Age of Wonder 4: what i would keep and trash from AOW3 and planetfall.

  • We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.
I totally agree. Thanks for that, i'll definitely check it, it will surely of a huge help. I'll try to contact the modders, it'd be great if a joint modding can be done, surely the AI improvements can be better and finished faster
 
  • 1Like
Reactions:
Why not also compare them with Shadow Magic? Some opinions (using all DLCs):

1)Mods vs Enchants : Enchants didn't have unit type restriction (very few did, like stone skin ban on flying units). Mods do give passive stats on top but I think most enchants could swing combat much more than the average. I doubt the devs are ever going back to the AoW2 Wizards so I don't think there is any chance of enchants returning though. AoW3 enchants don't count here because they were just tactical spells. AoW4 would be better with an enchant alternative than none at all.

2)There is way too few summon spells. So far it seems like factions can summon a tier 1 and maybe another tier 1 from their secret tech if it has one. This is way less strategic than using your resources to sneak a tier 3 into a fight in the last moment or fortifying a city with summons imo.

3)Free buildings: Buildings shouldn't be free. Why aren't higher tier buildings limited by Cosmite like higher tier units? This encourages spamming cities instead of having few strong cities spamming strong units in aow3.

4)Leaders\heroes are too similar to each other now without casting points. You can't take a goblin sorcerer with you and hope he manages to get off a lightning strike cast to turn the battle. Instead you get whatever and jam him into a tank to never see him again. or use him as a tank healer.. Everyone get to the laser tank it is our only healer!

5)RNG hit system.. You either love this or really hate it\tolerate it.

Finally I hope if they are re-introducing sectors that they either reduce the number of sector upgrades or give them their own production queue independent of units \ structures. Introduce sectors that don't give anything at all sector upgrade wise to make it harder to decide which areas give the best cities. If it is too much then allow forts \camps to mine whatever replaces energy \ research like AoW3.

Things I really want to be gone in next game: buildings that let you get Prime units without ever stepping foot into a battle. Units I used a lot in combat don't matter because a fresh recruit is going to be the same power.. if not even stronger than they are with higher tier mods on top. This was also in AoW3 and it wasn't great there either. Please limit new unit training to bronze\silver unless you manage to get a strategic building to get levels in AoW4!
 
  • 1Like
Reactions:
I'm not sure that's mostly because of the mods, i haven't try any AI mods though, but being attacked by full 5-12 stacks (that is 30-72 units full of modded t1 and t2) by turn 22-30 happen regularly in my game without AI mods, the most extreme is if i'm surrounded by 4 AIs, they often attack with 14-20 full stacks from 4 directions, each with 1k-1,2k power, happens usually around turn 25-35. This is why i (and few other posters in this forum) often complain about late game AI instead of the early game one, this is also why for now my focus on my AI mod is improving their use of higher tier units then later for them to mod those units.

What i found out so far by tinkering with the modding tools, the AI plans after mid game are quite weird, they still use early game plans at a turn where said turn is defined as mid or late game, by default the game consider mid game at turn 30+, late game at 55+, by default the AI seems to have different perspective regarding this though. I suspect this is part of the reason why they only use tier 3 and tier 4 in quite high number after turn 150+ (i never saw it myself, but looking at Ericridge pictures), there are several other reasons like unit production, research and unit upgrades priorities and the shifting between plans is influenced by chance. Personally the level or depth of access for the game resources is less than i expected, but i still think it's possible to create a decent if not good AI, it's however quite tricky for now, that being said i experienced decent amount of success so far in regards of making the AI use t3 and t4 much earlier than default, but too extreme for now, so i'm still figuring the right values for the variables

That is correct but becuase of how I play i have a tendency to face really numbers of enemies on regular basis. Its one of reasons why I keep on stating if AI's able to produce quality units faster and earlier than normal. Their production bonuses will have to be toned down greatly. Unless you feel great joy in deploying a bunch of unmodded t1s and t2s to square off eighteen laser tanks equipped with Plasma Disintegrator Module and Laser Precision modules equipped plus a random module at minimum. What would happen to human players? The human players will always get slaughtered every single time. No matter what. There's just no overcoming certain amount of technological advantage. My guess is that the devs is trying to nail down the teching speed balancing + resource gathering income speed. Planetfall ain't Civ5.

I am happy to see that you have finally noticed it.

I have a couple more screenshots to show.

20210103191636_1.jpg

20210103192722_1.jpg

If a battle happens early enough, usually only T3s on field will be your heroes that have taken up advanced piloting skill. Its one hell of a lifesaver for me. Especially when you're unlucky like I am stupid t4 guns never drop for me super early :p Yet i see AIs parading with them all the damn time. Makes me kill them and take fancy guns off their cold dead hands tho.

