Author Topic: SMACX AI Growth mod  (Read 38380 times)

0 Members and 1 Guest are viewing this topic.

Offline tnevolin

Re: SMACX AI Growth mod
« Reply #180 on: October 22, 2018, 05:50:24 PM »
Keep also in mind that the research rate is generally increasing throughout the game. It is something 10-20 per tech at the beginning shortening to 2-3 at the end. So the accelerated weapon-armor progress that you see on a chart gets awfully more accelerated in practice.
Tim

Offline bvanevery

  • Emperor of the Tanks
  • Librarian
  • *
  • Posts: 3614
  • €2930
  • View Inventory
  • Send /Gift
  • Allows access to AC2's quiz & chess sections for 144 hours from time of use.  You can't do without Leadship  Must. have. caffeine. -Ahhhhh; good.  
  • Planning for the next 20 years of SMACX.
  • AC Text modder Author of at least one AAR
    • View Profile
    • Awards
Re: SMACX AI Growth mod
« Reply #181 on: October 22, 2018, 06:23:49 PM »
Overall much better than vanilla but still curious why you don't use a smooth numeric progression when you could. Like in weapons you have 2,4. It's a 100% jump. You could use smoother progression 2,3,4.

For early game weapons, it's the progression of the original game though.  I wasn't under the impression that there was any problem that needed correction here.  Bear in mind, the current approach uses all the weapons artwork that's available.  There isn't new artwork for a 3 gun, it'll probably look like a 2 gun.

Quote
For the same purpose you could modify values for special weapons like R-Laser and R-Bolt to occupy free spots instead of just repeating same values.

Those slots are hardwired with the 25% anti-indigenous life form bonus.  Any size gun you make with them, they will have that ability attached.

Quote
It would add some variability. Like you may give them weaker basic rating in lieu of special power that makes them stronger against psi.

The ability to change that doesn't exist at the .txt mod level.  It's hardwired into the binary.  In general, Firaxis was fairly generic but not perfectly generic about how they implemented stuff.  To me it's a big lesson how to write or not write a game's data files.

Quote
Same story with armor. Look at 5,8,10 progression. Your second step is smaller than first one. Not that such small differences are critical but still.

The problem is there isn't any more unique artwork for armor.  I think I can add more armor, but it's going to look the same as some other kind of armor.  I'm not sure this is worth it, as it's going to create more pressure on the limited number of Unit Design Workshop entries.  I will contemplate it though.

Quote
Regarding keeping 1:1 ratio. I've analyzed technology research index and your progression seems more or less keeping to that on all path except first quarter of the research path when you have armor 3 for quite a long time while you already have weapon 5 and it takes some time until armor catches with weapon at 8 both.

Not sure if your analysis is a static analysis of the tree, or dynamic in terms of actual length of time to gain armor as measured in a game you played.  Silksteel is strength 5 in my mod.  In the real world it could take a long time to gain Silksteel, maybe longer than Superconductor, because there are so many Conquer techs.  You would get other things of equivalent value in the same C5 tier of tech though.

Quote
# Disable healing.

I'm not aware of any .txt mod for that.  Also I don't see why it helps.  The issue is determined by who wins combat.

Quote
# Disable reactor progression.

That's possible, but I haven't committed to doing it.  I've wanted to hear from multiple playtesters whether they think the current state of affairs is a problem or not.  Fusion Power does at least seem to come later in my mod, in practice.

Quote
# Decelerate weapon and armor progression.

This would have to be severe to work.  Basically, weapons and armor would have to stop progressing by midgame, leaving the job of progression to the reactors only.  It would completely mess up the unit artwork for weapons, making the cure worse than the disease.

Quote
# Make armor progress faster than weapon.

It won't matter.  Bigger reactors will still blow the armor to pieces.

Quote
Out of these former two are pretty difficult exe modding.

Incorrect: reactor types are easily disabled in alphax.txt.

Quote
Reversed weapon to armor ratio will keep weaker factions from quick total annihilation maintaining overall game power balance.

