SMAX - The Will to Power - mod

[Hagen0]

Those are two different observations. I didn't express that well.

Combat outcomes felt completely arbitrary. (I understand they aren't really.) By giving the winner of each combat round a bonus you inreased the volatility of combat. You can easily see the same moderately close fight won by each side taking hardly any damage. It also has the effect that you have a hard time wearing down units. I lost 2 Empath rovers vs the same 1 HP fungus tower since my planet rating was low.

The armor issue is a matter of taste. I don't like the way you can't remove units sitting in your territory. On the other hand, it does give a boost to Morale as experienced units are worth more than in vanilla and it forced me to build more units.

[Alpha Centauri Bear]

Hmm, 3% energy per effic isn't too bad. It might need a little more than that. Not much though any more than 5% would be overkill

Corrected numbers. I didn't think straight before.

1. Fixed guaranteed income: 25%.
2. 0-50% for EFFICIENCY, 1/2/8 = 1/16 = 6.25% per level.
3. 0-50% for HQ proximity.

This way HQ has 100% at 0 EFFICIENCY. So no energy loss at the game beginning. At the same time the worst case (far base, -4 EFFICIENCY) is dropped from 50% to 25%, which I believe is more modest.

It would fix the EFFIC scale. Right now my SE set doesn't allow for lower than 0 for that reason.

What do you mean? You cannot physically set it lower?

75% efficiency at any distance from the HQ at +4 EFFIC is a lot. I do think Thinker AI and various speedruns demonstrate how strong ICS is. That horizontal development (base count) is more to blame for the energy/research explosion than vertical (facilities, SPs).

It may change overall pace of the game one way or another but what more important is relative change per each level of effect. In vanilla change from -3 to -4 was just absolutely unbearable. Whereas the one from +3 to +4 was just unnoticeable. I hope this new system will level them up a little.

75% at each base with +4 EFFICIENCY may be a lot but try to get to it. One would need to sacrifice a lot of everything else. So it's a fair price.

Rather than hammer far bases with terrible energy or mineral production, I do wonder if dynamic colony pod costs would work. That is the cost of a pod goes up by some amount of minerals for every base & colony pod you have...to match how citizen costs go up in nutrients

Yes, that is a good suggestion. I always thought reducing the price of non combat units with reactors is uncompensated. Why do so if overall production grows?
How do you suggest it should happen? Just proportionally to game time? Like double in 100 turn, triple in 200, ... ? Wouldn't it effectively stop further expansion especially on large maps?

[Alpha Centauri Bear]

Combat outcomes felt completely arbitrary. (I understand they aren't really.) By giving the winner of each combat round a bonus you inreased the volatility of combat. You can easily see the same moderately close fight won by each side taking hardly any damage. It also has the effect that you have a hard time wearing down units. I lost 2 Empath rovers vs the same 1 HP fungus tower since my planet rating was low.

Yes it should feel less predictable with added randomness. That what randomness is for. I can assure you that it doesn't impact you on a large scale. Meaning if you attack fungal towers 10 times in a game you average losses stay the same. Whereas, PLANET rating **does** affect your average losses.

I am sorry if you feel frustrated about losing units in appeared sure cases more than you get happy when your units win in impossible cases. You can crank my randomness configuration parameters in thinker.ini to make it look more like vanilla.

The armor issue is a matter of taste. I don't like the way you can't remove units sitting in your territory. On the other hand, it does give a boost to Morale as experienced units are worth more than in vanilla and it forced me to build more units.

Are they enemies? Can you bombard them? If they are friends - ask them to leave politely. 😉

[lolada]

We probably just need to remove whatever AI restriction left to let human win and this should do the trick.

Based on numbers above it seems that they rushed it way past 50% completion. Did they?

I don't understand why you want to limit this again after you just proposed it. This is the last project. You should be able to overtake AI by far by this point as you did in previous games.

Not appreciated you need to take comments more seriously. Ok maybe its me - i need to better explain it - i presumed you would know what i mean.