My hero driving a laser tank in that screenshot has Plasma Disintegrator Module, Laser Precision, and Hard Impact Modules equipped. Then for the troopers, they have Rail Accelerators, Interlocking Armor, and Purification Field equipped. The battle occured in the forests right next to my capital. I had to seize the initiative to attack because I would've been at a rather grave disadvantage trying to fight it out in the city.

Do note that the Laser Tank equipped like that costs a beefy 64 cosmite if I remember right. Energy is variable. It depends on city's upgrades. :p Coming out of a city with lvl 4 energy sectors if early on and lvl 2 energy upgrade isn't researched from tech tree yet so the pain isnt bad but. 64 cosmite is real cost. Losing even just one is a huge setback. Because it now requires reinforcements and takes away from other areas that also need your attention.

That's right. This isn't the only battle I fought.

My Main army is away from capital, doing battle against other armies. Also being sent from green kir'ko. It is composed of 18 Units, of three which is laser tanks my leader included in that. It also has three monitors supporting plus various autonom units with just one unmodded justicar with few troopers and assault bikes. Its my primary murder all enemies army group.

Wait, there's more. I also have another army group. This one's main task is to keep the angry shakarans in check. It was previously t1-t2 crew with only one laser tank piloted by a hero. Then it also have another hero with light saber. *cough* I mean sorry, a laser sword with Plasma Disintegrator Module XP They managed to get reinforced very recently with three laser tanks and three assault bikes.

But wait, there's even more~!

In city of Bodifahl I have another group of mostly cannon fodder but with some laser tanks for support, they is still unmodded. It takes time to get more cosmite. So Each turn I buy is more cosmite I gain. There's 24 units in it, a mix of troopers, assault bikes, purifiers, and three unmodded laser tanks. One modded laser tank was looking after them but had to move towards my main army group to replace the destroyed laser tank.

It's a delicate balancing act to ensure I don't suddenly get steamrolled in one direction just because I don't have sufficient T3 numbers to put a stop to it.


The situation is gradually improving, i've been at war against four AIs at same time for most of this game. Two of them is dead now. I just managed to never let them catch me at same time otherwise the sheer weight of numbers would've just forced me off planet new haven.


edit: with crazy fights like what i'm experiencing I haven't seen any need for the improved AI mod. I doubt i will be able to even survive the murderhorde if AI suddenly got smart.

Oh and this is for the other guy who couldn't see the entire picture, make sure to mouse over the picture image and then you will see a grid shaped icon in bottom right. Click it, it'll push the bottom menu bar down and let you see the entire picture mostly without requiring to open a new tab.


One more edit: I think part of my difficulty is because of a combination of traits I select for my favorite Ericridge the Great character aka me.

I pick Vanguard + Veteran train so I automatically just plain anger most kir'ko in existence by merely existing I think. They thirst to feast on my flesh and drink my blood. There's nothing more than that they want dead.
 
Last edited:
Ericridge,
I don't see the problem. Correct, it is not civ and it means you don't need tech superiority to beat AI. I am playing 12 teams gane with that mod which makes AI 4 times stronger than extreme AI. True, I am fighting tier 3-4 units on turn 50 while I am last in science ranking and second to last in military ranking, but the AI is still rather stupid in tactical battles (leaving turrets area when under siege, not moving some units in outside battles) and does not mod its units properly. I have killed 3 AIs already and while it is hard/fun, it is rather doable. Just a simple example: I have 4 amazon huntesses, evolved tier 3 animal and commander, they kill everything without losses: 7 shakarn fliers in one fight, 2 tacticians, 3 firebrands etc. In another fight, 15 units with a leader in third etc. AI stupidly attack my animal with 10 defence , "return 50% damage back" mod, regeneration mod and "resurrect after death at 40% HP" mod while basically ignoring my other units. I guess the reason is that all huntresses are prime and thus have 15% evasion while the animal is large and can't use covers. They don't care about anything else it looks, just accuracy is important for them. They deal 3 damage to the animal and return 2 back to them, killing themselves. None of my huntresses/leader have the "return damage" mod. It's rather silly.

Thank you for the hint about viewing images!

Wait, why do you have just 289 research and 11 operation points? If you have completed all future techs, there should be 14 operation points. If you haven't completed research yet, why is it so low? Here is my game at turn 53 vs 4 times stronger AIs with never ending wars. You can see I have more operation points and more doctrines.
40A92B964B9B92788248839397665E7701AB1F9E
 
Last edited:
Wait, why do you have just 289 research and 11 operation points? If you have completed all future techs, there should be 14 operation points. If you haven't completed research yet, why is it so low? Here is my game at turn 53 vs 4 times stronger AIs with never ending wars. You can see I have more operation points and more doctrines.
40A92B964B9B92788248839397665E7701AB1F9E

Because I lost the science city to green kir'ko.

It was a choice between keeping the science city or keeping my capital city. I chose to let the science city go. And it was a wise decision, my leader got killed at least twice.