Although I do not want the game to turn into a cakewalk, I also do want the game to be over at some point, and not drag out forever in a WW I style trench warfare stalemate.  This is especially true in Single Player, where between my quality of life and the AI's quality of life, only 1 matters!  The game has often penalized the player for having the temerity to try to win it, such as dire consequences for using chemical weapons or nukes.

The footrace doesn't have to go on forever.  It's ok to be badder ass than everyone else and wipe them out easily at some point.  To me it's all a question of how much work and challenge you had to do to get there. 

Offline tnevolin

Re: SMACX AI Growth mod
« Reply #182 on: October 22, 2018, 07:06:54 PM »
Overall much better than vanilla but still curious why you don't use a smooth numeric progression when you could. Like in weapons you have 2,4. It's a 100% jump. You could use smoother progression 2,3,4.

For early game weapons, it's the progression of the original game though.  I wasn't under the impression that there was any problem that needed correction here.  Bear in mind, the current approach uses all the weapons artwork that's available.  There isn't new artwork for a 3 gun, it'll probably look like a 2 gun.
I didn't mean to add a new one. Merely rearranging existing ones. Like making the one 4 to have 3 rating and 5 to have 4 rating, etc.
Same applies to armor. Just rearrange their ratings to make the progression look better. Just a single number change in alphax.txt. Easy and flexible.

Quote
For the same purpose you could modify values for special weapons like R-Laser and R-Bolt to occupy free spots instead of just repeating same values.

Those slots are hardwired with the 25% anti-indigenous life form bonus.  Any size gun you make with them, they will have that ability attached.
Who mentioned the ability? Keep the ability! I meant to lower its attack rating keeping the price. This way you have the extra ability but lower attack rating and fill up in the gap at the same time.

Quote
Regarding keeping 1:1 ratio. I've analyzed technology research index and your progression seems more or less keeping to that on all path except first quarter of the research path when you have armor 3 for quite a long time while you already have weapon 5 and it takes some time until armor catches with weapon at 8 both.

Not sure if your analysis is a static analysis of the tree, or dynamic in terms of actual length of time to gain armor as measured in a game you played.  Silksteel is strength 5 in my mod.  In the real world it could take a long time to gain Silksteel, maybe longer than Superconductor, because there are so many Conquer techs.  You would get other things of equivalent value in the same C5 tier of tech though.
It's static, of course. Dynamic changes with every game. It gives the rough understanding of what follow what, though. See also my comment in next post regarding actual research rate. You can keep it in mind while looking at the chart. I am sure you have your own sense of technology discovery rate at different game stages being played so many games already.

Quote
# Disable healing.

I'm not aware of any .txt mod for that.  Also I don't see why it helps.  The issue is determined by who wins combat.
This is exe mod required for that, obviously. That what I mentioned below.
The issue is not about who wins a single combat. It is about how much each side invest to advance a front line. But this is completely different and much bigger discussion. If you like to have on it - go to my recent combat discussion topic.

Quote
# Decelerate weapon and armor progression.

This would have to be severe to work.  Basically, weapons and armor would have to stop progressing by midgame, leaving the job of progression to the reactors only.  It would completely mess up the unit artwork for weapons, making the cure worse than the disease.
I don't see what artwork has to do with it. Do you mean the way weapon look affects game strategy??? O_o
You probably confused "decelerate" with "decrease". It definitely should increase but the advancement speed should decline in term of discovery time. In other words, you should invest more and more time (or more and more labs) to discover same incremental improvement. That would compensate to kaleidoscopic technology research rate toward the end.

Quote
# Make armor progress faster than weapon.

It won't matter.  Bigger reactors will still blow the armor to pieces.
Common, man! Why are you feeling yourself obliged to criticize my every single sentence even if you cannot find a descent argument??? Don't you have better things to think of? If you feel like it is not important - just leave it alone. I am not a robot, you know. I am trying to formulate my thoughts to the best of my abilities to make it clear to you and other listeners. NOT to make them right. I am just thinking out loud, you know. Trying to offend me and showing how stupid I am for every letter I wrote is not constructive dialog and is boring. Let's keep it interesting.
Regarding the question itself. I does matter. Everything matters. Every single detail. All of them combined improve or break balance. The question is about quantity and proportion. Make armor progress at about the rate of reactor and voila! Nothing stops you. Make it even stronger and reactor won't be enough to penetrate it.


