Races, Units, Buildings

[Seravy]

One thing I realized meanwhile about Amplifying Towers.
They have a huge disadvantage, similar to military buildings.
Other resource producing buildings produce the resource per turn, until the building is lost. At that moment, you lose the building and any future income, but get to keep everything the building already produced.
In case of an Amplifying Tower, if you lose it, you retroactively lose all the benefits. The 7 skill is gone, and the "phantom" SP gain/turn by making your skill higher than what you pay for also disappears : If that extra 7 skill is gone, SP you saved by paying for the 50th skill point instead of the more expensive 57th is no longer saved because you now actually have the 50th skill point only. Furthermore, not only do you lose everything, if the loss is from losing the city, the benefit transfers to the opponent immediately, and it'll be most likely higher than yours - the AI capable of taking away your cities usually tends to have a higher skill. If all of that wasn't bad enough, the gold you paid for maintenance won't come back either - although in exchange you were able to use that extra 7 skill while you had the building. Also, because the enemy gains the benefit immediately, losing even a few cities with these can turn a war in the opponent's favor, as the combat casting skill gap shifts by 14 for each one transferred to the enemy. This isn't the case even for military buildings, while the enemy can start producing troops using them, they need time to do so.

I think this makes Amplifying Towers a much higher risk building than others because no resource is gained if the build gets destroyed by anything (combat, earthquakes, etc), which is one more reason for keeping it powerful.

I'll pay a bit more attention to it in my future games and see if any cost or maintenance adjustments are needed (in case I end up buying them too often/early), but I think it'll be ok to keep them in the 800-1000 cost, 7-12 maintenance range somewhere.

[Nelphine]

I'd like to know your goal for how much skill should really be gained from Amp towers. Which in some ways also means, what kind of casting skill do you really think is necessary? I think a mid game casting skill of 50-120 is fine. A late game casting skill of 200 is excellent. With amp towers, you could have 200 casting skill without ever putting a single point of power into casting skill.

I think at +5 per amp, they'd still be worth a huge risk. +3 per amp is where i'd start to have second thoughts about it. +7? It's simply the best building, even if it cost 15g per turn.

[Seravy]

I think 300 is acceptable for late game, 500 is excellent. 200 is too low. This assumes "late game" refers to a game state where you control about 1/3 to 1/2 of the world and are preparing for the big war against the remaining wizard who controls the other half, most likely at turn 250 or later.
With 200, it takes 5 turns to summon a very rare creature, or 45 turns for a stack of 9. Assuming there are about 100 turns between you researching your first such spell, and the strongest enemy reaching spell of mastery, that rate is slow, even if you cast no other overland spells, which is unlikely. You probably want to cast global enchantments, recast them if dispelled, and dispel at least some enemy global enchantments. In same cases, you also want certain city enchantments on all cities, especially Consecration, Gaia's Blessing or Spell Ward to counter global spells that wreck terrain or kill units. (Final Wave, Great Unsummoning, Armageddon, Great Wasting)
White it's not typical, sometimes you also need some artifacts, and those can easily cost 2000-5000 each.

In my experience, even 300 late game skill is barely adequate to stand a chance in those big endgames, but I have to say not every game reaches that state. Often the game is decided in the midgame, and the late game is just about mowing down the last enemy with superior armies and spells taking advantage of having a stronger economy. In those cases, 200 skill is certainly enough but those games don't have a real endgame.

50-120 for midgame sounds about right.

I'm starting to think this is not a case where we need to compare it to other buildings. It's the best building for increasing casting skill...and the only one. Technically power producing buildings can also increase skill if you invest the power into that, but that effect is so slow I'm not sure anymore there is a point in comparing "I got 7 skill now" with "I'm going to get 7 skill 50 turns later". It's just not the same thing.
I have to agree with zitro here, Amp Towers are like Sawmills. They are obviously the best, but the game needs them. The question is the cost, not the effect.

[Nelphine]

Right, but its completely rrssonsble for a strong AI, when all 5 wizards are alive, to have 20 cities. 30 cities isn't unheard of. That's 140-210 skill, right there. And it would never take close to turn 250 for that. I'd wager an AI would have that by turn 180.

That's 200 casting skill, 6 years before late game is reached, without actually looking at casting skill gained from power. And that's largely regardless of difficulty (although lower difficulty will mean AI are on the lower end of the city count).

That's why I think the concept is OK, but the bonus is too high. It artificially pushes you to end game casting skill levels simply because of the size of your empire. Dropping the bonus to +5 would reduce your casting skill to +50 to +150 instead, which wouldn't push you to end game skill while still in mid-late mid game.

[Seravy]

Right, but its completely rrssonsble for a strong AI, when all 5 wizards are alive, to have 20 cities. 30 cities isn't unheard of. That's 140-210 skill, right there. And it would never take close to turn 250 for that. I'd wager an AI would have that by turn 180.

