Alpha Centauri 2

Sid Meier's Alpha Centauri & Alien Crossfire => Modding => Topic started by: Alpha Centauri Bear on May 19, 2020, 01:37:28 PM

Title: WtP SP costs
Post by: Alpha Centauri Bear on May 19, 2020, 01:37:28 PM
This is a discussion about SP cost in WtP. I'll start with excellent analysis by lolada.
Keep in mind that this is preliminary discussion and cost may change if we curb economical growth in here:
http://alphacentauri2.info/index.php?topic=21451.0



Early special projects

They feel good - those in first 100 turns - cost of 300-600 minerals and techs trade between AI. So usually factions get at least 1 special projects and player can snatch 1 or 2. Its expensive enough that player can't start and finish two at a same time on trascend - and thats great. I find it when I can get two of the projects i want game becomes super easy. (For example get HGP and Virtual world and drone problem solved easy game).
I like that you priced The Merchant Exchance at 200. Its fringe SP. The Command Nexus is strong, but 800 was way too much. Its 400 now.. but you are right that morale is strong. I wouldn't mind even 500 min cost there.
Conclusion here: early to mid plays fine.

Note 1: crawlers are unused here, maybe even not researched.

Mid projects

This is the area of 400-800 cost projects. I think it stills plays fine, bases gets 20-30 minerals and in general it takes 20-30 turns. Now one can get projects quite faster - for example Drones are notorious with 20% discount, but ok thats their strength. Presumably player can pick up +2 industry (Fund + Planned). Actually one can trick this and pick this SE config near the end of projects just to finish it and save 20% turns - say 20% of 30.. save 6 turns. Or switch to +2 industry (or more) and just rush it cheaper.


Note 2: Crawlers are available - and they can swing into player favor significantly (120 minerals is a lot). Apparently AI can use crawler - but i doubt its anywhere as good as player - i didn't notice they use it but i want watching carefully
Note 3: Energy rushing - again  Thinker AI can energy rush but i didn't observe it yet. I think it heavily favors the player.

Now its not that bad yet this can stay as is.. its hard to build crawlers and there is no many bases. Energy is still limited. And many AIs are alive and well and they compete. So lets say its fine.

Note 4: There's one offender here: The Cloudbase Academy - its 1200 minerals. Now thats a lot and its not good. Bases with 30 minerals are rare - at this point there's no even Genejack Factory or it just gets available and its hard to get 20+ minerals due to ecological damage. This means it takes 30-50 turns to build it.. reduce for good industry somewhat. Here's the problem: it takes so much time to build that player has ton of time to steal techs and prepare now 3-4-5 crawlers or more and build this project. I played 4-5 full games and i could build Cloudbase Academy in every if i wanted. AI actually got much faster to tech than me and they can't finish the project anyway.

I think this big issue - and it gets worse later. Now good thing is The Cloudbase Academy is not that good  ;lol Its good but Needlejets are awful for base attack. And they die in raids so its not like they are so great - one can build them without Aerospace Complex. I usually build 2-3 Aerospace Complexes for morale edge. So it turns out that its mostly good for Interceptor morale and for very late game Satellites. You moved them back to E11 - good move - so i often didn't even rush to get this special project. But for example in my AAR i got it anyway just because AI was taking so long to build it.

Note 5: If you had made this SP cost 600 or 800 then its totally possible AI would get it before player which would be good thing. I lost other cheaper SP becuase i couldn't get tech in time.. Gains got Superconductor, Drones got Manifold Nexus (i didnt even get the tech in time).. Angels got Theory of Everything - and i started that one - it took me some turns to amass crawler. Or as you said if you amass crawlers and spend on one SP you can't get other. But if its so expensive you have time for 3-4 waves of crawlers.

Late mid-game and late game


Here are 800-3000 special projects. Now these have that same issue, but there's more to it. First base mineral production does not go as fast.. you are lucky with 50 mineral base - and it does go to 70 in late midgame. Eco damage slows down mineral rise a lot.

That means that for example The Cyborg Factory (2000) takes at least around 30-40 turns. Depends on when you get it.. its quite possible to get those techs earlier and it takes 50 turns to build.. industry might help a bit. Now this is infinite time for player - one can always build this before AI. Singularity Inductor (+50% minerals everywhere) is same and there are a few.

Note 6: you have made Cloudbase Academy issue worse. AI will never get this projects unless player lets it. Its not fail all SP - AI will manage to build some 600-800 lategame secret projects (like Neural Amplifier, Nethack Terminus and thats great) - because AI do get these techs before player.

Note 7: game scale - now its not ok to build secret project for 30+ turns.. Its ok to build secret project 30 turns at mid game.. say the game takes 250 turns. If i build it at t150 i get to use it for 50+ turns thats ok. Even 20 turns is ok. But if I get SP at turn t200-t210 its not ok to build it say 30-40 turns. AI will never get any use of it and that perspective matters. Late projects should note take as long or longer to produce than early projects. Its game over, it should be quicker.

The result is in my 4-5 games so far and in many in vanilla/thinker. I don't build these projects at all - they are not needed, it takes too long. I build two for fun (I got one in AAR) and i could build them anyway whenever i wanted because AI will take forever to finish them. Finally for example Ascent to Transcedence - 5000 minerals. I think it is impossible for player to lose ^^ I started it in one game and Zakharov started it in base that needed 300 turns to complete it. I started it in 100 mineral base and i needed 50 turns - i completed it in 4 or 5 with crawlers.

I hope that was clear ^^. Now to some more disscussion and possible changes.

Leaving it expensive

- It has all downsides described above. Positive is that you force player to spend a lot of crawler or give up getting secret projects. But slowing player this was is not really possible - most will just quick the game and consider it won. I really don't like it because i feel i can win at any time once these OP project become available - i could monitor it but i doubt AI will ever rush these projects. So I'd like this changed.

Making it much cheaper than now

- My first idea was to limit costs a lot - thats one simple and likely good solution. AI is perfectly capable of building up 800 mineral projects in late game before player - I've seen that much. I think this mod is about challenge and trying up features - and good challenge is to let AI get many projects before the player.

Suggestion 1: limit all non-broken projects to 800 mineral cost.. 1000+ is too much for AI.. if player gets ahead its ahead let it be. But on transcend and WTP not many can do that - i think i play at decently high level. One could object but there's no point.. making it expensive doesn't work as explained above.

Suggestion 2: Broken projects

Totally broken: move to Tech 11+ or remove from the game.. removing is a shame i like what you did with Cloning wats just no so excessive - i suggest 1600 min max
Broken - move late and 800 - 1000 min cost to allow players to try them without tediousness

I made secret projects list below at the end for details.

Suggestion 3:

Removing rushing secret projects. Both energy and crawlers. I start to really like this. They can be built manually if they are not too expensive and its fair race then. Someone have better tech, someone has better industry.

alternative: just remove crawlers completely if you can't stop them from rushing projects

Suggestion 4:

Rushing with energy - 1. remove. 2. At least make it even more expensive. If the projects are cheaper in minerals - as they should be i think - then i would double the rushing cost to 8 energy per mineral.

Secret projects - changes and some comments

- Cloning Vats (totally broken) you can leave it at tech 13 and put something normal like 1600 minerals. People can try it for fun.. it is fun.. it takes 16 turns for 100 mineral base someone will build it that way.

- The Bulk Matter transmitter (strong, not broken) - 2 minerals per base - i don't know why you rated this as 3000 minerals - its not that great. its 4-5 mineral per base with % bonuses, it hardly matters lategame.. this can as well be 800 mineral Tech 10 project - it will be fun to see AI gets it before player

- Clinical immortality (broken in sense game over literally) - double votes - just should be late.. once you click this its game over anyway.. you can make it expensive.. now if you want AI to be able to win ever.. price it 1000 minerals not more.

- The Cyborg Factory (strong, kind of broken) - elites everywhere - should be maybe C11, 12 where its now but at  800 cost its fun but it takes time to put to use.. build units and do warfare. There's no point in pricing it high - just place it late enough, tech 10 can be gotten relatively early.

