Caster of Magic II Brainstorming Megathread

[Seravy]

Road building costs. Are we happy with these or do we want it different?

LTGrassland,LTDesert : Result:=30; // 2 turns
LTHill : Result:=15; // 4 turns
LTMountain, LTVolcano,LTChaosNode : result:=12; // 5 turns
LTSwamp : result:=20; // 3 turns
LTTundra : result:=20; // 3 turns
LTRiver,LTRiverStart,LTSorceryNode : result:=20; // 3 turns
LTForest, LTNatureNode : result:=30; // 2 turns

Note the engineers are fairly cheap to produce so higher costs don't make all that much difference, players can build more engineers. The AI is not good at handling it though (as it can't protect their engineers nor do they build multiples) so the costs are more for flavor than a real obstacle to slow down the road production. In general, the difficulty in road production is building them without losing the engineers in battle, and considering the downsides to having roads (faster melee movement in combat can make cities harder to defend and enemy troops can also use them to reach your cities outside battle).

[Aeron]

Road building costs. Are we happy with these or do we want it different?

LTGrassland,LTDesert : Result:=30; // 2 turns
LTHill : Result:=15; // 4 turns
LTMountain, LTVolcano,LTChaosNode : result:=12; // 5 turns
LTSwamp : result:=20; // 3 turns
LTTundra : result:=20; // 3 turns
LTRiver,LTRiverStart,LTSorceryNode : result:=20; // 3 turns
LTForest, LTNatureNode : result:=30; // 2 turns

Note the engineers are fairly cheap to produce so higher costs don't make all that much difference, players can build more engineers. The AI is not good at handling it though (as it can't protect their engineers nor do they build multiples) so the costs are more for flavor than a real obstacle to slow down the road production. In general, the difficulty in road production is building them without losing the engineers in battle, and considering the downsides to having roads (faster melee movement in combat can make cities harder to defend and enemy troops can also use them to reach your cities outside battle).

I think it should take more turns to build roads in forests compared to building them in grasslands/deserts and it should take more turns to build roads in swamps compared to tundras. I believe this would add a bit more scale and variety to road building.
I also wonder if it's possible to add more types of terrain. For example, we have swamps, which are flooded counterparts to forests. Could marshes, which are flooded counterparts to grasslands, maybe be added? This might be very difficult to program, so I hope I am not bothering or pressuring you with this suggestion.

[Seravy]

For example, we have swamps, which are flooded counterparts to forests. Could marshes, which are flooded counterparts to grasslands, maybe be added? This might be very difficult to program, so I hope I am not bothering or pressuring you with this suggestion.

Possible but I need a very good gameplay reason to do that as it's a lot of additional work and I don't think such a reason exists. The current types of land cover all kinds of resource bonuses and amounts well enough.

[massone]

Road building costs. Are we happy with these or do we want it different?

LTGrassland,LTDesert : Result:=30; // 2 turns
LTHill : Result:=15; // 4 turns
LTMountain, LTVolcano,LTChaosNode : result:=12; // 5 turns
LTSwamp : result:=20; // 3 turns
LTTundra : result:=20; // 3 turns
LTRiver,LTRiverStart,LTSorceryNode : result:=20; // 3 turns
LTForest, LTNatureNode : result:=30; // 2 turns

Note the engineers are fairly cheap to produce so higher costs don't make all that much difference, players can build more engineers. The AI is not good at handling it though (as it can't protect their engineers nor do they build multiples) so the costs are more for flavor than a real obstacle to slow down the road production. In general, the difficulty in road production is building them without losing the engineers in battle, and considering the downsides to having roads (faster melee movement in combat can make cities harder to defend and enemy troops can also use them to reach your cities outside battle).

If the AI can't handle building multiple, how about just disabling multiple engineers stacking? Make the player build the same way as the AI, one engineer at a time. If that's too slow, then just speed it up for everyone.

[Seravy]

They can handle it in the sense of using the feature, but they don't build multiple engineers so it only happens if they hire a stack as mercenaries or they accidentally end up on the same tile.
However if we assume 5 engineers building 5 separate roads is as good as one stack of 5 building one road which actually is mostly true, we can actually solve this by raising the AI engineer production limit to more than one per continent, we just need to decide what's a good number. In fact for this same reason, engineers not stacking is not relevant and has no effect on either the AI or the human player.

So if we follow that direction we can have more expensive roads but only if engineers themselves become more expensive as well (otherwise it's trivial to build more of them). Whether that's good or not though, is something to decide. I can't come up with a good reason for raising the costs so the safest bet is "if not broken, don't fix", and keeping the current range of costs.

[prokolyo]

Stacking engineers gives the human strategic advantage, no? Because sometimes it's imperative to build a road fast.

Sent from my Pixel 3 using Tapatalk

[massone]

Stacking engineers is definitely a strategic advantage, the only question is how big. The AI doesn't know how to do effective build orders, and can't transport engineers either. Roads are more valuable in high pop towns, and if it takes 15 turns to build 3 squares on two roads with 1 engineer working each road, that's 7 extra turns of money for the player who stacks engineers and finishes the first road before starting the second, but that's only about 210 gold on 2 25-pop cities.

The AI does build a fair amount of roads in my games when they have access to engineers in their starting race, or in their early conquests close to their first towns. The overall advantage is probably quite small, especially on higher difficulties when they get gold bonuses even without the roads.

The military advantage is not really solvable, I think. The AI isn't capable of coordinating their forces in a manner that allows for rapid reinforcements and high speed doom stacks, even if they had roads.

[namad]

IMO, making the ai build more engineers would be a nerf to the ai. The ai already builds too many roads imo. They build more roads than are strictly necessary and they connect their cities to the road network even if they have a city in enemy territory, a city that means connecting is a strategic disadvantage. The best way to keep the ai from building unsafe roads, is to not increase their quantity of engineers any further than it already is. Also keeping in mind that the ai gets, typically, some cheating based gold, their need for road gold bonus is extremely low.

As it stands I'd say the ai needs to have no changes to engineers, fewer engineers, OR a totally new ai subroutine that determines which roads are a negative to build and then doesn't build those.

[Seravy]

27. Spell Blast to rare
With larger maps and more players, Spell of Mastery wins will be more relevant and the number of Sorcery wizards the player has to eliminate before they can cast it without getting countered should likely be lowered by making the spell more rare.
With this change it also becomes possible to remove the "do not use at random if the human player has below 200 casting skill" which is a very artificial solution. If the spell is rare and low on the AI's research priorities, this is simply not an issue, unless they find it in treasure but even that has some turn based restriction. If necessary we can also add a condition for the AI to not use it before a certain turn count. The problem here is Sorcery having no Rare spells that could be moved down to uncommon.

[massone]

27. Spell Blast to rare
With larger maps and more players, Spell of Mastery wins will be more relevant and the number of Sorcery wizards the player has to eliminate before they can cast it without getting countered should likely be lowered by making the spell more rare.
With this change it also becomes possible to remove the "do not use at random if the human player has below 200 casting skill" which is a very artificial solution. If the spell is rare and low on the AI's research priorities, this is simply not an issue, unless they find it in treasure but even that has some turn based restriction. If necessary we can also add a condition for the AI to not use it before a certain turn count. The problem here is Sorcery having no Rare spells that could be moved down to uncommon.

I disagree with this. The number of Sorcery wizards using Spell Blast to counter other spells should be increased, not decreased. With larger maps and more players, you need to move larger distances and simultaneously stop multiple Wizards from casting game-winning spells in a competitive game. If the game only has 1 Wizard casting SoM or other game-winning spells like Time Stop, Power Link, Armageddon, Spell Binding, etc, then Spell Blast isn't even relevant--it's not a competitive game. I emphasize that the game should not be about player vs. all AI like PvE, it should be every Wizard vs. every other Wizard.

Spell Blast should not only stay Uncommon to ensure it's always in play, the AI needs to make it a top priority in preparation for late game and consistently use it to stop dangerous Rare global Enchantments and war status enemy summons worth 300+ casting skill (trading only 50 casting skill for it), and ensure they have Detect Magic active as soon as possible.

Particularly on high difficulty (or multiplayer--the assumption is that all human players are skilled), larger maps, it should be next to impossible for rush conquering strategies to succeed, and fewer Wizards will be eliminated. Why? Because with the current game balance of RP and SP calcs, assuming that every player is of reasonable or similar competence, wasting resources in fighting an offensive war all the way to banishment is not worth it in 95% of cases as you lose relative progress in RP, SP, and unit build up against all non-participating wizards, and get very little in return because defenders have an overwhelming advantage in combat.

Spell Blast is a vital spell to balance out the extreme power of Very Rares.

Even on Lunatic Difficulty Huge maps now, in the game I just played, Very Rare spells appeared in the spellbook as early as 1409, and AI started casting Very Rares in 1411. I fought no wars whatsoever and focused entirely on research and beat the Myrran AI to VR by a single year, while the other Arcanus Wizards who had fought between each other only reached VR 1-2 years later. I was exceptionally unlucky this game and had only 10 towns, while finding 3 wrong realm spellbooks that slowed down my 2 primary realm research by almost a year due to the large amount of common/uncommon spells that took 1 turn each to be researched and wasted the excess RP.

In 1410 the Myrran Wizard and strongest Arcanus Wizard had 5 times and 4 times more Power than me respectively, and more than twice my town count each, but I reached VR even faster than the Myrran Wizard, and had 3 researched by the time she had 1.

The result?

The Arcanus Wizard who wasted resources aggressively expanding and appeared powerful was crushed by the Myrran wizard who nuked half her cities in a year before they had any VRs, despite having more towns and almost the same military/power rating. She had a 4-Sorcery book play that unfortunately didn't have Spell Blast to protect herself. 

The Myrran Wizard who wasted resources--I don't know doing what--was crushed by me, who not only had more VRs, but continued researching new ones during the war. All of her massive advantages, territory, adamantium, better racial units, Power output, all of it was irrelevant.

After I joined the war seriously and conquered most of Myrror, the Arcanus Wizard had a comeback and occupied the remainder of Myrror. By then, she once again appeared to be the strongest other than me, at least on paper. But it was false--the other Arcanus Life-Sorcery Wizard, quietly researched Enlightenment, Sky Drakes, Crusade, Power Link, and Spell Binding, then bound Time Stop from me, and instantly became the biggest threat despite owning only 15 towns. If they had been an actual competent human-level intelligence player, then I might have have lost the game, because I couldn't Spell Blast the Time Stop casting when there were two Power Links in play (not mine), and it would only take 5 turns in Time Stop to strike my Fortress and raze half my cities. And they definitely had enough mana reserve for 10+ turns. It was pure luck that one of two doomstacks I had that was strong enough to take on their highly buffed Beastmaster Sky Drake filled Fortress was within 2-turn range by Flying Ship.

Again, this showed that the time and resources I wasted conquering the Myrran Wizard turned out to be a trap--I gained all this territory and tripled my Power, had the highest casting skill and military rating, and in the end it all came down to one spell -- Spell Blast, which I should've cast earlier to stop the Power Links. It wasn't because Power Link provided an overwhelming magic power advantage to these AIs--that actually turned out to be irrelevant, it was the fact that they blocked Spell Blast that made all the difference. And I couldn't have used Time Stop myself earlier, because that same Wizard did have Spell Blast and Detect Magic, and I didn't have Power Link to protect myself (they wouldn't have used it against me as their ally, but that's the point I'm making. I don't take advantage of AI exploits as a house rule, but the AI could have and should have used Spell Blast to counter me, regardless if I was an ally.).

If Spell Blast is even less accessible, then the only way to stop these game-winning spells is to banish the Wizard. On even bigger maps, there is no chance of getting to their Fortress before they can finish casting, as it only takes 2-3 turns.