I didn't have that much choice in the matter. I was at 2-3 cities for most of the game. Around turn 30ish, I was shooting at 24+ oathbound dudes trying to take my capital. And I was fighting blue kir'ko to keep control of four energy sectors, and then I didn't have troops needed to retain control of the science city so I took what I had out of the crimson fastness city and moved them to bodifahl. Which was just two owls. And i lost control of bodifahl and regained it several times over the game.

Which naturally leads to the result of having low science income.

I just understood that I had to hang in there no matter what. Gradually I forced a opporunity to take out the oathbound who has been obessed with me for the whole game and took him out for once and all which gained me +4 cities instantly but in the process I lost control of my capital to green kir'ko who seized the moment to smash it I also lost Bodifahl at same time to the blue kir'ko. But with oathbound dead I was free to retake my capital and retake bodifahl and made the fronts much more stable.

I didn't know that losing my capital meant i had to rebuilt my influence/cosmite structures again. Now i Know. First time it happened to me.

My lands was a major battlefield for my enemies :p Almost every few turns I was fighting more than 30+ units from at least two opponents in total. So I could get home, and then get started and wage four major battles and it takes up several hours and then I have to log off for the day and I progress only 1-3 turns xD

It didn't help that the oathbound cities I took had terrible location for science tho. His cities was set up for energy/food. Even if I tore them down and set up science, they would've been lvl 2 at most. Much more easier to take the slow approach than to spend precious cosmite on colonizers. I wanted to set up a new science city but I needed the cosmite going into the units more.

Its yet another moment of ballooning colonizer costs royally fucking me over.
 
Wait, why do you
Because I lost the science city to green kir'ko.

It was a choice between keeping the science city or keeping my capital city. I chose to let the science city go. And it was a wise decision, my leader got killed at least twice.

I didn't have that much choice in the matter. I was at 2-3 cities for most of the game. Around turn 30ish, I was shooting at 24+ oathbound dudes trying to take my capital. And I was fighting blue kir'ko to keep control of four energy sectors, and then I didn't have troops needed to retain control of the science city so I took what I had out of the crimson fastness city and moved them to bodifahl. Which was just two owls. And i lost control of bodifahl and regained it several times over the game.

Which naturally leads to the result of having low science income.

I just understood that I had to hang in there no matter what. Gradually I forced a opporunity to take out the oathbound who has been obessed with me for the whole game and took him out for once and all which gained me +4 cities instantly but in the process I lost control of my capital to green kir'ko who seized the moment to smash it I also lost Bodifahl at same time to the blue kir'ko. But with oathbound dead I was free to retake my capital and retake bodifahl and made the fronts much more stable.

I didn't know that losing my capital meant i had to rebuilt my influence/cosmite structures again. Now i Know. First time it happened to me.

My lands was a major battlefield for my enemies :p Almost every few turns I was fighting more than 30+ units from at least two opponents in total. So I could get home, and then get started and wage four major battles and it takes up several hours and then I have to log off for the day and I progress only 1-3 turns xD

It didn't help that the oathbound cities I took had terrible location for science tho. His cities was set up for energy/food. Even if I tore them down and set up science, they would've been lvl 2 at most. Much more easier to take the slow approach than to spend precious cosmite on colonizers. I wanted to set up a new science city but I needed the cosmite going into the units more.

Its yet another moment of ballooning colonizer costs royally fucking me over.
Well, now I am not sure why you play vs extreme AI (if you do) and then complain that it will be impossible to beat extreme AI if it gets stronger. 2-3 colonies is what I have before turn 10 and 5 colonies is my goal for turn 20, you play in a very suboptimal way. Why do you have a single science colony? It is not civ where you stack wonders in a single city. It is optimal to play in a way where you can lose any colony (including capital, yes) and still win. That means thousands of science/energy, hundreds of cosmite, dozens of influence, multiple armies. And not on 100+ turn, but around turn 50-60.
My current game is hilarious, I haven't built any non-secret tech units yet which are researchable and just look at the armies and income at turn 60. 4 times extreme AI still does not have any mods on tier 3 units and I have seen just 1 tier 4 unit. AI definitely needs an improvement.
615F90F17768797E04DCAB1650B1DDDD9E71B977
 
Last edited:
Because anything below extreme would be too easy for me. Extreme is sufficiently tough yet beatable for me. I don't know about you but I strongly suspect if you were penned in by one super close AI neighbor with two AIs allied + keeping watch with their armies ready to move in the moment you move out you would be struggling too. The map I'm playing on at the moment is a galactic empire map. Difficulty is significantly higher than a normal RMG map.

The two kir'ko was in a alliance, they went oh we met you. Ok now you die!
 