- The Manifold Harmonics (strong, broken) - this should be 800 and at least tech 8-10 in my opinion.. maybe not if you nerf that fungus we discuss

- The Nano Factory - (50% upgrade, heal) - its really great project both healing and upgrades are awesome - you priced it only 400 min.. this should be 600 at least

- The Network Backbone (broken) - having it at D13 or where it is does the job. But then no need to price it more than 800-1000. You won't get to use it more than 10 turns or so

- The Self-Aware colony (very strong or broken) - i never built it actually - its so late.. no need to price more than 800-1000

- The Space Elevator - same as above

- The Telepathic Matrix (strong) - now this one is weird - it comes late but could be used a bit for fun. Its not really needed because everyone solved drones by this point. To put it to use one should scrap all Recreation Commons/Hologram theatres and will save some money. In that sense its not even broke.. it would be broken in midgame.. So I don't see why price this more than 800.. and it can be at tech 9-10.. no need to move it later.

- The Ascent to Transcendence (5000) - if AI won't rush it it should be much cheaper.. 2k or something like that.. its mostly clicking end turn. Someone might even lose the race but i doubt it.


Ah that was a lot, but it was fun ^^. Maybe we can make experimental version and test some of these ideas - it looks good on paper to me.  But i have been wrong plenty of time so it will be interesting to see what you think.


Title: Re: WtP SP costs
Post by: Nexii on May 19, 2020, 01:50:57 PM
One thing about SPs is their value changes a lot depending on map size, like EFFIC SE. If you have 10 bases the value is much different than 30. I do wish they had put some kind of map scaling for SPs like they did with tech.

My opinion is there's really only 2 what I would call 'broken' SPs in the stock game. Weather Paradigm, and Cloning Vats. CVs you have right, I think I have them at around E12. WP I had requiring higher tech, B5 or so. I guess 600 cost kind of has the same effect.

I cap out my SPs at around base amounts, 500 or 600 at the top end. But I can see how they ended up doubled in cost or so with heavy crawler use/abuse.

I'm not sure how much the AI cancels in progress SPs in WtP, but in the stock game it does it a lot. I found that increasing SP costs actually degraded AI play.
Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 19, 2020, 02:07:15 PM
First of all, I've increased their cost for a single purpose: to let factions compete for them more fierce. Discovering it first should not guarantee building it first. Less advanced but stronger economy should have a chance too. It seems I achieved it. However, some late game projects got too expensive. I should adjust them based on actual economical strength at the time.
That is actually strange why late projects seem so difficult to build as I assume later project will be build using crawlers with the combined empire strength. Looks like this is not the case.

Now when we are talking about human vs. AI then former will always be more cunning and will find more exploits. That is why AI is given raw economical advantage on tougher levels. The best way to level that is to teach AI to do the same. Like rushing SP with crawlers and money in this case. However, this is pretty tedious and difficult task and result is not guaranteed either. Wrong tuning may render AI stupider. SMACX AI is mostly single turn heuristics. It doesn't carry plans like multi turns SP rushing preparation, etc. In this regard the only option to help AI would be to disable crawler rush. I believe it is possible to teach it money rush since it is single turn operation. This disables whole empire participation in SP building which is sad. I always imagined how people across whole empire contribute to a single epic building.

Another option would be to limit crawler/unit rushing by some portion of initial cost: 50-40-30-20%. This way player can still concentrate whole empire efforts on it but for hefty penalty. Essentially, this translates to building SP faster but for double-triple price and corresponding economical impact. I think this is fair. SP hurry cost need to be increased too to match it.
Let me summarize it. We can allow external aid for SP building but with severe penalty to make project choice more fair. Let's say you have three equally productive bases. You can build three projects (one in each) at normal speed risking to not finish it in time or you can build just one and let two other bases contribute at 50%, thus, rushing it twice as fast. Proportion can be adjusted.
Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 19, 2020, 02:10:39 PM
One thing about SPs is their value changes a lot depending on map size, like EFFIC SE. If you have 10 bases the value is much different than 30. I do wish they had put some kind of map scaling for SPs like they did with tech.

I hear you, man. However, it would be a revolution to implement something like that. Besides, it would not be easy just to correctly adjust their cost by map size. It would require tons of play testing. Unfeasible.
🙁
Title: Re: WtP SP costs
Post by: Nexii on May 19, 2020, 02:18:19 PM
I've seen the stock AI rush SPs sometimes. It's more apparent on bigger SPs like Ascent. Though it's inconsistent about doing it.

Yea rushing is a harder thing to fix. Maybe crawlers should turn in at 75% or something. Better than disbanding units at 50%, but worse than direct transfer at 100%.
Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 19, 2020, 02:51:43 PM
The Command Nexus is strong, but 800 was way too much. Its 400 now.. but you are right that morale is strong. I wouldn't mind even 500 min cost there.

Morale/defense SP do not scale that well as economical ones. It doesn't really make difference for them between 400 and 500.

Note 4: There's one offender here: The Cloudbase Academy - its 1200 minerals. Now thats a lot and its not good. Bases with 30 minerals are rare - at this point there's no even Genejack Factory or it just gets available and its hard to get 20+ minerals due to ecological damage.

You may have researched it too early by luck. Cloudbase Academy is L10 and Genejack Factory is L6.
I agree it is too expensive and not that important. May reduce the cost of AC as well.

I played 4-5 full games and i could build Cloudbase Academy in every if i wanted. AI actually got much faster to tech than me and they can't finish the project anyway.

That is normal when you have crawlers already. Then you can cherry pick.

Suggestion 1: limit all non-broken projects to 800 mineral cost.. 1000+ is too much for AI.. if player gets ahead its ahead let it be. But on transcend and WTP not many can do that - i think i play at decently high level. One could object but there's no point.. making it expensive doesn't work as explained above.

I plan to do that if we agree on reducing external help proportionally.

Rushing with energy - 1. remove. 2. At least make it even more expensive. If the projects are cheaper in minerals - as they should be i think - then i would double the rushing cost to 8 energy per mineral.

I believe they are already quite expensive to buy. 800 minerals project is 2400 credits. That is like 20 turns savings in mid game. Especially when we nerf economy in general. I consider it to keep it at x4 for hurry cost and to reduce mineral contribution to 50% for all units and crawlers. Then crawlers will be no more important for SP mineral rush as you can do the same with units.
We can try it out and adjust hurry cost later if needed.

- Cloning Vats (totally broken) you can leave it at tech 13 and put something normal like 1600 minerals. People can try it for fun.. it is fun.. it takes 16 turns for 100 mineral base someone will build it that way.

Agree. Actually, I was pondering whether I could nerf the project itself a little. Say to stop base size growth beyond nutrients/4. So one would need more than 4 nutrients per citizen to grow it further. Just a thought.

- The Bulk Matter transmitter (strong, not broken) - 2 minerals per base - i don't know why you rated this as 3000 minerals - its not that great. its 4-5 mineral per base with % bonuses, it hardly matters lategame.. this can as well be 800 mineral Tech 10 project - it will be fun to see AI gets it before player

Help typo. It is actually +50% (as any mineral multiplying facility). So it is broken.

- Clinical immortality (broken in sense game over literally) - double votes - just should be late.. once you click this its game over anyway.. you can make it expensive.. now if you want AI to be able to win ever.. price it 1000 minerals not more.

Agree. Same note on nerfing effect. Could be 25-50% instead.

- The Cyborg Factory (strong, kind of broken) - elites everywhere - should be maybe C11, 12 where its now but at  800 cost its fun but it takes time to put to use.. build units and do warfare. There's no point in pricing it high - just place it late enough, tech 10 can be gotten relatively early.

This is just a free facility project. It is no more broken than facility itself. I set default morale to Very Green now. So even with all facilities you get them at Veratans only and then it is hard to get promoted and easy to die.

- The Manifold Harmonics (strong, broken) - this should be 800 and at least tech 8-10 in my opinion.. maybe not if you nerf that fungus we discuss

Eh. It's pretty situational. You need to be locked at +3 PLANET to get most of it losing other opportunities. And even then with recently modified fungus yield it is not a threat to others anymore. Planet liking factions could seriously benefit from it, though.

- The Nano Factory - (50% upgrade, heal) - its really great project both healing and upgrades are awesome - you priced it only 400 min.. this should be 600 at least

Again, great but situational. Upgrade is limited - you can upgrade only what you have and it is just a discount. So some limited economical advantage. Healing is speed up just slightly. So you'll end up with 100% health instead on 80% and this advantage is only in a field. Definitely some combat and economical advantage but you won't crush twice as strong opponent with it. Units still keep dying often.

- The Network Backbone (broken) - having it at D13 or where it is does the job. But then no need to price it more than 800-1000. You won't get to use it more than 10 turns or so

Agree. I actually was completely at lost how to price it.

