SMAX - The Will to Power - mod

[Alpha Centauri Bear]

It doesn't appear to say what your combat formula really is. The example you give is a bit odd. A unit with 2-1 combat strength advantage should of course win with very high probability. However it should take damage in the process.

You mean multi-round combat calculations or just round odds? Round odds are just proportion of strengths as displayed in combat statistics window. The multi-round combat calculations with adjusted loss divisor is very complex. I don't put it in readme to not scare people. 🤣

There are sample charts visualizing different value of this parameter effect.
Vanilla: http://alphacentauri2.info/index.php?topic=21371.msg122835#msg122835
This mod: http://alphacentauri2.info/index.php?topic=21359.msg122935#msg122935

[lolada]

Hello Tim, glad to have you back. I'll try to leave as much feedback as i can - i played ton of hours ^^ but still its only a few games; AC is micro intensive.. at least i tend to play a bit slow.

I'll start with native units, PSI combat and Planet rating - as i already commented earlier as i  played a lot more with worms due to my AAR. Its mostly positive if not completely.. lets see where i get.

- Native lifeform is dangerous - thats great - but not too dangerous. Its really beneficial to have Trance units, and they rarely die. Think you got good balance there. It looks its hard to mod native units to be dangerous.

- Planet rating is valuable - it is in good spot. Its not too strong nor too weak - its also hard to stack. For example picking up Green is no joke with that -2 Industry penalty. I picked it up later in game just for that bonus attack and ended up picking Fundamentalism for 1 Industry to counter Green penalty.

- Its tough to attack into worms with negative Planet rating - thats cool. Positive in that sense helps a lot to reduce casualties. Someone like Miriam can have real trouble getting rid of worms as you can just attack them and hope for 3 vs 2. Artillery appears to be strong counter.. and empath units are valuable.

- Planet rating is also great for fungus resources. For example Miriam really suffers with -1 planet - in my game i was lucky to be near Manifold Nexus that moved Planet to 0 - so i could use fungus - its awesome for new bases once you have +2 food on it

- on combat worms seem to be well balanced - they were my best attackers in any tougher situation vs strong defenders or perim. defense. As they should be for planet based faction. I complained earlier, they are really not good for attacking non-combat units, but honestly, it looks like thats their main weakness. It keeps them in check - and its very wise to have conventional attack units to save worms from dying on formers or on low tech units. So they are not self-sufficient as they are quite expensive at 80 minerals. I suppose thats ok, i don't know how would i improve them without making them broken.

- i've seen ai building both empath and trance units - so they do counter worms. AI also spams probes and they can often kill worms in defense

- i also really like what you did with Biology Labs - great change

- i used ton of Spore Launchers - their cost is 4 minerals row.. so relatively cheap - they are great police unit for Cha Dawn. They are also quite good defenders.. in attack worm artillery is kinda meh, but they do a bit of damage sometimes. I much prefer conventional artillery for bombardment. Probably good move to leave them cheap - otherwise they wouldn't be really used. One might abuse them to ruin AI economy - but there's ton of ways to exploit AI i am not into that.

- Mindworms and Sealurks are 80 minerals, Isle of the Deep, Locusts are more expensive a bit - i'd say thats good - one can get them cheaper with industry rating; it restrict these units from being op. They are quite unique and with great abilities and for example they counter probe teams well

- That said - Brood Pit is awesome, but feels like it comes way too late - i think its E10 tech. Kind of sad that I can't use one of The Cult factions features until the end of the game. Its goal of the mod to make all features usable reasonably. Its very strong facility, especially for The Cult - but other factions would have to build those Brood Pits. With 25% Brood Pit discount i could really swarm enemies with worms this late in the game - they became unstoppable.
One suggestion could be to have Brood Pit earlier - maybe like tech 8, but nerf the facility with removing +lifecycle. I'd keep discount, that even could go to 10-15% or 20% instead of 25%. Or leave +lifecycle, but reduce discount a lot.. just bland +lifecycle without discount removes uniqueness from it. I could at this point produce Demon Boils (Elite) worms straight from some bases - thats kind of op - i had some special projects with +lifecycle.

I guess thats plenty for this post. The Manifold Nexus and fungus resource yields - i'll comment on that in separate post i think these could be changed somewhat, i am not sure how exactly yet.

