens.  If people want to sandbox the hell out of that stuff, who am I to stop them?  That said, I did banish the Transcendi to the very end of the tech tree, along with the Telepathic Matrix.  Sure you can sandbox... when you've finished the game.

For really stripping down the game, my personal energies are going into a new 4X TBS title.  I'm not going make content indefinitely for SMAC, that was never my goal.  My goal was to prototype the difficulties of tech trees etc.  At this point I've certainly done that.  My conclusion from all these months of work, is that SMAC's tech tree is at least twice as long as it needs to be.  I mean from a modern face-to-face board game usability perspective, this tree is junk.

[bvanevery]

I like them, I want them early,

Supply Crawlers are game breakers.  You are experiencing the tension of actually having fun getting a huge economic advantage in the game, vs. the reality that it makes the rest of the game a cakewalk.  This is a problem in other genres as well.  Say a dungeon crawling RPG.  You probably like getting the rare item that enables you to really kick ass.  Even though it may turn the rest of the game into a foregone conclusion.  Power Fantasy and Challenge are contradictory game design goals.  Put another way: I want to get rich quick vs. I want to be made to earn everything I do by the sweat of my brow.

I started out putting SCs very late in my mod.  Eventually I decided the game had become too dry though, like eating tuna salad without enough mayonnaise in it.  So I moved them a bit earlier, but still a lot later than the unmodded game.  Then recently in version 1.21 I moved a lot of things earlier, like Cruisers and Hovertanks and Supply Crawlers and Genejack Factories and yadda yadda.  SCs are still later than in the unmodded game, but maybe only 1 generation of tech later now.  I may have broken something doing this.  But in my own playtesting, I was just getting too jaded to care.  The game felt dry and I wanted the players to have more options.  Why should it always be about getting Needlejets and then everyone else just dies?  I've abdicated responsibility for the balancing of these elements for now.  I'm awaiting the verdict of other playtesters whether things are now fine.  Or whether having, say, hovertanks too early has made it the way to win the game now, so now the game is stupid.  Time will tell.

[dino]

It's a simple trick of my design. I've made disabled User tech a prerequisite for progenitor psych, so you can't start developing it, you can only get it from aliens that start with it.
I've also took out of these techs non alien related stuff and made them not required for transcendence, so it's a completely optional techtree branch now, silly trick but I really like it.

99% is my wild guess, based on activity on forums like this one and an amount of copies garbage like civ 5,6 sells.
I took it out of my ass basically, but I have no doubt that gamers like we are irrelevant to game publishers.
Sid's paraphrased quotes about design philosphy and target audience are real, but I red it long time ago and have no idea where.

Not moving overpowered stuff upward, was related mostly to crawlers and satellites and directed at Inductio.
Satellites should be super easy for an AI to handle, so they can be a great tool for a mod that aim at challenging AI.
But in the current form If AI will handle them to their full potential, the game will degenerate completely into who gets them first.
I want them available as an easy, safe and good development strategy, just not the best possible, not to mention a completely game breaking runaway for an already winning faction.

SCs are gamebreakers, but increasing their cost and maybe removing the ability to disband full minerals, should be enough to fix them.

[bvanevery]

Belatedly I edited my post about the Alien techs.  I get it now.  "Faction bottlenecked" starter techs could be done the same way.  What if only Santiago has got the lasers??!  Problem: AI would never understand the value of the tech exclusions.  Well unless there was a Secret Project guarding each.

99% is my wild guess, based on activity on forums like this and an amount of copies garbage like civ 5,6 sells.

Oh, well yes, Firaxis went in some garbage populist direction in recent years.  I'm done with them as a company.  But there are other companies, and a "more serious" 4X TBS scene, exemplified by https://www.reddit.com/r/4Xgaming/ .  Civ is definitely not king of the hill there.

I took it out of my ass basically, but I have no doubt that gamers like we are irrelevant to game publishers.

Hardcore 4X TBS is a niche.  But it's a niche that exists, and some developers are attempting to serve that small market.  Unlike some other market spaces, I think there's room in that market, to be noticed for doing a good job.  The job of, say, doing a title similar to SMAC properly, still hasn't been done yet, AFAICT.