Because anything below extreme would be too easy for me. Extreme is sufficiently tough yet beatable for me. I don't know about you but I strongly suspect if you were penned in by one super close AI neighbor with two AIs allied + keeping watch with their armies ready to move in the moment you move out you would be struggling too. The map I'm playing on at the moment is a galactic empire map. Difficulty is significantly higher than a normal RMG map.

The two kir'ko was in a alliance, they went oh we met you. Ok now you die!
Not really. I killed first AI before turn 15. I lost a sector of capital to another AI while doing so. The trick for early game is to build roads between all your colonies even if it means attaching a bad sector. My games are harder when there are no adjacent opponents because then I cannot double my colonies by early conquest.
I stopped playing empire mode because only a few first planets are hard, after that you start with so crazy mods/units that AIs have no chance even if allied.
My point was if you struggle with default extreme AI, you cannot judge if AI needs an improvement, it works fine for you as is while I stopped playing because of boredom before being reminded about the great mod I am using. The game is rather enjoyable now, but the mod is required for that.
 
Last edited:
I find it a bit annoying that a potentially useful thread with an interesting topic is being ravished by some who are discussing way less interesting things.
I am sorry, but I believe adding new features to the game is a bad idea as long as AI does not use them or is bad overall. If you play aow3, you may know that AIs never use builders meaning no roads, no forts. Planetfall AI never replaces mods and does not mod tier 3-4 units until very late game. Using these abilities by human player feels like cheating.
Don't get me wrong, I like new features. But if it means AI is unable to effectively attack another continent, water should not be in the game because it causes more disappointment and complains than fun and good reviews. My derail started with disagreement regarding "AI is fine as is and should not be buffed".
 
Last edited:
  • 1Like
Reactions:
After tinkering a bit with the mod, i'm quite staggered by the possibility it can achieve.

Now i'm probably right that the reason why the dev decided to stay silent while there are many things people deems lacking for the current game is because of the possibility the modding tools alone provides

Here is the example:
Why not also compare them with Shadow Magic? Some opinions (using all DLCs):

1)Mods vs Enchants : Enchants didn't have unit type restriction (very few did, like stone skin ban on flying units). Mods do give passive stats on top but I think most enchants could swing combat much more than the average. I doubt the devs are ever going back to the AoW2 Wizards so I don't think there is any chance of enchants returning though. AoW3 enchants don't count here because they were just tactical spells. AoW4 would be better with an enchant alternative than none at all.

2)There is way too few summon spells. So far it seems like factions can summon a tier 1 and maybe another tier 1 from their secret tech if it has one. This is way less strategic than using your resources to sneak a tier 3 into a fight in the last moment or fortifying a city with summons imo.

3)Free buildings: Buildings shouldn't be free. Why aren't higher tier buildings limited by Cosmite like higher tier units? This encourages spamming cities instead of having few strong cities spamming strong units in aow3.

4)Leaders\heroes are too similar to each other now without casting points. You can't take a goblin sorcerer with you and hope he manages to get off a lightning strike cast to turn the battle. Instead you get whatever and jam him into a tank to never see him again. or use him as a tank healer.. Everyone get to the laser tank it is our only healer!

5)RNG hit system.. You either love this or really hate it\tolerate it.

Finally I hope if they are re-introducing sectors that they either reduce the number of sector upgrades or give them their own production queue independent of units \ structures. Introduce sectors that don't give anything at all sector upgrade wise to make it harder to decide which areas give the best cities. If it is too much then allow forts \camps to mine whatever replaces energy \ research like AoW3.

Things I really want to be gone in next game: buildings that let you get Prime units without ever stepping foot into a battle. Units I used a lot in combat don't matter because a fresh recruit is going to be the same power.. if not even stronger than they are with higher tier mods on top. This was also in AoW3 and it wasn't great there either. Please limit new unit training to bronze\silver unless you manage to get a strategic building to get levels in AoW4!

1) can be partially achieved with the current mod tools. I mean there is currently two type of strategic buff towards stack, one is lasting until said stack enter combat, the other is duration based. The tools allows modification to the latter, for the former, it's probably possible but tricky. The only thing the tools can't replicate (as far as i know) the enchant system of AoW2 is it can't introduce strategic buff to single unit, however can still be achieved by the player by just making a stack has one unit

2) while i agree, it can be achieved with mod too

3) this is interesting idea, fortunately it can partially be achieved by the modding tools. I mean the tools can make building cost cosmite, but cannot make it has cosmite upkeep, perhaps there is a way on this, but if there is it will be tricky

4) whilst i agree, the tools once again can partially achieve it.

5) probably can be achieved too because there is some resources related to units's properties like "accuracy", but i haven't checked directly the tools relating to those properties

The points below number 5, is surprisingly can be partially achieved too. The tools allow quite a huge modification to what sector exploitation defaultly do, like specification buildings tied to them, and probably even change sector exploitation upgrades do.