[Alpha Centauri Bear]

Thank you, man. Great feedback. Mostly positive.
🤣

SMACX is a very feature reach game. A singe introduction of psi combat is huge addition. Unfortunately, when you have so many alternative ways to wage a war it is almost impossible to balance them nicely that no one is OP. I didn't do much about psi except 1:1 land odds and somewhat elevated native units cost. The rest is updated combat odds resolution that seemed to do the trick. Now no combat alternative seems to be a universal weapon. Which is good enough. However, it is still open field for experiments.

I like you liked Biology Labs. That's my pride. 😛
Previously it was completely neglected facility.

Brood Pit is a very powerful facility as you correctly pointed out. We can move it a little earlier with caution. One way to offset its earlier usage is to set exorbitant maintenance on it. Thus player would suffer economically to offset advanced benefits.
I wouldn't do other modification to its features like native unit cost discount, life-cycle, police rating. My rule of thumb is to stick to original game features as much as possible. I am not introducing new experience. Merely fixing existing one.
So if you are fine with offsetting it with high maintenance we can try it out.

[Hagen0]

You still could have told me. From what I got from the other thread you do only two combat rounds and then randomly assign the damage of the victor within the window given by round wins. Is that correct? It seems extremely random.

Btw, I really do think vanilla combat has its upsides. It is very unique favoring sting operations and hit-and-run over normal combat since you essentially can't reasonably build a stack of more than 3-4 units.

[Alpha Centauri Bear]

You still could have told me. From what I got from the other thread you do only two combat rounds and then randomly assign the damage of the victor within the window given by round wins. Is that correct? It seems extremely random.

Vanilla combat works like that. There is a round odds that determine who wins the round randomly. Loser then incurs some damage. The battle then is the series of rounds with same odds repeated over and over again until one unit dies.

My modification is that after first round the round odds are changed so that last round winner has its losing probability decreased in fixed proportion. That proportion is a loss_divider and is exposed in thinker.ini. Obviously loss_divider = 1 does not change anything and, thus, works exactly as in vanilla.

Here is an example. Let's say two Scout Patrols fight. Their base round odds are 1:1. Meaning each unit has 50% winning chance and losing chance. Randomly first unit wins first round. If we set loss_divider = 2. Then victor losing chance in next round will be 50% / 2 = 25%. That makes its winning chance = 75%. This stays like that for as long as first unit keeps winning rounds. If at some round second unit wins the round it flips to second unit benefit and its winning chance becomes 75% and so on.

Don't try to find a logic in this. This is just formula we have come up with that keeps the same average damage proportion in large scales as in vanilla (that is important). At the same time it randomizes individual battle outcome a little bit more avoiding horrendous vanilla skews.

In other words, massive armies battle outcomes and average unit economical combat effectiveness stay the same as in vanilla with more randomized individual battle outcome.

Btw, I really do think vanilla combat has its upsides. It is very unique favoring sting operations and hit-and-run over normal combat since you essentially can't reasonably build a stack of more than 3-4 units.

Have you listened to Sid Meier's presentation on how they tune battle outcomes to satisfy players? Well, yea. This is kind of upside vanilla combat has. It has nothing to do with unique flavoring. The fact that you can use sting operations is just unplanned consequence of a massive cheating on all levels to give player sure advantage over AI and guaranteed win. Think about it. 100% first place out of 7 factions who have like triple economical advantage over you on toughest difficulty - what is it if not cheating on large scale?
Induktio and I worked on AI improvement. That also included removing all this "sting operations" stuff that AI is not taught to understand. Now game is much more fairer. You can easily lose on highest difficulty. And that is awesome for challenge oriented people like me. This increases game replay-ability 100-fold.

By the way, I removed collateral damage completely. Now you can stack infinite number of units without consequences and wage large scale war any way you like.
😎

Again. I am not discarding any opinions. Any suggestion is welcome. If you really think these sting operations are valuable addition to the game let's open a discussion and think how can we add them to the game without breaking human-AI balance. I'm up to it.
😉

[lolada]

No problem ^^

On changing PSI i guess its ok - its hard to tell whats right balance. Nexii i think for example said he left psi at 3:2 and gave worms Rover chasis ^^ quite interesting solution - native worms now attack first hehe. That might be too strong in player hands.. but everything has its ups and downs.