- The Self-Aware colony (very strong or broken) - i never built it actually - its so late.. no need to price more than 800-1000

Agree.
Also can be tuned to save not 50% but 25% or so.

- The Space Elevator - same as above

Not sure this one is broken. What do you get besides unlimited insertion range?

- The Telepathic Matrix (strong) - now this one is weird - it comes late but could be used a bit for fun. Its not really needed because everyone solved drones by this point. To put it to use one should scrap all Recreation Commons/Hologram theatres and will save some money. In that sense its not even broke.. it would be broken in midgame.. So I don't see why price this more than 800.. and it can be at tech 9-10.. no need to move it later.

I still think it is broken as it is a complete drone solution. Nothing in game should be absolute. Better be to multiply drone facilities effect of something.

- The Ascent to Transcendence (5000) - if AI won't rush it it should be much cheaper.. 2k or something like that.. its mostly clicking end turn. Someone might even lose the race but i doubt it.

Agree. I priced it that high to let the game continue even after all tech are discovered! I guess it is difficult to achieve.
🙁

Title: Re: WtP SP costs
Post by: lolada on May 19, 2020, 03:05:41 PM
Quote
First of all, I've increased their cost for a single purpose: to let factions compete for them more fierce. Discovering it first should not guarantee building it first. Less advanced but stronger economy should have a chance too. It seems I achieved it. However, some late game projects got too expensive. I should adjust them based on actual economical strength at the time.

Yeah i believe currently at your mod good pricing is 400-800 (early - late) minerals for competition. Over that player easily wins if he wants to. Player wins due to either crawler or energy rush. Pricing them less than 400 means first faction that techs SP will most often build it as well.

As you said.. if we now nerf economy, boreholes or minerals, growth it should be taken into account.
Quote
That is actually strange why late projects seem so difficult to build as I assume later project will be build using crawlers with the combined empire strength. Looks like this is not the case.

Its not strange ^^ they are not difficult to build with combined empire power, they are easy for the player. AI doesn't rush or rush rarely - feels to me like that. We don't have to change this quickly - this could be tested by watching AI; there's chance I am wrong here. But from my xp so far i had tons of time to build as many crawlers are wanted for those heavy 1000+ projects.

I think I read somewhere in Thinker thread that AI will energy rush secret projects in some cases - but its coded to let player win. We could ask Induktio maybe this can be changed. If AI rushes only near finish.. its no wonder they can't get those big projects.
Quote
Now when we are talking about human vs. AI then former will always be more cunning and will find more exploits. That is why AI is given raw economical advantage on tougher levels. The best way to level that is to teach AI to do the same. Like rushing SP with crawlers and money in this case. However, this is pretty tedious and difficult task and result is not guaranteed either. Wrong tuning may render AI stupider. SMACX AI is mostly single turn heuristics. It doesn't carry plans like multi turns SP rushing preparation, etc. In this regard the only option to help AI would be to disable crawler rush. I believe it is possible to teach it money rush since it is single turn operation. This disables whole empire participation in SP building which is sad. I always imagined how people across whole empire contribute to a single epic building.

Yeah its a bit sad.. but AI programming is a pain. I would try simple solutions. If people don't hate disabling Crawler project rushing that can be removed. Then energy rush can be a thing - it just needs to be expensive enough; provided that we can tell AI to rush projects with most of its money.
Now i picked up on rushing - I usually play with no rushing self-rule.. but its almost impossible to build those projects without crawlers ^^. I'd rather have it buildable in reasonable time and give AI chance to win.

Quote
Another option would be to limit crawler/unit rushing by some portion of initial cost: 50-40-30-20%. This way player can still concentrate whole empire efforts on it but for hefty penalty. Essentially, this translates to building SP faster but for double-triple price and corresponding economical impact. I think this is fair. SP hurry cost need to be increased too to match it.

This looks like half-measure on its own. If you make cheaper projects (up to 800 for example) and heavily restrict this rushing it might be ok-ish. Then player can spam ton of crawlers to get project faster or to just get 1 turn ahead of AI to win. Its still player only advantage likely but at a heavy cost. It also allows AI more space to win because player might not have that many Crawlers.

Quote
Let me summarize it. We can allow external aid for SP building but with severe penalty to make project choice more fair. Let's say you have three equally productive bases. You can build three projects (one in each) at normal speed risking to not finish it in time or you can build just one and let two other bases contribute at 50%, thus, rushing it twice as fast. Proportion can be adjusted.

I am for severe option or no crawler rushing. It doesn't even matter how much crawler cost - but i think its good at 120 minerals i wouldn't change it. Its valuable to crawl in mines and condensers, or borehole.

So if at 50% its 60 minerals.. and thats around 7.5-12.5% of the project. I'd cut it to 5% = 800*0.05 = calc.. 40 minerals. So if Crawler turns 33% resources into minerals - you get 40 minerals per crawler.

Player must use 20 crawlers to complete project from zero to 800. And its doable i can tell you that in AAR i have i dont know 20-30 bases (its normal map) and they get strong. Player can also build a project for say 5 turns, smack in few crawlers and energy rush the rest.
Title: Re: WtP SP costs
Post by: lolada on May 19, 2020, 03:23:56 PM
Quote
You may have researched it too early by luck. Cloudbase Academy is L10 and Genejack Factory is L6.
I agree it is too expensive and not that important. May reduce the cost of AC as well.

Yes - well AI got it and then they steal so much its crazy there's tons of probes. It took me quite awhile to get the tech and I don't know if AI even started the project. I got it in trade i think from Deirdre. Then sometimes AI start the project in weak base (they choose one base in top 1/3 of best.. they don't go for best base). So it can take them something silly like 60-80 turns to build such expensive SP.

Quote
I plan to do that if we agree on reducing external help proportionally.
;b;

Quote
I believe they are already quite expensive to buy. 800 minerals project is 2400 credits. That is like 20 turns savings in mid game. Especially when we nerf economy in general. I consider it to keep it at x4 for hurry cost and to reduce mineral contribution to 50% for all units and crawlers. Then crawlers will be no more important for SP mineral rush as you can do the same with units.
We can try it out and adjust hurry cost later if needed.

Yeah they are expensive at start and to midgame. Lategame there's so much energy i had 500+ per turn it goes up to 1000. So player can rush buy complete 800 min SP = 3200 energy; in 6-7 turns at 50% eco.

It would be interesting if this formula would scale somehow simply. SP initial cost/10.. so 800 projects 8 energy per min.. 400min project 4 energy per min. As simple as that. But its not really that much important i suppose.

About changing/nerfing specific project, we could have that discussion. Some sure deserved to be nerfed. Still its part of fun to have them strong - total drone solution - yes please.


Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 19, 2020, 03:51:19 PM
Yeah they are expensive at start and to midgame. Lategame there's so much energy i had 500+ per turn it goes up to 1000. So player can rush buy complete 800 min SP = 3200 energy; in 6-7 turns at 50% eco.

Then we have to reduce energy development. However, keep in mind that total energy collection grows with empire size. There is nothing you can do about it and that is why SP cost should grow somehow at least. 😜
Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 19, 2020, 03:54:43 PM
Here is preliminary plan. Conservative changes at first.

Title: Re: WtP SP costs
Post by: lolada on May 19, 2020, 05:21:02 PM
Quote
    Crawler help to project at 50% their mineral cost. That makes it same as disbanding units including crawlers themselves.
    SP hurry cost is x4 as now.
    Teach AI to hurry SP.
    Lower end game SP cost a little.

- Disbanding units gives back 50%.. Crap. Well 50% is step in right direction.

- About SP pricing cost - suggestion is that AI can get project in reasonable number of turns so that player is not guaranteed to win the race. That can be done with cheaper cost and teaching AI to rush. We just need to find a balance. If its late SP it must not take too long because the game actually ends.

- A little more than little for some ^^

I wanted to comment on few SP.

The Weather Paradigm - its value drops quickly when its moved later in tech - WTP already moved it to D3. I think this SP is overpowered with Thinker mode and E1 tech in vanilla, but in WTP is ok. Its pricey at 600 minerals and by the time you get it, it doesn't feel like a win. So i'd rate this project as good - but i'd rather have some other projects if i could.

The Cloudbase Academy - its not nearly as good in WTP as in vanilla because defense is so strong. Simply said Needlejets die ^^. Combat is interesting - Infantry attackers are actually back - that 25% vs base is  really important. Copters are also heavily nerfed in this sense - they simply cannot kill more than 1-2 armored units as they get badly damaged. They are useful to snipe several unarmored formers out of the base tho.
Satellites are very late tech so that part is of limited use as well.

The Hunter-Seeker algorithm - I think i overvalued this project a bit - its awesome, but there's but. AI uses lots of ench. probes once they get the tech and that counters it. Nethack Terminus also counters it - AI builds this SP. Both come later so Hunter Seeker is still great help.

The Virtual World - This project is exceptionally good - on border of broken good. Usually you want Network node in every base. One good fix is to make Hologram Theatres cheaper a bit and cost less maintenance. They are 60/3 now and every base needs them. Hologram theatre is basically Recreation Commons + 50% psych. Psych is great for big bases.

All Xeno - Green - Fungus projects - they are great for green factions and should not be cheap.



Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 19, 2020, 05:51:57 PM
The Virtual World - This project is exceptionally good - on border of broken good. Usually you want Network node in every base. One good fix is to make Hologram Theatres cheaper a bit and cost less maintenance. They are 60/3 now and every base needs them. Hologram theatre is basically Recreation Commons + 50% psych. Psych is great for big bases.

Guys, what's the big obsession with The Virtual World? It is one of these free facility unremarkable projects. They all should be proportional to facility cost/main. That's all.
Title: Re: WtP SP costs
Post by: Nexii on May 19, 2020, 06:24:14 PM
VW is quite strong because Holo Theatres are pricy for what they do. Not so much the construction cost but the maintenance. When you push back boreholes, 3E maint is kind of steep. The only real way to pay for them is +2 ECON SE, so you're kind of forced into that
Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 19, 2020, 06:53:53 PM
Hmm. Then this is Hologram theater problem. Should it be cheaper or lower maintenance?
It's a combination of Recreation commons (40/1) and +50% psych. The latter means that if you additionally spend 4 energy to quell 2 drones it makes it 3. Not bad but this usually comes later in the game. So in practice it quells 2-3-4 drones depending on game stage. I'd say it should be about 50% more expensive than Recreation commons. Or maybe it is designed to be later stage second tier facility when it is more useful.
Should I change it to 60/2. Will it be more fair?

If you browse through other psych improvement facilities you'll see they all are quite expensive especially in maintenance.
Hologram theatre (60/3)
Tree farm (120/3)
Hybrid forest (240/4)
Research hospital (120/3)
Nanohospital (240/4)
Title: Re: WtP SP costs
Post by: lolada on May 19, 2020, 07:09:48 PM
I think Virtual World is my favorite special project - its top 3 in the game. Basically the way i play the game is to spam colony pods up to a number and then grow cities. You can't grow cities past 3-4 really unless you have Hologram Theater. Then it takes a lot of time to build Network Node as well. If you can kill 2 birds with 1 stone its awesome. Especially due to how much maintenance it costs.

These are must have facilities in every base in standard game so its huge time saver. Hologram Theater is also kind of Prereq to get better Special Projects - higher pop = more minerals and SP. With low pop its not safe to build special projects. Lately i changed playstyle a bit so i don't focus research nearly as much - but its still standard good play.

So as to how to change it - well maintenance to 2. Remember we want to reduce tech pace a bit - i think its good to have some maintenance and 1 is too little for what it does. Virtual World certainly increases tech pace - thats part of why it is good.
Maybe it can be 1 row cheaper it makes some sense - its indirect nerf to project.

Other facilities (except recreation commons) are indeed expensive but they give other benefits.
- Research Hospital is important its +50% labs as NN.. + psych + 1 drone quell.. quite awesome for those 6-7 pop cities. It stops Prometheus Virus - thats is actually important as I place cities 4 tiles apart and virus has that range.

- Tree Farm, Hybrid have huge other benefits..
- Nanohospital is way overpriced imo i almost never build it.

There's no much point to science too much in late midgame.. Quantum smth +50% science is also kind of useless in SP unless you race to the max. At some point you get so much science that its better to build other things like army or probes, secret projects and what not.

Title: Re: WtP SP costs
Post by: Nexii on May 19, 2020, 07:26:25 PM
Yea but the others have big modifiers to ECON, nutrients, LABS. The psych is more of a side benefit. I made Holo theatres on the cheap side, 50/1 like Rec Commons. Compared to 2 sentinels at 40/2 (minerals), assuming 1 or 2 POLICE SE. Although at 3 POLICE or non-lethal methods, that price drops way down to 20/1. If you have both then it's even lower.

It's hard to say what they really intended in the early game. Police State SE was really bad. In Civ2 the early governments had good POLICE equivalents while energy was still low. So you'd build a Temple (Rec Commons) and use 3 police units to control the population. The next worker control facilities did require more tech. But then again so did the economic government.
Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 19, 2020, 07:38:55 PM
Would I satisfy everybody by decreasing HT maintenance and increasing VW cost? Would it close the gap and reduce the urge to have it?
Hologram theater = 60/1
Virtual World = 800
Title: Re: WtP SP costs
Post by: bvanevery on May 19, 2020, 08:01:13 PM
It doesn't carry plans like multi turns SP rushing preparation, etc.

So the AI would need to maintain a fleet of some amount of supply crawlers, to be used for whatever purpose in any given turn.  And it would need to decide each turn whether to replenish the fleet or not.  And that would be contingent upon keeping them safe, as otherwise it could lose all of them to strafing or invasion.

Quote
I believe it is possible to teach it money rush since it is single turn operation. This disables whole empire participation in SP building which is sad. I always imagined how people across whole empire contribute to a single epic building.

Wat, you mean you can't imagine all the Scrooge bankers signing little bits of paper as they deprive their employees of lumps of coal?  Think of the sacrifices everyone has to make in their accounting hovels.

You could go the other way with this and deny anyone the ability to Hurry a SP.  That would greatly change the game, but it's a valid game design decision.  The human player wouldn't have this "hoard a pile of money" ability over the AI.  On the other hand, one might rightly ask what piles of money are for.
Title: Re: WtP SP costs
Post by: bvanevery on May 19, 2020, 08:25:28 PM
I hear you, man. However, it would be a revolution to implement something like that. Besides, it would not be easy just to correctly adjust their cost by map size. It would require tons of play testing. Unfeasible.
🙁

It also doesn't make any kind of sense for early game SPs.  Nobody has a sprawling empire yet, they have to actually build those things.

Also, who says I want to win by building more than 20 cities, no matter the map size?  Who says I want to conquer?  When I conquer, who says I want to leave enemy cities standing rather than Obliterate them?  The risk in "tying cost to map size", is strongly biasing towards larger horizontal empires.  Which can be very tedious to manage.

The Netherlands used to be powerful.  They were small.  They had population density.  They got that because all the people who didn't like being oppressed by Catholicism, flocked to their more tolerant country.  Compared to the "map size" of Earth, they did quite well for themselves at one point in history.  In Civ terms they were a "vertical" empire, at least in their home territory.  Strong empire core.

The best way to understand SPs is they are a race.  Someone needs to win the race, and they get a reward for having done so.  It is not an actuarial calculation of what it would take to build all those buildings individually.  It's supposed to be an advantage.

Some advantages are clearly worth way more than others, i.e. free Aerospace Complex vs. free Command Center.
Title: Re: WtP SP costs
Post by: bvanevery on May 19, 2020, 08:47:09 PM
Agree. Actually, I was pondering whether I could nerf the project itself a little. Say to stop base size growth beyond nutrients/4. So one would need more than 4 nutrients per citizen to grow it further. Just a thought.

The most obnoxious thing about the Cloning Vats is not the growth, it's the IMPUNITY to Power and Thought Control.  It's a total giveaway.  If I were a binary modder I'd get rid of the latter.  In my own modding I've at least made it one of the most expensive projects.
Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 19, 2020, 08:48:51 PM
I agree with everybody that SP cost should not correspond exactly to its benefit. It should be placed correctly on time scale and then it should be buildable in 20-40 turns or so just to support competition. That's all.

I am using my way to calculate the price based on facility cost just to ease this weighting for myself. Otherwise, it would be just completely arbitrary judgement.
Title: Re: WtP SP costs
Post by: bvanevery on May 19, 2020, 08:57:48 PM
VW is quite strong because Holo Theatres are pricy for what they do.

So I made them cheap in my mod.  Because they're not interesting, they're grossly expensive to make, and grossly expensive to maintain.  It's going to a freakin' futuristic movie theater, the narrative on this is completely stupid.  You know what else was stupid?  The overpriced Colosseum in Civ II.  That's the play mechanic they're imitating here, this idea of sticking it to you if you want people happy.  Screw them.  Only costs 50 to build and 2 to maintain in my mod.

Since these are cheaper, VW is not exciting in my mod.  It can be useful, but it's not exciting.  Also, I have the HT and the VW available from different techs.  This is because if they come with the same tech, the AI will stupidly build HTs when it doesn't need to.  Even after it builds the VW those HTs will still be around.  So I put them on different tech paths.

I downgraded the AI weight in alphax.txt since it's not an exciting SP anymore.  Human Genome Project is always better, because it's not a building replacement SP.  There's no other way to get that +1 Talent.

Oh and BTW my Recreation Commons only costs 30 and 1 to maintain.  Instead of costing 40 in the stock game.
Title: Re: WtP SP costs
Post by: bvanevery on May 19, 2020, 09:06:52 PM
I am using my way to calculate the price based on facility cost just to ease this weighting for myself. Otherwise, it would be just completely arbitrary judgement.

I establish costs based on playtesting.  When things are too easy to get, I raise costs.  I don't usually raise lots of cost all at once.  It can take many mod releases to arrive at cost.  Most of my SP costs have been stable for a long time, but recently I shifted all the mindworm oriented SPs much later in the tree.  Some of them had their costs go up, to match the general trends in the tree.  The biggest things I picked on lately are the Neural Amplifier, the Dream Twister, and the Manifold Harmonics.  The NA and DT are clearly very powerful in combat in the real world.  MH I already had expensive, I just made it more expensive and moved it even later.
Title: Re: WtP SP costs
Post by: Nexii on May 19, 2020, 09:31:45 PM
Yea enduring unique benefits tend to be a bit better than free facilities. I'm noticing this in faction balancing a lot since I gave most of my factions a free facility of some sort. Although there is a limit, some facilities get too powerful to start with without huge downsides. And then when everyone gets that facility you're still left with those huge downsides.

Title: Re: WtP SP costs
Post by: bvanevery on May 19, 2020, 09:45:23 PM
Hologram theater = 60/1
The base cost is close to what I do, 50.  You don't need to be so generous on maintenance.  It removes 2 drones and gives +50% PSYCH bonus, it's ok to make people pay 2 to maintain it.

Quote
Virtual World = 800
By the standards of my own mod this would be completely silly.  I price the Cloudbase Academy and the Cyborg Factory that high.  However my mod has factory, crawler, and borehole only coming late game.  I don't know what the timing is like for your mod, because I gave up after mindworms trounced me.  My gut feeling though is this is completely absurd, unless it's your design intent to stop anybody from ever completing a SP earlier in the game.  In my mod, getting a 500 mineral project done mid-game is actually a challenge.  For me that's the Planetary Energy Grid.  You are unlikely to get all the important ones done, the AI factions are likely to complete some of them.
Title: Re: WtP SP costs
Post by: lolada on May 19, 2020, 09:47:41 PM
I like bvanevery reasoning here for i suppose almost everything including that tall vs wide example and race for the projects.

Neural amplifier and Dream Twister are not exactly game breaking - worms have their weakness - it helps to win the game without losing them 50% of the time ^^. I wouldn't touch that.

So suggestions:
- Hologram Theater 50/2 looks good
- Virtual World - 600 is plenty you still have to build all those Network Nodes.. (lucky Zakharov.. but he does need it the most).. and its less op now with cheaper HT

- About rushing - i doubt we can teach AI to be really competitive with rushing - i would second that removing Crawler rush of Secret Projects completely. Crawlers are still awesome for resource extraction. They can rush prototypes as well - not that i ever remember to use it - even when i needed that badly. If thats out of the question 50% effectiveness is better than 100% we have now.
Teaching AI to rush with energy is something i suppose - someone like Morgan can put it to good use and some other AIs sometimes have lots of energy.
Title: Re: WtP SP costs
Post by: bvanevery on May 19, 2020, 10:02:14 PM
Basically the way i play the game is to spam colony pods up to a number and then grow cities. You can't grow cities past 3-4 really unless you have Hologram Theater.

Sounds like you're too much of a spamming horizontal Infinite City Sprawl player.  The Bureaucracy penalty is there to punish people like you.  I routinely get cities up to size 5 without issue.  Some get jerky face, many don't.  I make Non-Lethal Methods available relatively early in my mod, so 2X police units come along soon enough to get me to size 5, with 1 per city.  One can always go Police State if one needs more early control over cities, although I tend to go Democratic because in my mod that's worth money.  Some factions of course are prohibited from choosing one or the other.  Some factions have inherent happiness benefits so city size is not such an issue for them.

I only play on Huge maps, which could make some scaling differences.  i.e. when Bureaucracy penalty sets in.  My mod is designed for Huge maps.  Standard maps are too small, they have no logistical challenge to them.
Title: Re: WtP SP costs
Post by: bvanevery on May 19, 2020, 10:12:09 PM
With low pop its not safe to build special projects.
False.  The enablers you're looking for are 1) a minerals special on Rocky terrain, 2) a Nutrient special combined with forests, 3) cashing in Artifacts to speed the SP along, and 4) spending money gained from popping supply pods out in the hinterland.  That said, my early SPs cost 300, not 800.  A 500 mineral SP is somewhat challenging to get done in my midgame.
Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 20, 2020, 02:11:18 AM
Released new version where AI hurry projects.

We can continue discussing project costs. However, we should also try how fast AI can build project with hurrying available and maybe it won't be that bad anymore.
Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 20, 2020, 03:19:25 AM
Reduced SP cost a little. Let me know if not enough.
https://github.com/tnevolin/thinker-doer#sp-cost-table

I clearly see that after 20,40,60,80 there is a jump to 200. Almost all end game projects are 200 or 300. However, it is also true that almost all of them are broken. That is why the cost.
Title: Re: WtP SP costs
Post by: dino on May 20, 2020, 07:48:53 AM
SPs should be a bit overpowered, this is the charm and a reward for beelining certain tech category, so I don't think it's a great idea to "balance" them like we are trying with regular stuff.
The problem araises only when the leading faction gotta catch them all, if they are are more distributed it just gives each faction some unique flavor.

So my gamist idea ( and SP are the most gamist thing in SMAC already ) is too keep the base cost of SPs more moderate but increase it by 25% for each SP a faction did build themselves.
I've just stopped to bother with even considering feasibility of actually implementing these ideas I'm throwing, it's not for me to judge ;)
Title: Re: WtP SP costs
Post by: lolada on May 20, 2020, 08:31:37 AM
Quote
Sounds like you're too much of a spamming horizontal Infinite City Sprawl player.  The Bureaucracy penalty is there to punish people like you.
Hm i should probably try to change the style a bit - i locked myself into this playstyle trying to outspam thinker AI i think. B-drones are a pain ^^ expansion in waves is certainly an option. I blame Democracy - once you want to grow it removes free minerals from new bases so i want to finish expansion before switching. Democracy is not even that good in WTP with 1 growth, i was more in fundamentalism for industry bonus and war troubles.

Quote
False.  The enablers you're looking for are 1) a minerals special on Rocky terrain, 2) a Nutrient special combined with forests, 3) cashing in Artifacts to speed the SP along, and 4) spending money gained from popping supply pods out in the hinterland.  That said, my early SPs cost 300, not 800.  A 500 mineral SP is somewhat challenging to get done in my midgame.

You are right i wasn't precise with that statement. 4-5 pop is enough with enablers, so nutrient and mine resource, one can get near 15 minerals and thats sufficient to get early SP. Its usually not hard to get 1 in will to power - i just picked up first SP in game with Morgan - and it was new Command Nexus at 400 min - i had 15 production with 4 pop. Its actually easier to bump up early production in WTP due to unlocked resource yields - i could use three 0-4-0 mines and go into eco-damage because i had 4 nutr and 5 nutrient tiles. For drones - two police units and recreation commons were enough (i switched to police state at start, its -1eff, -1 planet it doesnt hurt much). Good Nexus is cheap because i again ran into Drones and it took <10 turns until we were in war.. again.
About pods and artifacts - i am playing without them - maybe i should consider turning them back in. Vanilla AI was awful so i considered it only to players advantage - its big source of money/artifacts for the player.. and i remember often getting production boosts.

Quote
SPs should be a bit overpowered, this is the charm and a reward for beelining certain tech category, so I don't think it's a great idea to "balance" them like we are trying with regular stuff.
The problem araises only when the leading faction gotta catch them all, if they are are more distributed it just gives each faction some unique flavor.

So my gamist idea ( and SP are the most gamist thing in SMAC already ) is too keep the base cost of SPs more moderate but increase it by 25% for each SP a faction did build themselves.
I've just stopped to bother with even considering feasibility of actually implementing these ideas I'm throwing, it's not for me to judge ;)

These are good points. Its good to have them strong makes the game interesting - AC is about those great tech/human advances that radically change landscape. Making projects cheaper even if they are strong lets AI gets them - thats then strong AI - its attractive proposition to me. Broken one can be nerfed or moved to tech 10+. I was also thinking about making each consecutive SP more expensive slightly - not really 25% that seems a lot. But 10-15% would raise costs significantly after faction gets 4-5.
@tnevolin Is that possible? And is it dumb idea? ^^

Quote
Reduced SP cost a little. Let me know if not enough.
Released new version where AI hurry projects.
Well looking forward to try it - some playtesting will be helpful. I have my doubts, costs are still over 1000 but anything near 1000 is in realm of possibility for AI so its improved. Depends on how well AI will rush - i'll try to playtest some quicker games and to observe if AI rushes projects now.

Did you change crawler effectiveness? And did you teach AI to rush both with crawlers and energy? Or just energy and we presume AI rush already with crawlers?











Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 20, 2020, 12:49:33 PM
SPs should be a bit overpowered, this is the charm and a reward for beelining certain tech category, so I don't think it's a great idea to "balance" them like we are trying with regular stuff.
The problem araises only when the leading faction gotta catch them all, if they are are more distributed it just gives each faction some unique flavor.

So my gamist idea ( and SP are the most gamist thing in SMAC already ) is too keep the base cost of SPs more moderate but increase it by 25% for each SP a faction did build themselves.
I've just stopped to bother with even considering feasibility of actually implementing these ideas I'm throwing, it's not for me to judge ;)

They are still OP in WtP just slightly less than in vanilla. Essentially, their cost is nothing comparing to their benefit on normal map. I never tried to "balance" them. Just to make sure neither human nor AI can build them in less than 10 turns. Like I I used to in vanilla to build first four projects with artifacts only. That is too abusive.

You just argued against "balancing" them and proposed same exactly thing in your second paragraph trying to distribute them evenly across factions.
🤣
I don't know yet. It could be interesting.
Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 20, 2020, 12:55:04 PM
Quote
Reduced SP cost a little. Let me know if not enough.
Released new version where AI hurry projects.
Well looking forward to try it - some playtesting will be helpful. I have my doubts, costs are still over 1000 but anything near 1000 is in realm of possibility for AI so its improved. Depends on how well AI will rush - i'll try to playtest some quicker games and to observe if AI rushes projects now.

Did you change crawler effectiveness? And did you teach AI to rush both with crawlers and energy? Or just energy and we presume AI rush already with crawlers?

Only too OP ones are over 1500. However, there are quite a bunch of OP ones.
🙁
Anyway, let me know if it still too much even with AI hurrying them. I try to be conservative. There are so many tangled features in this game.

Didn't do anything with crawlers yet. It is more difficult change. I take it AI needs 2000 cash (30% less on Transcend) to buy out a project outright. This should happen quite often in midgame. So maybe even crawler help won't save you.
Title: Re: WtP SP costs
Post by: lolada on May 20, 2020, 01:33:18 PM
Ok time to playtest. I argue for and against   ;lol

If Ai gets too good new meta will be to let them build and take it by force.. there's that. Extreme version of that is ugly lets hope we don't get there - reminds me of some civilization games/version where its absolutely not good idea to build wonders - but to conquer them. Some you couldn't even hope to build before AI.

I think/hope these changes could be ok..  This will likely affect only mid to lategame where AI is weak and player takes over. Now if player can (at least sometimes) insta build rush and energy buy projects - its fair that AI do that sometimes.
Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 20, 2020, 01:38:50 PM
Ok time to playtest. I argue for and against   ;lol

If Ai gets too good new meta will be to let them build and take it by force.. there's that. Extreme version of that is ugly lets hope we don't get there - reminds me of some civilization games/version where its absolutely not good idea to build wonders - but to conquer them. Some you couldn't even hope to build before AI.

I think/hope these changes could be ok..  This will likely affect only mid to lategame where AI is weak and player takes over. Now if player can (at least sometimes) insta build rush and energy buy projects - its fair that AI do that sometimes.

I thought I made conquering hard in WtP. Are you still able to cut through AI defense easily on Transcend?
Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 20, 2020, 02:10:49 PM
You know what, dino. Why am I defending this SP cost at all? It's not like it is some cornerstone of my mod. It is not something difficult to change either. This is all in txt configuration.

Just let me know what you want them to be and we be done with it. Even easier you can change them yourself in alphax.txt until you are satisfied and then send it over for me to merge into my release.
Title: Re: WtP SP costs
Post by: lolada on May 20, 2020, 03:28:43 PM
I'd stop with changes for now - its all theory until we test it - so lets see how rushing works and new prices and there are other changes as well. Then tweak further this if needs be.


Quote
I thought I made conquering hard in WtP. Are you still able to cut through AI defense easily on Transcend?
Its hard ^^ and fun its good. No its not easy to cut through. Late midgame/lategame becomes easy - tedious because player gets those huge SPs.
Lets hope with these changes lategame is challenging as early - mid.
Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 20, 2020, 04:34:02 PM
Late midgame/lategame becomes easy.

Any thoughts as why it becomes easy? Did you just economically outgrow them or they cannot withstand your tactics?
Title: Re: WtP SP costs
Post by: Nexii on May 20, 2020, 05:26:12 PM
I think general 4x AI does better with small empires. Once you get larger a human gets much stronger and can overcome the production deficits. It doesn't see the big picture the way a human can.

I once requested a sort of 'scaling up' AI bonus as the game went on from Yitzi. But I guess it would have been very difficult to implement. The idea was to make the bonuses mild at the start, so factions like Spartans don't just crush you right away. But scaling up so the game doesn't become trivial by the midway point
Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 20, 2020, 05:45:33 PM
I think general 4x AI does better with small empires. Once you get larger a human gets much stronger and can overcome the production deficits. It doesn't see the big picture the way a human can.

I once requested a sort of 'scaling up' AI bonus as the game went on from Yitzi. But I guess it would have been very difficult to implement. The idea was to make the bonuses mild at the start, so factions like Spartans don't just crush you right away. But scaling up so the game doesn't become trivial by the midway point

I think SMACX already have this scaling up bonus on the high difficulty level. AI has 30% faster growth and production. Meaning when human doubles their economy, AI triples it. And this keep happening every next period over and over again, exponentially!
You will perceive it if you play as dumb as AI. The fact that you don't means that exploit you use are an order of magnitude more effective than sheer exponential growth.

Besides these huge bonuses AI should be taught to use them, actually. Otherwise, it may just waste them at times.
Title: Re: WtP SP costs
Post by: Nexii on May 20, 2020, 05:47:26 PM
No I mean as the game progresses, by turn or techs. Not by difficulty level
Title: Re: WtP SP costs
Post by: dino on May 20, 2020, 06:05:39 PM
You know what, dino. Why am I defending this SP cost at all? It's not like it is some cornerstone of my mod. It is not something difficult to change either. This is all in txt configuration.

Just let me know what you want them to be and we be done with it. Even easier you can change them yourself in alphax.txt until you are satisfied and then send it over for me to merge into my release.

It's fine, I was under impression that discussion is going toward increasing their costs even further.
It's your mod, don't design it completely by cometee. I'm unlikely to take it as it is anyway.
( but will try it again at some point for sure and I'm forever grateful for the changes to the combat you've made).
Title: Re: WtP SP costs
Post by: lolada on May 20, 2020, 06:37:55 PM
Quote
Any thoughts as why it becomes easy? Did you just economically outgrow them or they cannot withstand your tactics?

Outgrow them certainly - it takes time for player to overcome AI advantage - and it depends which faction you play, some are way easier than others. Once player achieve industrial parity - with same number of units human wins.. if you start to have more units than its overrun and going easy. So its snowball in that way and thats kind of expected and normal. Someone always snowballs - and if player is losing then one usually restarts the game ^^.
Its also tactics - because all those abilities give player edge. For example amphibious or drop make game easier - AI can't use it as well as player. Another one is mag-tubes.. i think AI should build magtubes even if player can use them, its lesser evil. Thinker disabled magtubes - i think thats mistake .. at least in WTP settings

Remind me of what Sid Meier talked in that GDC talk.. they envisioned game where civilization rises and falls (as in history) and you are supposed to rise again. And what happened is that noone would play it because players quit or reload -_-. Lel.

Anyway even when outgrowing in WTP it still takes time to conquer AI - you did great setup with combat - it really takes lots of time and effort to conquer the world. I've finished playing AAR -  just will need time to write it down so my struggles to bring down AI will be in pictures there^^. One thing that turns struggle to easy in the end is op superprojects - and i found i could get them all because they are expensive. If they were cheap AI would actually build them more often because i can't get tech in time - thats the logic behind my comments about price. Now with rushing that might change - and lategame will be harder. Cool.

Title: Re: WtP SP costs
Post by: Nexii on May 20, 2020, 07:00:27 PM
Yea for rise and fall you'd have to have more catchup mechanisms than ended up being in the Civ/SMAC series. That was more in line with how the Civilization board game was though. The larger your empire, the harder it got to keep cities up. There was a cap on the number of cities you could have. And as you got ahead other players would become a bit more wary to trade with you. It made for a very different game than one of exponential growth. Where SMAC feels like a sprint, Civillzation board game felt like a marathon.

There's a bit of that in SMAC, the leader tends to have a harder time getting treaties/pacts. And EFFIC hits you a bit as you grow. But then they put in the governor which is a huge benefit to a leading faction. EFFIC probably should have also applied to minerals (waste) like in Civ2. And specialists.
Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 20, 2020, 08:59:48 PM
No I mean as the game progresses, by turn or techs. Not by difficulty level

I meant the same. Both you and AI grow exponentially. However, AI exponent base is bigger at higher level. So it outruns you more and more as game progresses. This is not a difference in speed but in acceleration. Meaning if it is twice ahead of you by turn 50 it will four times ahead of you at turn 100 and eight times at turn 150. It will double relative advantage every 50 turns. (number are just an example)
Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 20, 2020, 09:03:20 PM
You know what, dino. Why am I defending this SP cost at all? It's not like it is some cornerstone of my mod. It is not something difficult to change either. This is all in txt configuration.

Just let me know what you want them to be and we be done with it. Even easier you can change them yourself in alphax.txt until you are satisfied and then send it over for me to merge into my release.

It's fine, I was under impression that discussion is going toward increasing their costs even further.
It's your mod, don't design it completely by cometee. I'm unlikely to take it as it is anyway.
( but will try it again at some point for sure and I'm forever grateful for the changes to the combat you've made).

I don't mind committee opinion in some areas at all. As you correctly said, this is pretty narrow purposed mod. Everything else is welcome. I like to even merge it with others if they find it beneficial. So by all means, share your preferred SP costs when you test them and find satisfactory.
Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 20, 2020, 09:10:08 PM
Someone always snowballs.

That's right. However, it was always difficult for me to snowball in Thinker. That is why I was curios how you beat it in every game. You might be more seasoned player than me.

Its also tactics - because all those abilities give player edge. For example amphibious or drop make game easier - AI can't use it as well as player. Another one is mag-tubes.. i think AI should build magtubes even if player can use them, its lesser evil. Thinker disabled magtubes - i think thats mistake .. at least in WTP settings

Yeah, yeah. However, this is why AI is given economical advantage to compensate for weakness.

Thinker disables magtubes??? I don't think it is disabled in WtP.

One thing that turns struggle to easy in the end is op superprojects - and i found i could get them all because they are expensive.

I guess this is what we are working on now. Next step could be actually limit their OP features directly.
Title: Re: WtP SP costs
Post by: dino on May 20, 2020, 09:28:02 PM
Thinker AI doesn't build magtubes at all, but with modified combat in WtP it definitely should, it would allow it to counterattack damaged invading force before it can heal.
Title: Re: WtP SP costs
Post by: Nexii on May 20, 2020, 09:31:48 PM
I meant the same. Both you and AI grow exponentially. However, AI exponent base is bigger at higher level. So it outruns you more and more as game progresses. This is not a difference in speed but in acceleration. Meaning if it is twice ahead of you by turn 50 it will four times ahead of you at turn 100 and eight times at turn 150. It will double relative advantage every 50 turns. (number are just an example)

In practice the AI doesn't keep the same acceleration is what I'm saying. Its doubling rate slows down due to inefficiencies in play tactics. A lot of it is in how it manages military movement and support. And also when to build facilities and which facilities. Or when to run growth to boom or +2 ECON safely. Thinker I'm sure has improved this with the AI using advanced terraforming, but nonetheless the rate will degrade. Whereas a human can maintain their rate, which while at first is slower will eventually overcome the AI.
Title: Re: WtP SP costs
Post by: lolada on May 20, 2020, 10:47:04 PM
Quote
That's right. However, it was always difficult for me to snowball in Thinker. That is why I was curios how you beat it in every game. You might be more seasoned player than me.

Hehe 1 - 0 for me :) Well its basicaly turn advantage from better management/rushing and as Nexii said better build orders, better efficiency. AI also sometimes chooses stupid SE - i see them ruin their efficiency. They build way too much military - but thats part of the challenge.. here more works for AI. AI with weak military is easy pray for player in vanilla or plain thinker mod.

Maybe one of the biggest advantages for the player is rushing and your mod made it even easier for facilites. Player can rush every recycling tanks, every recreation commons at some point - last x turns at least and thats ton of turns over AI. And eventually with base spam it all adds up. Then for example one can neglect research for industry for example and steal lots of tech. Combination of those works for me. For example changing facilities to 3 energy per mineral would make game visibly harder  - i don't suggest that :D

I've not even once seen magtubes by AI - Thinker taught AI NOT to build them - reason was: player uses it to steamroll AI - AI patfinding can't use it to full extent. AI ignores magtubes so it (i presume) stupidly goes of and lose movement points - but even that is better than no magtubes i think. You can't steamroll AI in WTP so they should be turned back on.

Before:
(https://i.imgur.com/6vaAtg5.png)
After:
(https://i.imgur.com/Ijs1w8T.png)
Looks like they rushed it with 200+ energy credits. They had 390 more so they could have rushed it earlier maybe but its ok i suppose. Cyb. Con. also was building project and was 3 turns behind Drones.

Secret projects in game so far. Btw i have weird bug.. maybe Fundamentalist+ is too long ?!

I've managed to get 2 secret projects with 4-5 pop capital and 14-15 minerals. Rushed last 3 turns both times.. I switched to +2 industry to finish second project.
(https://i.imgur.com/OZ9uJ4S.png)

q: Hm why did you put The Manifold Nexus in build tech? Somehow to me makes sense its on explore branch. This is second time Drones get it and they can't really use it (to full potential). Once i saw Morgan get it. Seems weird. It might be better idea to put somewhere where Deirdre and Cha-Dawn can get it.

Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 21, 2020, 12:06:05 AM
Thinker AI doesn't build magtubes at all, but with modified combat in WtP it definitely should, it would allow it to counterattack damaged invading force before it can heal.

Wow. Didn't know that. Probably I should enable them.
Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 21, 2020, 12:09:21 AM
I meant the same. Both you and AI grow exponentially. However, AI exponent base is bigger at higher level. So it outruns you more and more as game progresses. This is not a difference in speed but in acceleration. Meaning if it is twice ahead of you by turn 50 it will four times ahead of you at turn 100 and eight times at turn 150. It will double relative advantage every 50 turns. (number are just an example)

In practice the AI doesn't keep the same acceleration is what I'm saying. Its doubling rate slows down due to inefficiencies in play tactics. A lot of it is in how it manages military movement and support. And also when to build facilities and which facilities. Or when to run growth to boom or +2 ECON safely. Thinker I'm sure has improved this with the AI using advanced terraforming, but nonetheless the rate will degrade. Whereas a human can maintain their rate, which while at first is slower will eventually overcome the AI.

Exactly! It doesn't and lose momentum while it should crush human when it could. That is not a problem of everincreasing bonus but the AI who cannot use it.

More specifically, AI heuristic should adjust to game state, empire size, technologies, etc. Tuning AI is hard.
:(
Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 21, 2020, 12:15:03 AM
q: Hm why did you put The Manifold Nexus in build tech? Somehow to me makes sense its on explore branch. This is second time Drones get it and they can't really use it (to full potential). Once i saw Morgan get it. Seems weird. It might be better idea to put somewhere where Deirdre and Cha-Dawn can get it.

Tech tree tuning and correct placement of all features is a very-very-very hard task. I believe it is now somewhat better than in vanilla (where it is a complete mess). However, I gave up doing it alone. Only if someone team up with me.
Title: Re: WtP SP costs
Post by: bvanevery on May 21, 2020, 06:34:18 AM
Remind me of what Sid Meier talked in that GDC talk.. they envisioned game where civilization rises and falls (as in history) and you are supposed to rise again. And what happened is that noone would play it because players quit or reload -_-. Lel.

When you free yourself of the artificiality of "immortal empires", it begs questions about what it means "to win".  Especially in the modern era, where having everyone lose is a much more straightforward outcome for the human race.

In other thoughts, I've seen no evidence that the Network Backbone is actually a powerful project.  It sounds powerful, but I've never noticed my tech actually being boosted substantially for having built it.  And I do of course do the "super capitol" thing.

In other thoughts, after 2 years of course I think my techs are in all the right places.
Title: Re: WtP SP costs
Post by: lolada on May 21, 2020, 09:03:30 AM
Quote
Tech tree tuning and correct placement of all features is a very-very-very hard task. I believe it is now somewhat better than in vanilla (where it is a complete mess). However, I gave up doing it alone. Only if someone team up with me.

Yeah its complicated it needs to be thought through,. Its just my initial observation - i didn't know if you had some special reason to place it in build branch. That said i've no idea if it would be better or worse if moved elsewhere  :-[. I'll suggest something if i see that it could help. Its too early for me to give good advice.

I find it interesting how things turn out when they are weirdly placed or priced. For example 800 min The Command Nexus was hard to build and it took quite a lot of time - so initial reaction is this is wrong. But it had effect to always be there in early game, meaning AI (or player) would switch to it and didn't waste production due to lack of tech when their SP is completed by someone else. So they then chain build second tier of projects one The Command Nexus was done or just switch in-between. That was convenient.. maybe one AI loses ton of minerals in the end its hard to predict.
Still when you take positives and negatives i think its better that its cheaper - it lets you finish SP and move on do other things, there's plenty of things to build in early AC.

The Manifold Harmonics SP was op - but it is out of the way of Planet factions so its nerfed that way balancing it. Basically chances are they won't build it, but they can conquer it. It also stops player from easily getting it. I've seen these factions use it to some potential with  0 or +1 positive rating - from what i've seen they won't switch to Green.. player can switch but there's big penalty in Green. When you take that into account this project loses on power.

Now fungus is changed so that needs to be tested first. I am using it with Morgan its 3-0-3 tile (i have +1 energy SE). Unexpected thing is they way you set up SE i am using Democratic/Planned/Wealth and have tons of pluses including +1 Planet (i landed near manifold Nexus). So for example its perfectly possible to have good use of the project with Morgan.

Quote
In other thoughts, I've seen no evidence that the Network Backbone is actually a powerful project.  It sounds powerful, but I've never noticed my tech actually being boosted substantially for having built it.  And I do of course do the "super capitol" thing.
Yes good point, for example I've never build the project. We talk about thing as we know them, but some were never tried. I can see it giving like 100+ science - its strong. But then thats like one extra strong city (only energy..) - my best cities at endgame approach or go over that number. Maybe its not even op - but i never built it so i don't know. Thats why playtesting is the best way to try things. As alternative its possible to ramp up one city to high potential quite quickly in endgame - maybe even easier than to build this project if its highly priced.

Quote
In other thoughts, after 2 years of course I think my techs are in all the right places.
This would be interesting to see. I still didn't try your mod - i got into this aar and wtp changes - but I'll see to try it sooner rather then later, it looks interesting.

Title: Re: WtP SP costs
Post by: dino on May 21, 2020, 09:26:40 AM
Thinker AI doesn't build magtubes at all, but with modified combat in WtP it definitely should, it would allow it to counterattack damaged invading force before it can heal.

Wow. Didn't know that. Probably I should enable them.

You have bad memory, you are being made aware of this at least third time ;)
Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 21, 2020, 01:37:02 PM
In other thoughts, I've seen no evidence that the Network Backbone is actually a powerful project.  It sounds powerful, but I've never noticed my tech actually being boosted substantially for having built it.  And I do of course do the "super capitol" thing.

On normal map you probably get somewhere 100 for commerce and 200 for other nodes on the planet. So 300 research. Not bad addition to like 1000-2000 total research you have in end game but not game breaking, of course.

That is why it is not that insanely expensive for end game project in WtP but I can make it even lower.
Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 21, 2020, 02:01:43 PM
Yeah its complicated it needs to be thought through,. Its just my initial observation - i didn't know if you had some special reason to place it in build branch. That said i've no idea if it would be better or worse if moved elsewhere  :-[. I'll suggest something if i see that it could help. Its too early for me to give good advice.

Thinking through is not a hard thing. I did this all the time. The biggest problem is that everything is inter dependent with everything. Any small change causes everything else to change. Fixing one thing breaks others. There won't be absolute perfections.

Therefore, I dropped all the unimportant conditions like "game lore" and "research path" because they are difficult to define. I only make sure that certain features appear at certain time and some features are also chained (like weapon and armor). Everything else is more or less arbitrary. I try to keep "game lore" whenever I can but its not a definitive factor for me.

That said the "lore" and "paths" could be slightly improved if one would spend tons of time trying different combinations. It is not possible to just set them without breaking everything else.

I also welcome any suggestions on moving techs and stuff around if they are currently doesn't feel in place. Some isolated changes can be done.

The Manifold Harmonics SP was op - but it is out of the way of Planet factions so its nerfed that way balancing it. Basically chances are they won't build it, but they can conquer it. It also stops player from easily getting it. I've seen these factions use it to some potential with  0 or +1 positive rating - from what i've seen they won't switch to Green.. player can switch but there's big penalty in Green. When you take that into account this project loses on power.

I don't think it is OP. It requires +3 PLANET to double fungus production and this have multiple drawbacks. Even with such rating it'll turn fungus from something 2-0-2 to only 3-1-4 in mid game which is nice but not broken. It is most time useless for PLANET impaired factions, etc. Not a universal weapon. I'd like to see someone getting it and grow from inferior to superior nation.

Title: Re: WtP SP costs
Post by: Alpha Centauri Bear on May 21, 2020, 02:03:04 PM
Thinker AI doesn't build magtubes at all, but with modified combat in WtP it definitely should, it would allow it to counterattack damaged invading force before it can heal.

Wow. Didn't know that. Probably I should enable them.

You have bad memory, you are being made aware of this at least third time ;)

Every time it is a surprise.
😂
Title: Re: WtP SP costs
Post by: bvanevery on May 23, 2020, 12:55:05 AM
Fixing one thing breaks others.

Yes, but when you do it enough times you get better at it.  You have memorized everything's effect much like the back of your hand.  It's like some metagame where your job "playing SMAC" is to move all these tech puzzle pieces around.

Quote
There won't be absolute perfections.

False.   ;lol  It just consumes a rather large amount of developer time, over a long period of real calendar time.

Even now I find myself shuffling a few techs around.  I realized a way that I could get Sky Hydroponics Labs to come later, by decoupling the notion of "spaceflight" from the notion of lobbing Planet Busters.  Who says that's space travel?  Could be a cruise missile skimming the surface.  So now I'm going back to a design I did a lot earlier, where Sky Hydroponics Labs is a late tech that I don't have to make excuses for where it's placed.  And a prereq for the other satellites.  I'm not 100% thrilled about that, as I wanted Geosynchronous Survey Pods to come 1st, but I can make that sacrifice to the UI gods.

Quote
Therefore, I dropped all the unimportant conditions like "game lore" and "research path" because they are difficult to define.

I actually compose upon that dimension simultaneously with all other concerns.  In a pinch I'll hand wave / do a lesser job with lore, but I try hard not to.  Sometimes it just takes me another month to realize how I can solve game design and lore problems at once.  Like just after I've released my latest mod version.   ;lol
Templates: 1: Printpage (default).
Sub templates: 4: init, print_above, main, print_below.
Language files: 4: index+Modifications.english (default), TopicRating/.english (default), PortaMx/PortaMx.english (default), OharaYTEmbed.english (default).
Style sheets: 0: .
Files included: 31 - 840KB. (show)
Queries used: 14.

[Show Queries]