How much a newly recruited unit rank bonus have certainly can be modified by the tools, i have tinker with this

Another surprising thing i found is there is a possibility that the size of a sector can be changed, like to total area (in hexes) it have, potentially making the map larger, haven't checked whether it clash with RMG resources properties though, if there isn't then yes the map can be made bigger. And even a weirder map templates can be made

In short, the mod can make the entire game truly a different game, or rather truly different experience for the players

My guess is, with planetfall, the dev take the opportunity to both try new things (sci fi settings) and to do some experiments. I guess there will be quite long time before the next AoW game because if i'm right the dev will see how far the community mods go, then collect all the data needed to make the next game better, this will take some time, because currently there aren't many mods with large or staggering scope (like the oftenly wished WH40k mod)
 
Last edited:
Another surprising thing i found is there is a possibility that the size of a sector can be changed, like to total area (in hexes) it have, potentially making the map larger, haven't checked whether it clash with RMG resources properties though, if there isn't then yes the map can be made bigger. And even a weirder map templates can be made


I, and at least one other person, have done this. I even made sectors as small as 7 hexes, just for fun.

I was experimenting with reducing unit speeds across the board, and reducing the sector size, so that it would take longer to traverse the map, hopefully leading to more discrete metropolitan areas, as opposed to having cities kind of bleed into each other as they do now.


It is a visual mess because everything is designed around larger sectors, but the principal is there.

Also, some other guy managed to make summonable units buildable, e.g. Valkyries, and it took me a while but I figured out how to do that as well.

Go on discord and chat in the modding channel there, the people there are far more knowledgeable than I am (I am an ignoramus :p)

Now, back on topic:

AI is crucial imho because most players are going to be single players, so whatever game systems are in place need to be handled by the AI, or you design the game in such a way that the AI plays by a different set of rules.

I would prioritise this over content as well.

If I were to make an AoW, I would:

  1. have relatively few mechanics, but that tie into each other,

  • e.g. resources that are used by every player, such as gold from mines and cities, food from farms, mana from nodes.
  • unit recruitments from cities (uses gold)
  • unit recruitment from independent sites, including independent cities (uses gold and influence)
  • unit recruitment via summing (uses mana)

  1. some sort of city and unit amount limiter, a soft limiter. I would propose a system similar to the "orders" mechanic from Old World
If said limiter works well, we can dispense with relatively arbitrary limitations, such as 6 units per stack (I think I should explain a bit what I am thinking here, namely an army to have an administrative cost ontop of the gold upkeep. The larger the army in one place, then in reality the harder it is to organise, so abstract that away by having an exponentially increasing administrative cost. In other words, the larger your army, the fewer "orders" you have available. Generate orders through buildings, research and special sites.

By tying nearly everything into the use of orders, you can govern/control/direct the gameplay, whilst still giving players options.

If a player wants to create a "stack of doom" with 20 units in it, well that might well cripple his administrative capacity, so cities don't get used etc. For example, assume a stack of 6 (the existing stack limit) is considered the default army size, and carries the usual upkeep cost and also, 5 admin or orders point per unit, so 30 orders points for an army of 6. Now, simply have that every extra unit (i.e. number 7 onwards) incurs an additional 10% orders point upkeep (round up), and what you will have is that an army of 7 does not cost 35 orders points, but instead is 36. And then add unit 8, which would cost the same as unit 7 plus another 10%)

It gets expensive quickly, but in return you get larger armies. Now give yourself a pat on the back if you thought Warlords could directly influence this ;)

Now, the upside of orders would be what you get in Old World, which is you can concentrate those orders on a single unit (in our case army) and allow that army to travel further than it normally would. In Old World, you can move your army it's movement points for free, and then pay orders after that to keep moving it, until you run out of orders iirc...


Go play Old World, it is an incredibly effective tool.


  • I would then create 4 races, no more, to test the orders system, and basic resource system.
These would be what I term "low complexity" races, and they would be Humans, Dwarves, Orcs and one more, the idea being that these races are classic, straight forward, uncomplicated and don't require much set up. Other races, such as Elves, would, in this design, be higher complexity, meaning they'd alter the basic formula in some way, for example by having very limited growth options. This is to iterate the core mechanics first, which are orders, unpacking the city, multiple city queues and combat (which I will talk about later).