Quote
Out of these former two are pretty difficult exe modding.

Incorrect: reactor types are easily disabled in alphax.txt.
Damn. I didn't even try to be correct all the time! There are things I don't know. You could just point me to where it can be done in alphax.txt, for example. I would appreciate that.

Although I do not want the game to turn into a cakewalk, I also do want the game to be over at some point, and not drag out forever in a WW I style trench warfare stalemate.  This is especially true in Single Player, where between my quality of life and the AI's quality of life, only 1 matters!  The game has often penalized the player for having the temerity to try to win it, such as dire consequences for using chemical weapons or nukes.

The footrace doesn't have to go on forever.  It's ok to be badder ass than everyone else and wipe them out easily at some point.  To me it's all a question of how much work and challenge you had to do to get there.
Err, you confused me. Why are you working on your mod then? ANY game is like that you just described. ANY game gives you a chance to end it at some point. Why did you complain about vanilla SMAX when you can practically "end the game" at about 100-150 turn? Enjoy. Somehow you claimed that you don't like it and want it to last longer. Not in these words, I know. Please don't argue with wording - I won't reply to it anyway. You says "balance" but it is the same. Generally more balanced game = less benefits from exploits = better AI playing = more challenging game = longer game. If you are working on more challenging but specifically shorter game - that is completely different story and different game. If so, make it explicit and I won't waste your time anymore.
My problem with SMACX and Civ 2 is that they were advertised to be played for 400 turns and they target Ascent to Transcendence as an ultimate goal and method of victory. It even cost 800 points more than any other victory in SMACX. That is what I want to fix. To make achieving that type of victory to be not boring and rewarding in term of scoring, for example.
Tim

Offline bvanevery

  • Emperor of the Tanks
  • Librarian
  • *
  • Posts: 3614
  • €2930
  • View Inventory
  • Send /Gift
  • Allows access to AC2's quiz & chess sections for 144 hours from time of use.  You can't do without Leadship  Must. have. caffeine. -Ahhhhh; good.  
  • Planning for the next 20 years of SMACX.
  • AC Text modder Author of at least one AAR
    • View Profile
    • Awards
Re: SMACX AI Growth mod
« Reply #183 on: October 22, 2018, 07:12:18 PM »
Keep also in mind that the research rate is generally increasing throughout the game. It is something 10-20 per tech at the beginning shortening to 2-3 at the end. So the accelerated weapon-armor progress that you see on a chart gets awfully more accelerated in practice.

I have realized that one consequence of eliminating reactor sizes, is your nukes, tectonic loads, and fungal loads won't have nice blast radii anymore.  That IMO is a serious consideration.  I may not use these things in every game, but when I do play a game where I want to use them, I want them to work.

I wonder if I made advanced weapons or armor disproportionately expensive, if it could even out the disproportionate cheapness of the advanced reactors.  You'd still have units that would wipe the floor with everyone else, but they could cost you a pretty penny to obtain.  That would mean your manufacturing infrastructure would have to be pretty robust to make advanced units.  It might also make mindworms the best possible deal, since they don't have reactors.

Offline bvanevery

  • Emperor of the Tanks
  • Librarian
  • *
  • Posts: 3614
  • €2930
  • View Inventory
  • Send /Gift
  • Allows access to AC2's quiz & chess sections for 144 hours from time of use.  You can't do without Leadship  Must. have. caffeine. -Ahhhhh; good.  
  • Planning for the next 20 years of SMACX.
  • AC Text modder Author of at least one AAR
    • View Profile
    • Awards
Re: SMACX AI Growth mod
« Reply #184 on: October 22, 2018, 08:15:23 PM »
There isn't new artwork for a 3 gun, it'll probably look like a 2 gun.
I didn't mean to add a new one. Merely rearranging existing ones. Like making the one 4 to have 3 rating and 5 to have 4 rating, etc.