This is about the argument bvanevery used - players have expectations. This is violating it ***heavily***, you can only go so far.  Now there's some tolerance, we get used to something, so it takes a while to accept change, and we get used to that and its ok. This is for example your combat.. its more random. but for example i got used to it after 2-3 games and its good.

Now... Do you want to force player to always insta rush The Ascent to Trascendence or risk losing the game? I presume no.
Do you want players to delay and bank 15-20k ducats before completing Voice of Planet.. so they can insta rush The Ascent on turn 1 to deal with insta loss?
Is it ok to expect from player to have to rush last project? Personally i really don't like it.. its like saying to win you need last tech +20k gold or you lose.

Some backwards AI with pathetic production instantly completes the game winning project for 4k gold. That is not good design... there are many things wrong here, here's some:

- AI gets access to project it shouldn't have (vanilla allows everyone to build this project once Voice of Planet is done)
- AI heavily cheats with rushing.. like i sad he rushed 90+% production with 4k gold... while the player needs many times more
- its the ultimate project - what is allowed for other special projects should not be allowed for this one - and there is good reasons for that i hope i don't have to explain that

Now don't confuse this with AI rushing projects - its a great change - it makes the game better, AI is competitive. Player rushes the projects, so can AI. Its a fair game, player rush one, AI other. One can't get them all - and its fun to fight strong AIs.. 

There are reasons why this works, there are restrictions:
- faction must have a technology!!! Its a big thing.
- faction must have energy.. and energy is limited during most game duration! (even if AI cheats)
- it has consequence! (spent energy means no energy for rushing other things.. can't rush units of followin secret projects)

The Ascent to Trascendence breaks these rules:
- no tech requirement: anyone can build it.... eghm its ok for vanilla for its terribad AI
- yeah AI needs to have energy but it cheats its ass of.. and at late endgame there's ton of energy
- no consequences.. insta win button

So The Ascent to Trascendence needs to be somehow dealt with differently to be reasonable thing. Some suggestions:
- require tech to build it.. then its fair game if player wants to risk AI rushing it.. (fine by me.. its easy solution)
- disable rushing it completely (its 30 ish turns or less fine by me)
- limit rushing it (fine also..)
- or limit rushing every project if you want.. (my original suggestion, fine by me..)
- nerf transcendence AI rushing cheat _significantly_ (i wouldn't do this.. AI needs help for getting lategame projects)
- you can leave it be - but thats lazy thing (i really wouldn't like this)

Why not fix it properly when so much other things were already improved?

ps. no harsh feelings haha ^^

[Nexii]

Yea that would be about right with the corrected numbers. -4 EFFIC is about half the energy output of +4 EFFIC. Which is roughly in line for the same range in SEs like INDUSTRY, RESEARCH. What I meant is that my SE choices don't currently give -EFFIC. Any more than -1 EFFIC is devastating and I used it as a faction penalty.

Colony pod cost would be something like
30 + X * MAPFACTOR * (# of bases + # of colony pods)

X would be some constant. Yea there is the issue of higher reactors reducing the cost a lot. It's a similar issue with crawlers and in some ways it's why I preferred Yitzi's cost formulas. Also unit costs can't grow too much due to native life always having constant cost

Re: Ascent. It was always kind of broken but few games went so long due to imbalances that it was never a huge concern.

I think it's more fitting with the lore to have no tech requirement. As essentially all factions 'ascend' you just become a less dominant part of the ascended entity if you don't win the race. So probably no rush for Ascent makes the most sense. VoP IMO should have granted some bonus to Ascent production - Space Elevator/satellites kind of idea.

[Hagen0]

In the current form of the game of course you buy the Ascent to Trancendence in one turn. I appreciate why you don't like it but that is true for all Secret Projects. The way you can use your whole faction to pre-build them is somewhat silly.