Just checked this Brood Pit thing: so i got Centauri Genetics E10 Brood Pits technology - in 2393, so that is 293 turns in the game . Its on tech stagnation, blind research so it plays a significant part. Zakharov was wiped relatively early so that slowed things a bit. Most AIs were also using Fundamentalism, including me, so research was slowed. On that topic - apparently AI really like industry and it really likes morale - from what i can see from forums discussions. You Fundamentalism has +1 Probe, +1 Morale, +1 Industry, -2 Research - it might be favorite AI pick. Both Aliens and Drones used it a lot.. Deirdre also likes to pick simply Frontier. Roze preferred Democratic. But thats another topic.

Anyway by this point i solved most issues i had in the game - so it was mostly nice finisher tech to make the faction op. Worm spam + Locusts to finish the job. I got other E10 tech (Secrets of Alpha Centauri) with Temple of Planet much earlier - at M.Y. 2358 - just took a look there. 30+ turns is huge difference this late in the game. Centauri Genetics is kinda hidden behind lots of discovery, build and conquer prerequisite techs so that delays it. Swap these two ^^. I don't know i might be completely off here - its just one game, maybe too early for any changes.

More data, in 2393 my research (Explore/Conquer focus whole game):

Explore 77%, Discover 70%, Build 66%, Conquer 93%. It might be my crap RNG luck as i almost completed every conquer tech - some were likely very expensive. I have the feeling game has more conquer technologies than any others.

Lastly on topic of costs and maintenance. This late Brood Pit is very cheap with only 80 minerals and 2 maintenance.. its the cost of one mindworm. It could easily be 150 minerals, 4 maintenance - but it doesn't really matter too much. Temple of Planet is for example 220 minerals, 3 maintenance. At this point i had +560 energy credits per turn on 50% economy.

- I forgot to comment on Dream Twister and Neural Amplifier - there are obviously great and strong, but not too op. I got one, missed other and  worms still die. They are ofcourse better but its no cakewalk - one must spam a lot of units to overrun AI.

[Hagen0]

Do you still have stack damage? Did you change self destruct damage? Sorry for asking so many questions but your readme is not as detailed as I would like.

It seems Centauri Ecology removes restrictions? I like that. You may have nerfed forests a bit too much. I'm not sure they are worth it at all anymore.

[Alpha Centauri Bear]

@Hagen0

Here is another thing to think about for you. Look at this "sting operation" in this angle. What if we teach AI to recognize and use these "sting operation" properly. Now it will destroy your units left and right at any opportunity run back for healing and then return to the front line reinforced only to damage you even more without retaliation and losses at all? How would you like that? Would it be fun to play if AI would beat you into pulp even on lowest difficulty?
😜😝

AI is currently exceptionally dumb. We are trying to improve it and raise to some average human level to get ourselves a challenge. No more and no less. Fair conditions for all sides.

[Alpha Centauri Bear]

Do you still have stack damage? Did you change self destruct damage? Sorry for asking so many questions but your readme is not as detailed as I would like.

No. Stack damage is removed completely. However, this is configurable in ini file. You can revert it to vanilla if you like.

No problems with questions. I enjoy this game, forum, and talking to players. The more the merrier.
😁

It seems Centauri Ecology removes restrictions? I like that. You may have nerfed forests a bit too much. I'm not sure they are worth it at all anymore.

Correct about restrictions. Actually, skim my readme. There is a lot of reasoning why I did all these changes. You can find a lot of answers there. It is also a fun reading too.
😅

Forest is disputable thing as I said many times. Propose you way to modify it and add some reasoning and we'll go from there.
Even in this nerfed form forest is still usable quite a lot. You convert barren land into 1-2-0 in 12 turn. One factor of production per 4 terraforming turns. That is as good as building a farm. Besides, you have no better options in dunes. So it is far from being unusable.

[Hagen0]

I did read parts of your readme. Chatting with you is more fun though. The problem with your logic is that you compare desert tiles with normal terrain. (In vanilla bad land doesn't matter that much. In your mod it does.) If you have a rolling rainy tile a farm (4 turns) makes it a 3-1 tile with the option of adding energy yield. A forest instead is 12 turns for a inferior tile. My point is that you would not forest desert tiles. You would avoid the area in the first place.

Did you change self-destruct damage? Also I didn't say that the hit-and-run style combat in vanilla did not have counterplay. Of course it had. There were overpowered ways to go about it (helicopters) though.

[Alpha Centauri Bear]

No problem ^^

No problems for Brood Pit maintenance? Please quote or mention what you are referring to. It is a flat forum without sub threads.