The weapons artwork cannot be readily changed like that.  The "5" currently looks like a Gatling gun.  If you change it to "4", it looks like a Particle Impactor.  There's a fixed numeric range where the weapon looks like what it's going to look like, until the next weapon type appears.  You simply cannot have something that looks like a Particle Impactor and is strength 3, nor strength 5.  In those cases it will look like a simple Laser or a Gatling Gun, respectively.  The weapons artwork is hardwired into the binary, you can't change it.  You can only adjust it within ranges.  I believe the permissible ranges are:
1, 2..3, 4, 5, 6..7, 8..9, 10..11, 12, 13..15, 16..19, 20..23, 24..29, 30+

Quote
Same applies to armor.

Armor fortunately does not have this problem.  You can assign any number to the armors that you like.  However, there are substantially fewer armor art assets than weapon art assets.  This requires compromises for spreading them over the entire game.  I'm not sure if a new armor can be added, or what artwork it will have.  If for instance it has to look like the Temporal Stasis shields, well that's simply not acceptable to stick anywhere earlier than the late game.  In that case I could add another armor to the late game and push armor back 1 slot, all the way to the beginning, and that might help alleviate "armor pressure".  But that may be all that can be done.  I will check if it's even possible.

Quote
Just rearrange their ratings to make the progression look better. Just a single number change in alphax.txt. Easy and flexible.

Don't be quick to assume that anything is easy and flexible unless you've actually modded it yourself.   :D  The implementation is quirky.

Quote
Who mentioned the ability? Keep the ability! I meant to lower its attack rating keeping the price. This way you have the extra ability but lower attack rating and fill up in the gap at the same time.

That might be reasonable.  It irritates me that Missiles are basically useless because of R-Lasers now.  The artwork for the Resonance attachment is generalized.  Any weapon can be made into a Resonance weapon.  It superimposes a module on top of an existing gun.

Quote
It's static, of course.

How do you account for different probability of discovery according to Explore, Discover, Build, Conquer weights?

Quote
I don't see what artwork has to do with it. Do you mean the way weapon look affects game strategy??? O_o

It affects mod enjoyability.  I'm not going to get in the business of throwing out the game's art assets without exceedingly good reasons to do so.  I'm not just trying to perfect progressions, I'm trying to get as many people as possible to play my mod.  People can rightly object to the stuff they like about SMAC, being taken out of SMAC.  They say "this mod sucks", stop playing, tell other people it sucks.  Or maybe they don't say any of that, they just stop playing, and don't think to say anything positive in the future.  That hurts my cause as well. 

Quote
In other words, you should invest more and more time (or more and more labs) to discover same incremental improvement.

Any game fixing that relies on "more and more research" is doomed.  It is not possible to control a player's research.  It will exponentially increase.  There are simply too many ways to become tech powerful to avoid this.

Quote
That would compensate to kaleidoscopic technology research rate toward the end.

It is not possible to compensate for this at all.  It's a fool's errand.  Not without severe "tech stagnation" changes, making the game pretty much unrecognizable.  Mart did something vaguely like that in the last scenario contest I participated in, some time ago.  There was this guy named Kirov, I think, who creamed the crap out of the tech regardless of all the super slow tech impediments that Mart put up.  Kirov and I got mad at each other, and I actually floated the idea that he might be cheating.  In hindsight, no, that wasn't the problem.  He's just an indefatiguable calculator and knew better ways to exploit the game than I did.  Part of his secret sauce was recognizing the Governorship is worth a lot of money, something that isn't high priority in my usual play styles.  Then I figure he did satellite abuse.  And he built like every lab you can think of.  These things are harder in my mod, but still not impossible.  It's all going to happen eventually.

Quote
Quote
It won't matter.  Bigger reactors will still blow the armor to pieces.
Common, man! Why are you feeling yourself obliged to criticize my every single sentence even if you cannot find a descent argument???

First off, if you're wrong, you're wrong.  If you think you're right, you can mod a test case yourself, and then in the course of events find out that you're wrong.  I've been at this a long time.  Although I can be proven wrong on some things, I'm most usually not.  Due to that 1+ year of AARs and 4 full time person months of modding and all that.  I've got a lot of coverage of issues already.