  • Then, I would work on the city management system, and "unpack" the cities,

so whatever you build, e.g. a warhall, gets put on the city map (meaning it can be targeted, say hello to a functional raiding system! Higher level buildings would require lower level building, e.g. warhalls would take the existing barracks and make it larger. Using the barracks chain of buildings as an example, with the Human race (and assuming for the moment that they follow the same design as AoW3) , things like the range, and Pike building would unlock that type of unit, barracks would provide hammers, warhalls would unlock Cavalry and provide Hammers and the Knights Building would unlock Knights (and a whole bunch of Knightly technology, doctrines etc.)
Now they key thing to take into account here is that buildings would have a footprint, meaning they would take up space on the map. AND, creating a building would require "orders" as well (they are considerable projects after all.) Now map space is limited, so by way of example, if a barracks takes up 4 hexes, and a warhall takes up 4 more (but must be attached to a barracks) then you need to plan ahead. and that real estate is going to be a precious commodity, because that is where naturally occurring items such as mines and mana nodes are, as well as landmark buildings. Tier 3 buildings will require multiple pre-requisite buildings. Our Knights Guild here would require a warhall, and a Masters' Guild (best armour after all.)This opens up considerable scope for racial variation, as you can place unique racial buildings at different points in the building tree, or condense them (e.g. Elves could very easily not have a range separate to a barracks, instead having them as the same building.) That racial scope is further increased by my next item.


  • I would then introduce multiple city queues, specifically:
construction => buildings and machines
recruitment => military units
support => support units

The way this works is that you have your population, and you allocate them to these queues. And you can produce multiple items simultaneously. So, you can recruit a spearman and build a warhall at the same time.


Now your hammers (production points) are affected by the buildings you make, so if you want to construct machines and buildings faster, you build Builders' Halls. And excess hammers here produce gold, or can be sent to other queues (but very inefficiently.)

Note that you could have 2, 3 or even 4 Builders' Halls, and they would make the production of buildings and machines faster, but not the recruitment of soldiers (except very inefficiently through spillover.)

To recruit soldiers faster, you need more barracks (and Warhalls etc)

So, you can have a Dreadnought (for example) player with very high building and machine capacity, but limited capability to recruit soldiers. And a Warlord be the reverse. And a Sorcerer would have a lot of stuff centred around recruiting more and better support units, but limited capacity to affect production and recruitment.

Already, having 3 queues here gives you immediate variation in designing the races and classes, and we haven't even made a single unit or spell yet. E.G. Humans could produce a flat 50 hammers per appropriate building, but Dwarves would produce 100 per construction, 50 per recruitment and 25 per support, and Orcs 30 per construction, 100 per recruitment, 20 per support.

Obviously these numbers are to illustrate, but the point is that this is a fairly simple system, and even at such a basic level, you now have quite differing gameplay. In the examples above, that means Humans will have a balanced army, Dwarves will favour machines and have fewer soldiers and support units, Orcs will have many troops but few machines and few supports.


Orders affect these as well, and orders (along with gold/mana) can be used to rush things.


 
I would use the above as building blocks, and make sure the AI understand them, or can use them somewhat effectively, e.g. code a "Sorcerer" AI script to focus on support units with a sub script "battlemage" to focus on destructive spells.

Using these, I would then design the races to alter, or outright subvert, some of these mechanics. Then I would alter how research etc work. I am inclined to think, for the sake of simplicity, to use orders for research. After all, we talk about research projects. And this would present some delicious dilemmas for the player, because if you are going all in on many/strong/large armies and prosecuting a war, your research might well suffer.

Goblins, for example, could get a huge growth and recruitment bonus, really rubbish production, bad order generation, poor economy. Thus you get swarms of lower level Goblins, as the more or less default playstyle.


Races would then be higher or lower in complexity, Humans being the simplest, the baseline, and something like Merfolk being the most complicated (meaning here adhering to different rules and having many more limitations)
After the races are more or less set, i.e. with a very strong identity (e.g. many but weak Goblins, Orcs with strong troops, Dwarves using machines and technology, Humans being balanced etc) then I would look at classes, with a two fold approach here, namely, how they would interact with the base mechanics (for example, Dreadnoughts would affect the production queue massively, and open up many options for new machines and fast building construction, and also equipment for troops*) and also how they would mesh with the races.

This is basically answering the question, how would race x approach the concept of this class?. How would Tigrans (who, being cats, are highly territorial, don't cooperate, are fast and deadly, disregard armour, and prefer to bask in the sun and have a high growth rate) approach the concept of using machines (answer: in a way that requires they do as little work as possible! and also allows them to fight in snow etc.)

From that, we can modify Dreadnoughts. Maybe instead of a Flamethrower they get a Sun chariot, that focuses the rays of the sun into a beam.

Also, some classes would be low complexity, and some high. Dreadnoughts, Warlords and Sorcerers would be low complexity, as they map the most directly onto the production, recruitment and support queues respectively.

The higher complexity classes would either affect a combination of these, or would have very distinct bonuses and maluses, or would depend upon other, unique to them, resources. E.G. A Necromancer would revolve around the number of bodies available, and largely disregard orders (as the dead are easier to organise!) and a Theocrat would both generate and use "faith" as a resource (and turn mana nodes into faith generating nodes) and a warlock would focus around blood sacrifice (and so would have to juggle happiness in cities etc)



equipment for troops* = I'll talk about it later, but think empire upgrades in general and unit mods in particular.
 