We were talking about the human player though.
The AI is a completely different story.
For starters, the AI builds randomly so it won't have the Amplifying Tower in every city, and it won't be built early. They do have a production and gold bonus, but that is scaled by difficulty. It's only enough to make up for building randomly on the highest levels. I'd say on Extreme the AI is expected to have the same percentage of towns Amp. Towers as the human player, on impossible it might have more - but that's kinda the point of impossible, to give the AI advantage. They might have more cities, but again, "might", ultimately, if the player was actually playing the game and expanding, they'll have an empire of a comparable size. Finally, over half a player's empire usually comes from the AI through conquest. It's possible, but rare to have more than 5-10 cities made through settlers. So even if the AI did build more amplifying towers faster, it is potentially helping the player.
And yes, a few posts ago I said the AI having more skill is the secondary design goal. The AI having an extra 140-210 skill on turn 180 (assuming extreme or impossible) is exactly what I want.

Your numbers seem to assume the game is played on HUGE land though. With Fair land size and 5 players in the game, 10 cities per player is more realistic - unless you mean the Myrran wizard having the entire plane alone in which case yes, they might have 20-30 cities and yes, the Myrran wizard is supposed to be powerful so that's perfectly fine. Of course, in some cases it's not the Myrran wizard, but ultimately, if one player has 20-30 cities, that means the others are that much weaker. You also seem to assume all cities are equally developed which is far from the truth. Yes, the AI might build 30 cities in 180 turns on extreme or higher, but the last 10 of those will still be 5 pop with minimal buildings. Settlers don't teleport on the world map, it takes time to fill up an entire plane with cities. If they got the cities through conquering each other, that's different but then they should have lower casting skill because wars eat a lot of mana crystals which means less power to spend on skill and research.

Finally, one thing you seem to forget, your numbers assume every city can actually build the tower. This isn't true. Barbarian, Gnoll and Dwarf cities cannot. That's 2/9 of the cities on Arcanus and 1/5 on Myrror on average. The actual number greatly varies depending on which was the starting race for each player, but assuming the 2/9 ratio, the 7/skill city is instead 49/9=5.44

[zitro1987]

Right, but its completely rrssonsble for a strong AI, when all 5 wizards are alive, to have 20 cities. 30 cities isn't unheard of. That's 140-210 skill, right there. And it would never take close to turn 250 for that. I'd wager an AI would have that by turn 180.

That's 200 casting skill, 6 years before late game is reached, without actually looking at casting skill gained from power. And that's largely regardless of difficulty (although lower difficulty will mean AI are on the lower end of the city count).

That's why I think the concept is OK, but the bonus is too high. It artificially pushes you to end game casting skill levels simply because of the size of your empire. Dropping the bonus to +5 would reduce your casting skill to +50 to +150 instead, which wouldn't push you to end game skill while still in mid-late mid game.

Early Game - 3-5 towns - combined 40 production / about 30 skill
Late Game - Large Empire of developed cities - combined 1500 production / about 400 skill

As an example (if half of empire is cranking out units):

Early game - 2 military towns - 1 swordsman/cavalry per turn   (mana: about 1 hell hound per turn)
Late game - 15+ military developed cities - 5 stag beetle/etc per turn  (mana: 1 stone giant per turn)

If you think about it, your production capacity as an empire is somewhat superior to the casting capacity of your wizard.

[Nelphine]

*shrug* all right.

Since this is how it's been for a long time, I admit it must be working reasonably well. It will make my playstyle stronger, but, that should be ok.

[Seravy]

Early Game - 3-5 towns - combined 40 production / about 30 skill
Late Game - Large Empire of developed cities - combined 1500 production / about 400 skill

As an example (if half of empire is cranking out units):

Early game - 2 military towns - 1 swordsman/cavalry per turn   (mana: about 1 hell hound per turn)
Late game - 15+ military developed cities - 5 stag beetle/etc per turn  (mana: 1 stone giant per turn)

If you think about it, your production capacity as an empire is somewhat superior to the casting capacity of your wizard.

I agree. Even with the current Amplifying Tower, the overland casting capacity can't keep up with the production capacity. However in exchange, spells are much more cost-effective than normal units and buldings.

Someone suggested Amp Towers to be a % bonus to skill and I just realized, this isn't a bad idea. The problem is the amount. Let's say we make it 5%. Then in the endgame (200-300 skill), each of them adds +10 to 15 skill, far too much, but earlier (50-100 skill) they add a mere 2.5-5 each which is low considering the relatively higher cost (1600 gold is a lot harder to pay on turn 100 than on turn 200) so the problem with this idea is, it doesn't scale well.
...actually, what if it was "This building adds +7% casting skill or +7 casting skill, whichever is lower"? So if you build it before you have 100 skill, the effect is weaker? This seems to be "the" solution, as everything stays unchanged yet it still prevents bumping up your 30-ish skill to double in a turn just by spending 6k gold in the early game, which is the only thing I don't like about the towers. Not sure if the player can keep up with the high difficulty AI without it, though.

[frutagode]

The dark elves production screen shows that many units have "scouting II", but when you pruce them they show  "scouting I", but actually they have no extended scouting range.

[Seravy]

The dark elves production screen shows that many units have "scouting II", but when you pruce them they show  "scouting I", but actually they have no extended scouting range.

The production screen uses a different subroutine than the unit view so it has some inconsistencies. I haven't had the time to fix that, and probably quite a few of them cannot be fixed at all - especially the abilities that don't appear. If they see 2 tiles ahead, then they have the correct scouting effect. Other units without the ability see 1 tile ahead only.