Quote
Don't you have better things to think of? If you feel like it is not important - just leave it alone. I am not a robot, you know. I am trying to formulate my thoughts to the best of my abilities to make it clear to you and other listeners. NOT to make them right. I am just thinking out loud, you know. Trying to offend me and showing how stupid I am for every letter I wrote is not constructive dialog and is boring. Let's keep it interesting.

I appreciate feedback, and you can ascribe all sorts of imagined motives for why I respond to a specific comment.  My responses are of the sort: "this might work... this is clearly not going to work."  Hey if you think you can brainstorm as a form of giving feedback, I don't see why I shouldn't be answering in turn.  You speak, I don't, what kind of arrangement is that?

Quote
Damn. I didn't even try to be correct all the time! There are things I don't know.

Right.  So... don't ascribe negative motive too quickly to me.  I just may know more about the "endgame" of some approach than you do, since I've been at this awhile.

Quote
The footrace doesn't have to go on forever.  It's ok to be badder ass than everyone else and wipe them out easily at some point.  To me it's all a question of how much work and challenge you had to do to get there.
Err, you confused me. Why are you working on your mod then?

Modding is authorship.  Games are not likely to be made or broken on one narrow concern, like whether the weapons progression has exactly perfect numerical intervals.  A proper analogy is writing a book.  There is no analytic formula that determines what a correctly written book is, you have to actually write the book.  Once you've done a draft, you see what you've got.  Then you might have to redraft, a lot.  In game design, certainly, your most basic tool is iterating on the design over and over again.  Until you personally find it to be "good".  And you get some corroboration from others that what you've done is good.

Quote
Somehow you claimed that you don't like it and want it to last longer. Not in these words, I know.

"Last longer" is a highly material design dimension here.  It is not a term I've generally used.  At the extreme limit, there are people who want the game to last forever.  They sandbox, because that's what they like to do.  I'm not into that.  The vast majority of the years I've spent playing SMAC, have been about optimizing my path to victory.  I have to experience the path in order to win, I have to make choices along it.  My mod is about being able to make more satisfying choicesNot about artificially extending the life of the game.

Sandboxers, BTW, want bigger reactors.  And I want big nukes and tectonics and fungus, when I choose to do that in a game.  So I think this means they have to stay in.  I will consider the feasibility of adjusting various costs to compensate.

Quote
Generally more balanced game = less benefits from exploits = better AI playing = more challenging game

With you so far.

Quote
= longer game.

Nope.  Furthermore, one must distinguish between game time and real wall clock time.  It's Turn Based Strategy.  You could think for an hour about one game turn, if you were so inclined.  Actually in very long games, that can happen organically, just due to all the units you're pushing around.  I think of this all in terms of "the mouseclick budget".  If a game is wasting my mouseclicks, I am not happy.

So for instance, some of your design ideas in some threads have been of the flavor, hey, let's "solve" problems by more turtling up!  Well I've played enough Civ-style games where turtling up was the order of the day.  The early game of Freeciv was heavily that way, a city wall on a hill is basically impregnable.  Pretty much had to wait until you got Cannons.  Catapults were too expensive for not enough firepower and would tend to die en masse.  This has the effect of reducing choice, your realistic choice is to put all your civilization energy into getting better tech as fast as you can.

Quote
If you are working on more challenging but specifically shorter game - that is completely different story and different game. If so, make it explicit and I won't waste your time anymore.

This mod is not "abbreviated SMAC".  I have imagined writing that mod, but realistically, I'm only ever going to write and champion one mod.  It is too much work to get a mod adopted by players, to split their attention 2 ways.  Let alone my development energy.  Any "abbreviated SMAC" ideas are going into a new game, to try to make money.  SMAC has lots more crap in it than is needed.

Offline tnevolin

Re: SMACX AI Growth mod
« Reply #185 on: October 22, 2018, 08:52:17 PM »
I got your points generally. Thanks for sharing.

You didn't share how to disable reactor in alphax.txt, though.
 :)
Tim

Offline bvanevery

  • Emperor of the Tanks
  • Librarian
  • *
  • Posts: 3614
  • €2930
  • View Inventory
  • Send /Gift
  • Allows access to AC2's quiz & chess sections for 144 hours from time of use.  You can't do without Leadship  Must. have. caffeine. -Ahhhhh; good.  
  • Planning for the next 20 years of SMACX.
  • AC Text modder Author of at least one AAR
    • View Profile
    • Awards
Re: SMACX AI Growth mod
« Reply #186 on: October 22, 2018, 10:04:33 PM »
Reactors simply have a prerequisite tech.  Change it.  For instance, "Disable".

I wonder if the game barfs if I change the prereq to Fission reactors to something other than "None" ?

Setting it to another tech, like Industrial base, doesn't do anything.  Units are still shown with Fission reactors, but the Fission reactor choice in the Unit Design Workshop is greyed out.

Setting it to "Disable", completely removes the reactor from the available choices.  It isn't even greyed out, it is completely gone.  Units still have Fission reactors though.

I just tried adding a weapon and an armor to their respective tables.  It doesn't work, they are ignored.  So, have to distribute the ones that exist as best as I can.

Verified that a strength 3 weapon does indeed look like a Laser.

« Last Edit: October 22, 2018, 10:21:41 PM by bvanevery »

Offline bvanevery

  • Emperor of the Tanks
  • Librarian
  • *
  • Posts: 3614
  • €2930
  • View Inventory
  • Send /Gift
  • Allows access to AC2's quiz & chess sections for 144 hours from time of use.  You can't do without Leadship  Must. have. caffeine. -Ahhhhh; good.  
  • Planning for the next 20 years of SMACX.
  • AC Text modder Author of at least one AAR
    • View Profile
    • Awards
Re: SMACX AI Growth mod
« Reply #187 on: October 23, 2018, 08:53:54 PM »
CHANGELOG
Changed from version 1.22 to 1.23:

- Free Drones: changed personality back to Erratic.  When they were Passive, I saw them get bottled up unnecessarily by the Peacekeepers, who are also Passive.  The Drones clearly should have fought for territory.
- Pirates: removed incorrect Faction Help entry saying they get +1 PROBE and -1 GROWTH.  They get neither.

- Heavy Artillery: moved to C1 Applied Physics.
- Bioadaptive Resonance: changed it from a C5 to a C3 tech.
- Soporific Gas Pods: moved to C4 Retroviral Engineering.
- Resonance Laser: now has attack strength 3 and cost 3.
- Silksteel Armor: now has defense strength 4 and cost 4.
- Photon Wall: now has defense strength 6 and cost 6.
- Probability Sheath: now has defense strength 8 and cost 8.

- Progenitor Psych: set power=4 and growth=3, making it a C2 tech.  Need it as a Conquer prereq.  Previously it only had growth=1, making it gratuitously hard to discover, which is probably not a good idea after all.
- Field Modulation: set power=4 and growth=3, making it a C3 tech.  Need it as a Conquer prereq.
- Nonlinear Mathematics: changed it from a C3 to a C4 tech.
- Silksteel Alloys: changed it from a C5 to a C4 tech.
- Carrier Deck: moved to C5 Doctrine: Air Power.
- Cruiser Colony Pod: moved to C5 Doctrine: Air Power.
- Submarine Probe Team: moved to C5 Doctrine: Air Power.
- Superconductor: changed it from a C4 to a C5 tech.
- Missile Launcher: moved to C6 Orbital Spaceflight.
- Chaos Gun: moved to C7 Probability Mechanics.
- Superstring Theory: changed it from a C6 to a C8 tech.
- Phaser: moved to C8 Superstring Theory.
- String Resonance: changed it from a C8 to a C9 tech.
- Sentient Resonance: changed it from a C9 to a C10 tech.
- Matter Compression: changed it from a C9 to a C10 tech.
- Clinical Immortality: moved to Matter Editation.
- Matter Editation: set power=3 and growth=3 due to Clinical Immortality.
- Temporal Mechanics: changed it from a C10 to a C11 tech.  Set wealth=0 and growth=0 due to removal of Clinical Immortality.
- Applied Gravitonics: changed it from a C11 to a C12 tech.
- Singularity Mechanics: changed it from a C12 to a C13 tech.
- Controlled Singularity: changed it from a B13 to a B14 tech.
- Intertial Damping: changed it from a C13 to a C14 tech.
- Threshold of Transcendence: changed it from a D14 to a D15 tech.
- changed many prereqs