  • 1Like
Reactions:
Next up:

  1. mods/enchantments
  2. combat
  3. a revised magic system
  4. leader generation and customisation
  5. a handicap system
  6. building upon the concept, introduced in Planetfall, of the "independents turn"
  7. What I would do with dwellings and independent cities (fairly radical)
  8. some examples of complex races (Draconians :D) and classes (including 2, a subset of Druid and Theocrat respectively, that completely upend the game)

I'll discuss these all later, but if you'll forgive me, I am off to play Spirit Island.
 
All of these posts about the AI, and I feel that I am in the minority in that I believe that the AI should be thought of being part of the PvE game, instead of being a fake player in a PvP environment.

To clarify what I mean: I want the AI to be focused in giving the player an interesting challenge, and not in trying to winning the game per se.
Obviously, an AI is not meant to be mere NPC going through very static motions, but an opponent who can react to the various strategies a player can employ:
  • If being raided, the AI should try to fight off the raiders. If it cannot, it should get weaker.
  • Taking cities from the AI should make it weaker. But taking the last city should be the hardest of them all, not the easiest.
  • When a big army is being set up at its border, it should try to concentrate its forces too.
  • Stealthy armies should be mostly ignored, but against a player with stealthy armies, it should have more defender and patrols to find them.
  • And some more that I am missing.
Beside that, I would be perfectly happy with AIs which cheats if they use those cheats to provide interesting challenges to the players instead of trying to run them down.

My ideal AI setup would be one where each AI had a role, which came with both cheats, and specific behaviors which made it uses less than its full cheating force against the player, such as to get fun things to do for the player.

As examples (in the context of Planetfall):
Crusader
Cheat: Extra unit production and energy
Behavior: Always try to be at war with a single target. Select a single city from the target, build an army and send it against the target. Attacks even if the target is sufficiently defended, and scales the size of the army before sending it based on how long the game is. Keep all others armies in its territory or close by to defeat neutral forces or raiders.
How to fight: Either try to spot the target city and defend it with a lot of force, or try to kill the troops before the army is set up by raiding. Do diplomacy to get them in others wars to reduce attacks frequency.
Hegemon
Cheat: Starts with mid-game technology from the start, but has a big research speed penalty
Behavior: Try to bully everyone else by extorting resources from them. Happy to let them live if they give resources or border sectors to the AI. When angered, perfectly happy to do a truce after having conquered a single city or two.
How to fight: Bide your time. Give up to the extortion at first, before attacking the AI with full force after having teched-up.
 
All of these posts about the AI, and I feel that I am in the minority in that I believe that the AI should be thought of being part of the PvE game, instead of being a fake player in a PvP environment.

To clarify what I mean: I want the AI to be focused in giving the player an interesting challenge, and not in trying to winning the game per se.
Obviously, an AI is not meant to be mere NPC going through very static motions, but an opponent who can react to the various strategies a player can employ:
  • If being raided, the AI should try to fight off the raiders. If it cannot, it should get weaker.
  • Taking cities from the AI should make it weaker. But taking the last city should be the hardest of them all, not the easiest.
  • When a big army is being set up at its border, it should try to concentrate its forces too.
  • Stealthy armies should be mostly ignored, but against a player with stealthy armies, it should have more defender and patrols to find them.
  • And some more that I am missing.
Beside that, I would be perfectly happy with AIs which cheats if they use those cheats to provide interesting challenges to the players instead of trying to run them down.

My ideal AI setup would be one where each AI had a role, which came with both cheats, and specific behaviors which made it uses less than its full cheating force against the player, such as to get fun things to do for the player.

As examples (in the context of Planetfall):
Crusader
Cheat: Extra unit production and energy
Behavior: Always try to be at war with a single target. Select a single city from the target, build an army and send it against the target. Attacks even if the target is sufficiently defended, and scales the size of the army before sending it based on how long the game is. Keep all others armies in its territory or close by to defeat neutral forces or raiders.
How to fight: Either try to spot the target city and defend it with a lot of force, or try to kill the troops before the army is set up by raiding. Do diplomacy to get them in others wars to reduce attacks frequency.
Hegemon
Cheat: Starts with mid-game technology from the start, but has a big research speed penalty
Behavior: Try to bully everyone else by extorting resources from them. Happy to let them live if they give resources or border sectors to the AI. When angered, perfectly happy to do a truce after having conquered a single city or two.
How to fight: Bide your time. Give up to the extortion at first, before attacking the AI with full force after having teched-up.
I am not sure we disagree here, at least after you realize that there is no challenge if AI cannot win and even does not try to. I think everyone is fine with AI cheating at hardest levels, that's the point of levels.
 