Regarding the combat. I don't mind unpredictable combat winners that much. It's the way the winner often takes low or no damage that I really dislike. As I said you can for the same battle have the attacker win and take no damage or the defender win and take no damage. It's very hard to plan for and early game with few units it can screw you over. Overall, the current combat implementation is worse than the vanilla one.

[lolada]

In the current form of the game of course you buy the Ascent to Trancendence in one turn. I appreciate why you don't like it but that is true for all Secret Projects. The way you can use your whole faction to pre-build them is somewhat silly.

Regarding the combat. I don't mind unpredictable combat winners that much. It's the way the winner often takes low or no damage that I really dislike. As I said you can for the same battle have the attacker win and take no damage or the defender win and take no damage. It's very hard to plan for and early game with few units it can screw you over. Overall, the current combat implementation is worse than the vanilla one.

No its not the same - there's big difference between regular secret project and game-winning project. If its the same why then allow everyone to build it without tech  It can't be treated the same.. it should not.

Tim is right about combat from what i can observe - in average you win.. he has formula you can't argue math and win really. You can sometimes lose unexpectedly or win unexpectedly - thats randomness, but with proper preparation you will consistently win. If you want to take base with one 1-3-1 defender bring in 3 attackers you will take it like 95%+ of time. Yeah it will fail sometimes.. if you are unluck and deal zero damage on first attack - smart thing is to retreat or risk losing all of them. Take note that healing is nerfed heavily - if you bring enough units they will break through anything. I bet several 4-1-1 units can break though 1-24-1 defender due to chance.. and he can't heal quickly. Then smart solution is to bring 1-2 artillery units - really helps reducing casualties.

Vanilla taught us often one xp 4-1-2 is enough to take early bases.. with 5-6 of these experienced player can conquer the map. In WTP just to random chance you will lose them all in 20+ fights. And its  a good thing.

[Nexii]

Voice of Planet creates fungus, increases ecodamage and gives lifecycle bonus. I don't think the devs really had instant Ascent in mind considering those mechanics. It was supposed to be more of a race

[Hagen0]

I'm sorry but you really don't understand the problem I have with the combat. It's not that you can win or lose unexpectedly. It's the way the damage dealt to the victor is so volatile. The bonus for winner of the last combat round makes combat outcomes more extreme, making results that should be extreme outliers (like the weaker unit winning without taking damage) commonplace. If you are attacking with weaker units this is particularly problematic. If you attack say with 2 vs 3 odds you can easily lose 3 units in a row without making a dent or the first one can win without problem. It's too much.

[lolada]

Yeah you are right - what you expect is model that simulates consecutive battles. If defender wins 1st combat in 2nd he should be weaker... ie. tired. Its more complicated but realistic

[Alpha Centauri Bear]

Not appreciated you need to take comments more seriously. Ok maybe its me - i need to better explain it - i presumed you would know what i mean.

What are you talking about? I was genuinely excited by your test results. No sarcasm was intended. It is sad if you understood it differently. I hope your feelings are not hurt.

For long time few modders me included tried to make AI more competitive and still didn't reach the level where it beats the human consistently. This one was an amazing discovery that was overlooked by others. Thank you for your idea!

This is about the argument bvanevery used - players have expectations. This is violating it ***heavily***, you can only go so far.  Now there's some tolerance, we get used to something, so it takes a while to accept change, and we get used to that and its ok.

This is the same argument Sid explained in his video about modern game design. Actually pleasing customer is not a modern concept at all but it seems Sid, being mathematician and slightly out of this world, discovered it for himself in quite late age. Everybody else in the industry knew that all along.

I know what is customer expectations. Throughout my career I used to collect them and formalize them into a business requirement documentation all the time. I should note that, with all proper respect to you and other users, you won't find there words like "violating" or "heavily" or "tolerance" or any other emotional colored words. Just because it is impossible to convert them into code. This may sound Star Wars episode I - ish but it literally "does not compute".

I would definitely like to hear your and others opinions. That is why I asked these questions. Let me repeat them move formally.

1. Generally speaking, do you expect to certainly win on highest difficulty all the time? If not all the time then what percentage?
2. If not at highest level, then what is the difficulty level you expect to win all the time?
3. Do you expect win the race for The Ascent to Transcendence all the time? If not all the time then percentage or times?
4. Do you expect AI to get to the level when it competes for AtT? Do you prefer it to compete but still get it to yourself?

Now... Do you want to force player to always insta rush The Ascent to Trascendence or risk losing the game? I presume no.
Do you want players to delay and bank 15-20k ducats before completing Voice of Planet.. so they can insta rush The Ascent on turn 1 to deal with insta loss?
Is it ok to expect from player to have to rush last project? Personally i really don't like it.. its like saying to win you need last tech +20k gold or you lose.

Some backwards AI with pathetic production instantly completes the game winning project for 4k gold. That is not good design... there are many things wrong here, here's some:

Come on, man! It was just first try of new feature. I didn't even know what could happen and neither did you. That is what play testing is for. Now when we know it we can sit and think if this is something we want. You are immediately attacking it as if I designed and tested whole game and you bought it and now you are not satisfied. I believe we are both designers in this case.

- AI gets access to project it shouldn't have (vanilla allows everyone to build this project once Voice of Planet is done)

What do you mean? AI built AtT before Voice of Planet is done? Then this definitely a bug and need to be looked into. Please send me a save.

- AI heavily cheats with rushing.. like i sad he rushed 90+% production with 4k gold... while the player needs many times more

This looks like another bug. I didn't understand first the completion percentage. However, you said 11 production * 101 turns to completion = about 1100 minerals. So it should cost about 4k credits. Please send me a save to look into it.

The Ascent to Trascendence breaks these rules:
- no tech requirement: anyone can build it.... eghm its ok for vanilla for its terribad AI
- yeah AI needs to have energy but it cheats its ass of.. and at late endgame there's ton of energy
- no consequences.. insta win button

So The Ascent to Trascendence needs to be somehow dealt with differently to be reasonable thing. Some suggestions:
- require tech to build it.. then its fair game if player wants to risk AI rushing it.. (fine by me.. its easy solution)
- disable rushing it completely (its 30 ish turns or less fine by me)
- limit rushing it (fine also..)
- or limit rushing every project if you want.. (my original suggestion, fine by me..)
- nerf transcendence AI rushing cheat _significantly_ (i wouldn't do this.. AI needs help for getting lategame projects)
- you can leave it be - but thats lazy thing (i really wouldn't like this)

Why not fix it properly when so much other things were already improved?

You think this is vanilla bug? I never experienced AI to build it out of order but this need to be checked.

Everything can be fixed we just need time and testing and post analysis. Don't expect it in a single iteration. A little patience. 😉

[Alpha Centauri Bear]

Colony pod cost would be something like
30 + X * MAPFACTOR * (# of bases + # of colony pods)

Didn't get your MAPFACTOR. I though you proposed to make them more expensive with time. Which is a great idea in my opinion.

Re: Ascent. It was always kind of broken but few games went so long due to imbalances that it was never a huge concern.

I think it's more fitting with the lore to have no tech requirement. As essentially all factions 'ascend' you just become a less dominant part of the ascended entity if you don't win the race. So probably no rush for Ascent makes the most sense. VoP IMO should have granted some bonus to Ascent production - Space Elevator/satellites kind of idea.

I think this is by design and intended to help **human** player in first place. So they don't feel behind the race and can compete for this project with all the crawlers they can build. Since vanilla AI does not hurry SP with credits and uses crawlers poorly, human had all the chances to beat it even if being greatly behind in development and technology!

[Alpha Centauri Bear]

In the current form of the game of course you buy the Ascent to Trancendence in one turn. I appreciate why you don't like it but that is true for all Secret Projects. The way you can use your whole faction to pre-build them is somewhat silly.

I agree that pre building them is silly. However, I believe putting all your faction efforts into it makes sense.
I don't know how to disallow prebuilding (either with crawlers or cash). It seems to be a big embedded game feature.

Regarding the combat. I don't mind unpredictable combat winners that much. It's the way the winner often takes low or no damage that I really dislike. As I said you can for the same battle have the attacker win and take no damage or the defender win and take no damage. It's very hard to plan for and early game with few units it can screw you over. Overall, the current combat implementation is worse than the vanilla one.

Unfortunately it is how probability and multi round combat work. Damage (un)predictability is mathematically equivalent to winning (un)predictability. The more unpredictable the damage to victor the more unpredictable who wins.

Let's say two units with relative strength 2 and 1 are fighting. How badly do you want weaker unit to be damaged if it wins? 80-90% all the time? What if they both have 1 HP left?

I'm sorry but you really don't understand the problem I have with the combat. It's not that you can win or lose unexpectedly. It's the way the damage dealt to the victor is so volatile. The bonus for winner of the last combat round makes combat outcomes more extreme, making results that should be extreme outliers (like the weaker unit winning without taking damage) commonplace. If you are attacking with weaker units this is particularly problematic. If you attack say with 2 vs 3 odds you can easily lose 3 units in a row without making a dent or the first one can win without problem. It's too much.

Here also replying to your similar claim in another post.

You are absolutely right that my formula increases combat outcome volatility. This is what it is for and this is an intended effect. 😜
Before you start criticizing this let me tell you that this was done specifically to counter vanilla extremely low combat outcome volatility. Yes, vanilla designers did this on purpose to implement the same exactly functionality you advocate in your posts. They solved their "phalanx cannot beat the tank" Civ1 problem but opened can of other worms.
Well, some of WTP mod players thought that it did it too much probably to an abusive level making combat exceptionally not volatile. So there was a huge-huge discussion about how much of that stolen volatility should be re-introduced back into WTP. It is very difficult to tell where the sweet point is. So I left it as an adjustable parameter. And you will notice that many people there play WTP with their own setting of this parameter. So WTP is somewhere in between Civ1 and vanilla in this regard. You don't need to criticize the idea itself but you can propose a sweeter point on a Civ1-Civ2 scale if you don't like the current one. Remember, you can always switch it off completely.

Better yet, review all what have been said on the matter to not repeat same questions.

[lolada]

What do you mean? AI built AtT before Voice of Planet is done? Then this definitely a bug and need to be looked into. Please send me a save.

No I mean i built it.. AI doesn't even the tech. But it then happened that AI rushed complete AtT at 90+% for cheap (i am not sure anymore) and won the game. What this mean is that if player wants to win - you delay Voice.. prepare crawlers.. money and have to rush it on turn 1.. or you really risk losing the game.

I was genuinely excited by your test results. No sarcasm was intended

Haha well i considered that.. shortly lol. My bad.. i thought you just glanced over it like its nothing. its kind of important imo.

This looks like another bug. I didn't understand first the completion percentage. However, you said 11 production * 101 turns to completion = about 1100 minerals. So it should cost about 4k credits. Please send me a save to look into it.

Now I am confused  Ah I'll see later to check saves again and i'll upload them. The project is 3000 minerals.. I don't remember what Pirates Industry was... they might have used several crawlers and then rushed with energy.. maybe they rushed it multilpe times -_- weird. I don't understand now how they could even get to 1000 mins to completion their production was so crap. I though Transcend AI can rush cheaper?! Oh crap.. gonna take another look at saves.. something doesn't add up here.

What do you mean? AI built AtT before Voice of Planet is done?

- No.. I meant now in your mod it is not ok anymore to give AI access to AtT, like in Vanilla. But maybe scratch that.. need to check saves again.. i thought AI can rush-buy SP for cheap. If AI pays 4 gold per mineral its a fair game..
Still its a bit weird.. for example 3000 minerals with 40% discount is 1800x4 = 7200 energy to fully rush AtT - thats all you need. Its cheap for insta win lel ^^ i had 7k gold just AI just caught me sleeping.

1. Generally speaking, do you expect to certainly win on highest difficulty all the time? If not all the time then what percentage?
2. If not at highest level, then what is the difficulty level you expect to win all the time?
3. Do you expect win the race for The Ascent to Transcendence all the time? If not all the time then percentage or times?
4. Do you expect AI to get to the level when it competes for AtT? Do you prefer it to compete but still get it to yourself?

1. No not ^^ actually its really reasonable to lose early game if spawned close to AI. I had a few close calls. I almost lost diplomatic victory - AI got double vote SP. Once player has enough space its hard to lose - maybe due to this transcendence rushing thing. It surprised me.
2. I feel i can win 90+% on Thinker and below  . Thinker seems way easier than Trascend. Its possible to lose real bad early start.
3. Actually this trashed my expectations. I think player can win AtT race every time once it knows AI can insta rush it. Basically you just prepare crawlers and money and win instantly.
4. Well I somehow expected that 11 production city can't complete AtT.. my bad. I would actually prefer to have to build AtT completely manually without Crawlers and energy rush, that would be real race. But AI would need to know to use the best city and industry settings for that feat.. i guess thats out of scope.

[lolada]

Let's say two units with relative strength 2 and 1 are fighting. How badly do you want weaker unit to be damaged if it wins? 80-90% all the time? What if they both have 1 HP left?

In general i like combat and got used to it. it happens sometimes, very rarely, that for example 16-1-2 just dies to 0-1-1 former dealing zero damage  There is so many fights.. even if it is 1/200 its kind of notably since disparity is so big. i just take it for granted i know it can happen.. i played Diablo 2 ^^. Every few months you get that 1 in a million item or rune. Lottery.

-One could use trickery to check damage levels and guarantee that Former gets hurt certain %.. but i don't know if thats good idea or would introduce bugs or predictability.
1. Something reasonable - tweaking formula a bit:

Say that former (rough terrain+territory) has 10% chance to damage 16 attack rover and it hits it actually.. So Rover is damaged 10%. Next round former odds are reduced to 5%.. It hits it again..- unlikely but it will happen. Rover is damaged 20%. Next round 2.5%... next 1.25 .. and then if odds<1 round to zero. So even in very unlikely scenario Former will get 4 hits and damage Rover to 40% and at that point Rover is guaranteed to hit the Fomer. So its 40% to 10% atm.. Once this happens just restart loop and goes same to starting 10% odds again. In this case Former can kill Rover.. very unlikely.. but it has to take at least 20% damage.. Also implicates that 90% already hurt former must lose vs healthy rover. In that sense combat would be predictable.
 Ie. 1% HP Worm in WTP combat could not really kill healthy attacker like ever with this combat logic... it can now, i've seen Morgan smashing 5 needlejets into my 50% damaged locust and losing all of them before finally killiing it in 6th attack. I also lost 2 full hp rovers with miriam at (-10%) attacking 10% hp worm - and thus losing freshly settled base.

Basically one could eliminate the chance completely that Former 0-1-1 can kill 16 dmg Rover without taking damage. Similarly this would lead to things like that 4-1-2 attacking 1-3-1 defender guarantees some % of damage. That looks interesting and fun. I bet there is some downside to this .

2. Or some mix... take your combat now - if one units wins n in a row (say 3..) reduce odds 50%.. so you actually get a scenario where one unit cant cleanly kill the other under certain odds. This would mean that superior units will pretty much always be at least 10% damaged on iteration 4, 5, or 6 of combat before it resets.. But it would be more realistic to see unit hurt a bit after every combat.

3. Alternative idea is to give attacking/defender units penalty for fighting multiple times in same turn. Say 1-3-1 unit gets attacked and kills attacker taking zero damage. On following attack (on same turn) it gets 10-20% penalty.. so it gets weaker. On 3rd attack is 20-40% weaker. This should lead to guaranteed damage - death if attacked multiple times. Attackers like Rovers and Copters then likely should be nerfed similarly. This is probably way more complicated than suggestion 1. But I think hagen0 expects something like this.