The attachment limit per post is 5, and I eventually ran out of room at the top of the thread.  Consequently, this becomes the new home of version 1.23.  It was downloaded 60 times before I moved it here.
« Last Edit: February 18, 2019, 06:36:30 AM by bvanevery »

Offline vonbach

Re: SMACX AI Growth mod
« Reply #188 on: October 25, 2018, 03:13:58 AM »
Passive drones never do well. You have to play aggressively with them. Oh and as I recall revolt doesn't effect the AI at all.

Offline bvanevery

  • Emperor of the Tanks
  • Librarian
  • *
  • Posts: 3614
  • €2930
  • View Inventory
  • Send /Gift
  • Allows access to AC2's quiz & chess sections for 144 hours from time of use.  You can't do without Leadship  Must. have. caffeine. -Ahhhhh; good.  
  • Planning for the next 20 years of SMACX.
  • AC Text modder Author of at least one AAR
    • View Profile
    • Awards
Re: SMACX AI Growth mod
« Reply #189 on: October 25, 2018, 05:53:50 AM »
Actually I thought the Drones were doing ok with Passive.  Thought I did some AI vs. AI test games when I set them that way in version 1.20 almost a month ago.  But then I saw them corralled by another Passive player, Lal, who really spreads all over the place now.  So I thought it was less that they couldn't do well, and more that they do badly with Passive when presented with certain situations.  Like a Passive early colonization bottleneck.

I think it's been eons since I've seen any city revolt and change hands.  It's not generally something I allow my own cities to do.  Haven't really seen Drone Riots inflicted by foreign meddlers in a long time either, probably because of the way I changed PROBE bonuses.  I think the AI doesn't value probe teams all that highly anymore.  It'll make some but not tons, which is fine by me!

Offline vonbach

Re: SMACX AI Growth mod
« Reply #190 on: October 25, 2018, 01:54:08 PM »
As far as I know Revolt doesn't work on the computer in any way whatsoever. I've never seen it work. My favorite faction is the Drones actually.
But the computer never did well with them at all. It was far to passive and unless left alone for a long time
they simply got flattened.  Try making them as aggressive as Santiago and they would be far more formidable.

Offline bvanevery

  • Emperor of the Tanks
  • Librarian
  • *
  • Posts: 3614
  • €2930
  • View Inventory
  • Send /Gift
  • Allows access to AC2's quiz & chess sections for 144 hours from time of use.  You can't do without Leadship  Must. have. caffeine. -Ahhhhh; good.  
  • Planning for the next 20 years of SMACX.
  • AC Text modder Author of at least one AAR
    • View Profile
    • Awards
Re: SMACX AI Growth mod
« Reply #191 on: October 25, 2018, 03:03:12 PM »
I made Santiago Aggressive in my mod.  She isn't by default, she's merely Erratic.  The Caretakers are Aggressive by default and I made them merely Erratic in my mod, to try to distinguish them from the Usurpers.  The Believers are Aggressive, like the original game.  I don't actually believe that Aggressive is the One True Way to win, I think some factions are actually better off making less war or no war.  This is particularly true on Huge maps or larger, when the faction has some inherent growth / infrastructure advantages.  The Spartans, Usurpers, and Believers are Aggressive because their main advantages are for attacking people.  +2 MORALE, +1 MORALE, and 25% Attack bonus respectively.

The Peacekeepers, Morganites, University, and Pirates are all getting away with Passive just fine.  Doesn't mean they won't make war.  It means they tend to concentrate on getting huge, get big factional resource advantages, and then decide they will clobber somebody.

Offline vonbach

Re: SMACX AI Growth mod
« Reply #192 on: October 25, 2018, 08:57:33 PM »
Santiago is default erratic? I did not know that interesting.

Offline bvanevery

  • Emperor of the Tanks
  • Librarian
  • *
  • Posts: 3614
  • €2930
  • View Inventory
  • Send /Gift
  • Allows access to AC2's quiz & chess sections for 144 hours from time of use.  You can't do without Leadship  Must. have. caffeine. -Ahhhhh; good.  
  • Planning for the next 20 years of SMACX.
  • AC Text modder Author of at least one AAR
    • View Profile
    • Awards
Re: SMACX AI Growth mod
« Reply #193 on: October 26, 2018, 01:20:58 AM »
Yeah I wouldn't have guessed.  I think making her Aggressive made her tougher, and it makes sense, since she's got the +2 MORALE advantage to feed.  Of course, I also removed her INDUSTRY penalty.  Although, I removed her POLICE bonus as I couldn't have her with better police than Yang.  She also has to walk at the beginning of the game, same as anyone else.  Most factions don't get starting tech bonuses either.


Offline bvanevery

  • Emperor of the Tanks
  • Librarian
  • *
  • Posts: 3614
  • €2930
  • View Inventory
  • Send /Gift
  • Allows access to AC2's quiz & chess sections for 144 hours from time of use.  You can't do without Leadship  Must. have. caffeine. -Ahhhhh; good.  
  • Planning for the next 20 years of SMACX.
  • AC Text modder Author of at least one AAR
    • View Profile
    • Awards
Re: SMACX AI Growth mod
« Reply #194 on: October 27, 2018, 04:24:29 PM »
I'm thinking it takes too long to get Command Centers.  Currently they become available with C2 Planetary Networks, which also gives Probe Teams and Police State.  Due to the breadth of Conquer techs available, it can take a surprisingly long time to obtain, even with a pure Conquer focus as the Spartans.

I'm not sure of any pattern with C2 techs, but it seems like Doctrine: Loyalty is more likely to be researched first.  The net effect is that learning how to make Command Centers is rather random, with some factions possibly gaining it much earlier than others.  Maybe it has made faction potential lopsided, I'm not sure, as I've never paid close attention to this issue.  But as a player I find it a bit frustrating, wondering "Where are my !@#$#!$ Command Centers??!"

I will move them back to Doctrine: Mobility and do a 1.24 release.  I will keep the "no more than once a week" rhythm and release on Oct. 30th.  If anyone has any other suggestions / issues, now would be a good time to make them known.

Once upon a time, I was worried about the AI building Command Centers obsessively in the early game.  The AI also seemed to obsess about building artillery, if it was available early.  The latter doesn't seem to be a problem anymore, so perhaps the former isn't either.  I actually enjoyed the delay in Command Center availability for awhile, but now it's taking too long.  It's as though a Conquer 1.5 tier of techs is needed.  It probably means tier 1 has gotten broader, and tier 2 as well, so just harder to get any particular tier 2 tech.

 

* User

Welcome, Guest. Please login or register.
Did you miss your activation email?


Login with username, password and session length

Select language:

* Community poll

SMAC v.4 SMAX v.2 (or previous versions)
-=-
17 (6%)
XP Compatibility patch
-=-
8 (3%)
Gog version for Windows
-=-
69 (27%)
Scient (unofficial) patch
-=-
25 (10%)
Kyrub's latest patch
-=-
14 (5%)
Yitzi's latest patch
-=-
85 (34%)
AC for Mac
-=-
2 (0%)
AC for Linux
-=-
5 (2%)
Gog version for Mac
-=-
10 (4%)
No patch
-=-
12 (4%)
Total Members Voted: 247
AC2 Wiki Logo

* Random quote

Until now the battle had been proceeding smoothly. The enemy was outflanked and had been driven from the reactor housing, but against the reactor itself the matter canons were strangely ineffective. Rounds simply-stopped-in mid air.
~Col. Corazon Santiago 'A Tactical History of Sparta'

* Select your theme

*