Wait, why do you

Well, now I am not sure why you play vs extreme AI (if you do) and then complain that it will be impossible to beat extreme AI if it gets stronger. 2-3 colonies is what I have before turn 10 and 5 colonies is my goal for turn 20, you play in a very suboptimal way. Why do you have a single science colony? It is not civ where you stack wonders in a single city. It is optimal to play in a way where you can lose any colony (including capital, yes) and still win. That means thousands of science/energy, hundreds of cosmite, dozens of influence, multiple armies. And not on 100+ turn, but around turn 50-60.
My current game is hilarious, I haven't built any non-secret tech units yet which are researchable and just look at the armies and income at turn 60. 4 times extreme AI still does not have any mods on tier 3 units and I have seen just 1 tier 4 unit. AI definitely needs an improvement.
615F90F17768797E04DCAB1650B1DDDD9E71B977


Seems like you had a very lucky start that's all. I started a new game on RMG and I got put deep into middle of no where rather far away from each AI. So I had to do a bunch of settling and grow the cities for economy. But in the same vein, the AIs left me alone so far except for one dvar who went on a vassalizing spree, he vassalized two AIs before it was even turn 20 lol. I had just enough military to discourage him from escalating the war and limited it to just skirmishing and stare off contests as I colonize the lands around me. I'll be nice and post what I have on turn 62 even when it puts me in a bad light but to be fair, please understand, I lost 12 units in the war. xD Highly armored units is tough enough for troopers to grind down without casualties. I also lost like 6 npc units. Maintained the status quo.

20210114221341_1.jpg


Do you reroll maps until you get very near AI neighbor and then murder that AI. Or do you reroll starts until you get a bronze science landmark? Or at minimum a city ruin so that you can have level 3 science sectors? From what I can see, most of your units is composed of NPC faction units + some T3 units from a landmark + animal recruitments. You're kinda playing like what I would do as a sorcerer in AOW3. lol Playing as a sorcerer can certainly screw the AI very hard because you're basically materializing powerful unit son the spot while the AI has to produce them first. See how you have very few produced units.

From what I'm guessing, you're probably putting all the influence into NPC unit recruitment and likely isn't purchasing any of the settlements or cosmite sites. You're gaming the game extremely hard. Devs should just create a new difficulty tier of the AI for guys like you and leave me in peace against extreme AI. I'm having plenty of fun vs Extreme AI. Its one of reasons why I do not want Extreme AI nerfed or made harder because the difficulties below is way too easy for me. So, let's say, a new AI difficulty above Extreme AI should be named as... Gamer AI? And they'll do same tricks as you do. :p

You need to understand that sometimes once in a blue moon you will get a start that royally fraks over the player no matter what and makes for a long game.
 
Seems like you had a very lucky start that's all. I started a new game on RMG and I got put deep into middle of no where rather far away from each AI. So I had to do a bunch of settling and grow the cities for economy. But in the same vein, the AIs left me alone so far except for one dvar who went on a vassalizing spree, he vassalized two AIs before it was even turn 20 lol. I had just enough military to discourage him from escalating the war and limited it to just skirmishing and stare off contests as I colonize the lands around me. I'll be nice and post what I have on turn 62 even when it puts me in a bad light but to be fair, please understand, I lost 12 units in the war. xD Highly armored units is tough enough for troopers to grind down without casualties. I also lost like 6 npc units. Maintained the status quo.

View attachment 670196


Do you reroll maps until you get very near AI neighbor and then murder that AI. Or do you reroll starts until you get a bronze science landmark? Or at minimum a city ruin so that you can have level 3 science sectors? From what I can see, most of your units is composed of NPC faction units + some T3 units from a landmark + animal recruitments. You're kinda playing like what I would do as a sorcerer in AOW3. lol Playing as a sorcerer can certainly screw the AI very hard because you're basically materializing powerful unit son the spot while the AI has to produce them first. See how you have very few produced units.

From what I'm guessing, you're probably putting all the influence into NPC unit recruitment and likely isn't purchasing any of the settlements or cosmite sites. You're gaming the game extremely hard. Devs should just create a new difficulty tier of the AI for guys like you and leave me in peace against extreme AI. I'm having plenty of fun vs Extreme AI. Its one of reasons why I do not want Extreme AI nerfed or made harder because the difficulties below is way too easy for me. So, let's say, a new AI difficulty above Extreme AI should be named as... Gamer AI? And they'll do same tricks as you do. :p

You need to understand that sometimes once in a blue moon you will get a start that royally fraks over the player no matter what and makes for a long game.
This is a typical game for me, I like to play largest maps with 12 players where everyone has adjacent neighbors. There are no camps/teleports and my bronzemark was food which I annexed at size 12 i.e. quite late (science sector first, residential sector next). Amazons don't need food early due to racial bonus. Yes, having more difficulty levels than current 5 would be a perfect solution indeed. With the mod I use there are 9 levels with 4 new levels above existing extreme. This game is rather easy because of animals and ascended teachers, I lost my previous game at this level. AIs have colonies everywhere and huge armies everywhere too.
 
Last edited: