Caster of Magic Release thread : latest version 6.06!

[Nelphine]

Giving people 2 starting settlers won't change the effect of on high difficulty, the best option is still to conquer the AI. It just means that at some point (and with slow races, which this is aimed at, outpost growth is usually low) you end up with 3 cities. You still use summons to take away enemy cities after that if you want to be competitive on high difficulties.

[Nelphine]

I'm sure its been suggested. And response given, and I'm sure it was rejected, but for the life of me I can't find the conversation.

If we (instead of giving free settlers) limit the AI to one less settler than currently, they would settle good cities slower. This would give the human more time to get good city spots. The human could make the choice of settlers or early rush but there wouldn't be as many cities to take (in the short term). Also in the short term, the AI wouldn't have as much to defend, so would actually defend each city better via summons. So while early rush would have less targets for the first AI, they would be better defended, so the human would come out worse (more damage taken and less cities gained) against the first early rush target. Alternatively they could choose to build settlers since the AI would be more limited in the medium short term due to fewer cities making it easier for the human to defend their new cities as they started growing, while by middle things would go back to expected because:

After 30-60 turns (probably based on difficulty) increase settlers back to current levels (since in the long run more cities wins FOR THE AI - humans can most certainly plan around winning with less cities with a wide variety of races).



Please note, again, I don't object to the extra settlers. I simply think they will not solve the problem you have said you are trying to use them to solve.

[Kaiku]

I'd be much happier with smaller incremental changes than drastic ones.

I'm a supporter of this as well. I've been convinced this is a real problem and that the human needs some early advantage in expanding, but dumping two settlers in our city on day one is such a massive... ugly... change it makes my head spin a bit. It even has some minor side effects as well, as usually I'd leave the two starting units at home and summon spirits to do scouting, or I'd scout with the two units. Now, I obviously send out the two settlers... leaving me with one useless unit at home (who needs a buddy). Again, it's not a big deal, but it's sub-optimal, doesn't feel good, and still only turn one.

Why not start incrementally? Give us one settler, and reduce their cost a little, to see how that works out first. The AI doesn't care about the cost anyway, so this should be for our benefit. This also gives us the chance to save up a bit for that first sawmill, and then the next - opposed to developing two cities simultaneously which we're not ready for anyway.

[Seravy]

If we (instead of giving free settlers) limit the AI to one less settler than currently, they would settle good cities slower.

That would mean going back to the vanilla game's AI. That's slower than a turtle. 1 settler at a time...
Currently it is set to 2, so there is no room to reduce it, we can't have it at 1.5 obviously and 1 is extremely sluggish.
Also, only 1 settler at a time means the AI can be prevented entirely from building cities by hunting for that one settler and killing it repeatedly. Obviously this can be done to 2 as well but it's far more difficult.
Look, I remember how much the AI sucked without the 2 settlers at a time change so I don't want it back. We don't want to see AIs with 3 cities in 1410 again.
Seeing how big difference the additional settlers Myrran are allowed to have make, even 2 is holding back quite significantly, but on Arcanus the AI can't risk going all settlers.

leaving me with one useless unit at home (who needs a buddy).

I can change the number of starting swordsmen to 2 if needed. Haven't had the time to bother with that last time.

Why not start incrementally?

I believe I answered that. It's easier to judge the effects of a change if it's large. This is a feature that might or might not be kept for later versions, and I rather find out if it's good or bad now, not 2 years later.
...besides, settlers don't come in halves. 1 and 2 are the first two integer numbers. So 1 for smaller maps, and 2 for larger because we obviously need the amount to be scaled by map size. There is no smaller possible number here to use.

[Kaiku]

So 1 for smaller maps, and 2 for larger because we obviously need the amount to be scaled by map size.

Why? What's wrong with just 1 for all map sizes?

[coldsteel]

Settler costs have been strange lately, and here the city of Cunaxa can apparently produce settlers for a cost of 3(150).

[Seravy]

Thanks, fixed for tomorrow's release!

[m59gar]

Howdy. New here and loving Caster of Magic.

Is it possible for us to have a setting where the AI is as smart as possible, but doesn't have absurd resource advantages? I've been playing on Expert and above and I have to be honest, there's no legitimate way to win at these levels without relying on save scumming both diplomacy (so the AI doesn't attack) and battles themselves. Hell, I just started a Tiny landmass game where my Draconian capital started with a bunch of gold squares and I immediately bought 9 bowmen and raced straight for an enemy capital while casting Heroism on them on the way.

By the time I got there, the AI already had 9 Ghouls, some of them with Weapon Immunity from Wraith Form. My forces were obliterated, obviously. So I reloaded and did the same thing going for a different wizard. By the time I got there, he had 8 Nagas with Focus Magic and again simply blew me away. Ok, third try, I save scummed until I crushed a capital purely on bad rolls by the enemy - a few turns later, while banished, he was still able to (from 1 outpost by the way) summon a doom stack of hell hounds and find and destroy my capital immediately. WTF? In every example, that's got to be hundreds and hundreds' worth of casting skill and mana in just a handful of turns, and these were all just on Expert.

At that point I realized that I would never have fun with difficulties Expert and above. With Expert, and Master and Lunatic even more so, you very quickly run into an obnoxious wall where your cities are literally surrounded by doom stacks from every wizard and the only way to progress is by save scumming to avoid war. What's the point of settings like that?

But at the same time, the smart AI is great. I'd like a setting with intelligence set to 100% but no resource advantage.


PS - Maybe I'm a noob, but I think Ghouls need to lose their ranged attack. They're an obnoxious standout every single time. Every other early game unit is balanced and interesting, but these things just freaking mow down everything in the same tier - and then resurrect them as undead! WTF?

PSS - Corruption is an insanely un-fun spell. I genuinely don't see the need for this to exist the way it does. The AI just kind of spits on you randomly with it and you have no idea who did it. I'm not even at war with anyone and sometimes they're Corrupting or Raising Volcanoes on my resources for no reason. Why is this a thing? I find myself relegated to Nature books just to fix this random damage that would otherwise be crippling.

[Seravy]

Change of plans, this bug is too severe to wait with a release, so here goes :

5.05
-Nagas now cost 92.
-Raised AI catapult production priority.
-Ghouls now cost 88.
-Skeletons now have 7 figures.
-Fixed 5.04 bug : Settler costs are sometimes random.
-Reverted 5.04 change : The trade gold bonus cap is now 5%/population but a maximal 75% instead of 3%/population. (see http://www.realmsbeyond.net/forums/showt...p?tid=9028)

Nothing significant this time except the minor adjustments on the units that were discussed.

[Seravy]

Is it possible for us to have a setting where the AI is as smart as possible,

I guess that's possible but...
1. It'd be part of the settlers menu, so people could turn it on/off during games which is not how difficulty settings should normally work
2. It'd take quite a bit of work to integrate it into every single AI decision that's difficulty level based. In particular there is at least one in a part I rather not touch unless absolutely necessary, and which might not even have room for this additional condition.
3. The difficulty levels are still quite new (the 6 level system was added in 5.0) and there is a good chance the resource advantages might get lowered further, albeit not in the categories I already consider set in stone (maintenance and overland casting).

I've been playing on Expert and above and I have to be honest, there's no legitimate way to win at these levels without relying on save scumming both diplomacy (so the AI doesn't attack) and battles themselves.

That's simply untrue - I've won games in Expert in the current difficulty system, and some people do it even on higher settings. However it takes a lot of experience with the mod and all the changes in it.
I do suspect Lunatic as is serves barely any purpose except masochism but it is very helpful for testing : if a strategy beats that consistently, it's a fairly good signal that it's overpowered. Master however is a legitimate level that is supposed to be beatable - whether it is or not I have no idea, as I said the system is new and I have only had the chance to play up to Expert so far. Games take 20 hours each so it's not something that happens overnight.

Ghouls - being death realm creatures which is the second best at early game summoning, are on the strong side but what makes them so is not the ranged attack, it's the poison. High resistance units should work well against them. Also, their power is reflected in their cost, they are (as of the latest update) the second most expensive common creature. It that was a thing, I'd give them 0.8 Poison strength but obviously it only comes at integers, so it has to be 1 even if it's a bit on the strong side. That said, they are on the "to test and modify if needed" list, and this current cost increase is part of that.

Corruption - Chaos wizards are about blowing up your stuff. If you have problems with that, what do you say when they nuke your city or corrupt the entire world? There are a few ways to prevent it from happening though. If you don't let them get in range to see your cities, they can't corrupt the area around it. If you are in a wizard's pact or other treaty with them, or they have a personality that doesn't allow being hostile without declaring war, then they can't do it to you at all. And of course there is always the solution of "you destroy my land, so I destroy you!". (Yes, if there is more than one with Chaos books it can be a bit tricky to know who is doing it but possible. It's the person who is also attacking your units, the two use the same diplomacy mechanic called hostility.)
Note that threatening a wizard gives you a peace treaty if successful, which does count and prevents casting such spells until it expires. So if you are more powerful you can just tell them "hey stop doing that or you'll regret it" and they'll do for a while. Or course, a Wizard's Pact or Alliance is a more permanent solution if possible and desired.