Anyway by this point i solved most issues i had in the game - so it was mostly nice finisher tech to make the faction op. Worm spam + Locusts to finish the job.

Are you saying native warfare is too OP at the end game? I would suspect so taking that conventional units become more expensive. However, didn't have a chance to test it much in end game.
From the other side it should be easily counterable by Hypnotic/Empath units along with AC. I believe AC does protect from Locust!

I got other E10 tech (Secrets of Alpha Centauri) with Temple of Planet much earlier - at M.Y. 2358 - just took a look there. 30+ turns is huge difference this late in the game. Centauri Genetics is kinda hidden behind lots of discovery, build and conquer prerequisite techs so that delays it. Swap these two ^^.

Secrets of Alpha Centauri and Centauri Genetics?
Centauri Genetics allows Locust - the most powerful native unit. I think level 10 is even too early for it. If anything it should be slightly delayed.

Explore 77%, Discover 70%, Build 66%, Conquer 93%. It might be my crap RNG luck as i almost completed every conquer tech - some were likely very expensive. I have the feeling game has more conquer technologies than any others.

That is completely normal. Conquer stays aside from others and many factions target it. Mostly your research focus is defined by Conquer or not Conquer. Therefore, if you focus on it 50% of the time it gets researched faster than other three.

Lastly on topic of costs and maintenance. This late Brood Pit is very cheap with only 80 minerals and 2 maintenance.. its the cost of one mindworm. It could easily be 150 minerals, 4 maintenance - but it doesn't really matter too much. Temple of Planet is for example 220 minerals, 3 maintenance. At this point i had +560 energy credits per turn on 50% economy.

Agree. Do you want to try it with 150/4?

- I forgot to comment on Dream Twister and Neural Amplifier - there are obviously great and strong, but not too op. I got one, missed other and  worms still die. They are ofcourse better but its no cakewalk - one must spam a lot of units to overrun AI.

Agree on that too. They are not a golden pill. That is why they are moderately priced. Just 800 each.

[Alpha Centauri Bear]

I did read parts of your readme. Chatting with you is more fun though. The problem with your logic is that you compare desert tiles with normal terrain. (In vanilla bad land doesn't matter that much. In your mod it does.) If you have a rolling rainy tile a farm (4 turns) makes it a 3-1 tile with the option of adding energy yield. A forest instead is 12 turns for a inferior tile. My point is that you would not forest desert tiles. You would avoid the area in the first place.

Don't quite get why vanilla and my mod are different in term of terrain yield? I din't change anything on that.
You still get good and bad spots in vanilla, right? And you still can place base on bad spot when all good ones are taken, right? And these poor bases still don't grow beyond size 2? You still would prefer to plant forest there regardless of the cost?
If you want to avoid placing base there it is your strategical choice. Forest is not that good on nutrients so its cost wouldn't change you mind about base in barren land anyway.

Forest primary purpose is mineral yield. It is a main competitor of the mine, not solar collector. As an alternative terraforming method it should be inferior to mine in general. You can build 0-4-0 rocky mine in 12 turns. Why 1-2-1 forest with its OP spreading capability should be any cheaper than that??? And its energy yield is clearly an additional bonus not related to its primary purpose that makes it even better.

Did you change self-destruct damage? Also I didn't say that the hit-and-run style combat in vanilla did not have counterplay. Of course it had. There were overpowered ways to go about it (helicopters) though.

What is self-destruct damage? Don't remember doing anything on that.

Let me reiterate it. The hit-and-run combat style does not have a counter play. Human can counter it. However, only human opponent is AI who is not aware of this exploit. It does not use it does not counter it on purpose.

[Alpha Centauri Bear]

By the way, guys. How do you like my Vader avatar? Is it too scary?
🤔

[Alpha Centauri Bear]

Version 46.
Here you go with more expensive Brood pit. Didn't move it earlier yet. It is still L10 (78% in tech tree). Where do you want it? 50%, 60%?

[Hagen0]

Self-destruct damage is one of the reasons you can't build a stack. Any unit can be self-destructed in the field (even worms I think ). They damage any unit on that tile and adjacent tiles for weapon strength * reactor level/2 rounded down. Units in bases are not effected. So self-destructing two fusion needlejets or a conventional missile and a laser needlejet will wipe out anything on adjacent tiles provided same reactor level is used. This is from memory from years ago so I can't guarantee it is perfectly accurate but should be about right.