Deity Empires

I find that I spend way too much time optimizing production and not enough getting care of the strategic side of things (ie, it sometimes feels like I am a manager rather than a wizard/warlord).
What settings could I change to focus less on building stuff and more on killing?

On a side note, it is one of the best maritime AI I have seen in a non space 4X. It keeps wrecking my ships, and it has been rather hard to sink its transports. I eventually had to give up on sea power, and use a portal to launch an invasion.

That’s interesting; we are having polar opposite experiences. In my game the AI builds nothing higher than galleys and sends out unprotected transports with high-level armies aboard. I keep them bottled up on their islands while I take them one by one.

What difficulty are you playing? And what map-type? I have one large continent with several large islands also. I guess I was lucky they started on the islands and declared war on me before actually settling on my continent.

That. And the way double right click to open the unit “card” seems to work erratically. And the way unit icons on the tactical map are oversized, obscuring adjacent depleted units and making it very easy to mis-click. Paladins in dungeons especially.

I was playing at the second highest difficulty.
The AI does build only galleys indeed, but massed galleys win easily against more expensive ships actually. Caravels can kite galleys, but are awfully inaccurate on the move, so I think a mix of galleys + Gallions could work, but the AI kept hunting down my ships with a stack of 3-4 galleys per AI before I could regroup them into a larger fleet.

Try opening the settings and setting that to a single right click rather than a double-click. First thing I did when they changed the default behavior from single-click a while back.

Reports are that starting on islands hinders the AIs so that may be a factor.

The difficulty level must be a big factor then because I have never lost a naval battle against galleys, using any combination of caravels & galleons, and whether I auto-resolve or not. And I leave my transports unguarded because unless the AI has 3+ galleys in its fleet, the auto-resolve combat usually ends in a no-result with no losses. But that is on normal difficulty.

It is true that when starting on an Island, the AI will take a very long time before getting a second city.
I think using the portal spell would really make its expansion much faster too.

An oddity to the game – it has the most conservative ideas I have seen of line of sight/spells, etc. around corners. What a contrast to Gloomhaven.

Corners are screwy. They’re not radially symmetrical w/r/t LOS, and it drives me bonkers.

Just stopping in with an observation instead of a question (since so far all I do is ask questions!).

I’m pleasantly surprised by how good this game is.

The AI is clever enough it very much reminds me of a Aggressors: Ancient Rome, odd how development teams of like 1 or 2 people can pull it off whereas hundreds at other studios can’t.

What does the mithril forge do?
It seems to give a mithril forged bonus (+ armor chance) but it is always on, or is it random?
And what about its production bonus? Is it a one time +10 prod, or +10/worker?
Is there any improvement to use with it?

My understanding is that it gives any units produced in an attached city +1 armor chance. So it’s always on. It has to be inside your city radius to work.

Not all that odd. If every great new system or feature is being put in (or vetoed) by the same guy writing AI… :)

Clever! Good point.

Yeah, just strikes me as odd larger development teams couldn’t find there was thru to working out a development process that would land them at the same end point.

No matter how big the team, most game developers just don’t have anyone that’s very good at coding AI. And it’s often the thing that gets the short end of the stick as far as resources delegated to the project. It’s just very fortunate that the DE devs seem to be very experienced and good in the AI arena. It makes me wonder about their background. There is zero info out there about them or their company.

Larger teams have business people that control the budget and the conventional wisdom seems to be that AI development isn’t worth the money. If it’s the developer’s baby, though, it’s different.

I suspect that one can be “good” at coding AI and still release a game in which the AI “sucks.”

Strategy game AI is a terrifying example of the kind of heavily internally interconnected system that is also heavily dependent on upstream data that has to be an absolute nightmare to deal with.

Even if your company is good and disciplined and your team respects a feature lock well in advance of a beta, for example, what happens when playtesting reveals that your game is more prone to city spam than you thought, and so the designer cranks up some coefficient that puts a drag on expansion? Oh shit, now the AI is over-expanding at the cost of military development or tall infrastructure or whatever. So you lower the weight on expansion activities in the AI’s decision table, okay. Wait, what the fuck, why is it spamming scouts now? Etc, etc.

I have a lot of empathy for anyone trying to wrestle with strategy game AI.

Is there any impetus to actually use any of these endgame global spells? I sort of feel like you’re way better off spending your meditation on summon spamming and buffing.

Take the Ice one for example. Endless Winter or whatever. So it turns squares to tundras. It’s useful to one race. That’s silly. There is no impetus whatsoever to take that spell if you aren’t playing that specific race. Yea, you might have access to one or two more summons with tundra walking, but you’ll also have access to adult Dragons, so haha, who cares.

Dominions has probably at least 20 examples of horrifying global spells you need to disrupt. Even MoM had some pretty nasty ones. But these ones seem kind of piddly.

Yes. There are actually a few pretty powerful ones.

One that I’m using right now is the death spell, ‘Living Dead World’. Combined with the ‘Living Dead’ battle spell it can be pretty ugly for opponents. What happens is that any time you kill an enemy unit, it rises as a random undead unit and fights for you the rest of the battle. After the battle, you get to permanently keep as many of those surviving undead units as will fit in your army. So if you keep 4 spaces empty in your unit stack, you essentially get 4 free undead units every battle. Talk about force multipliers. If any AI has that up, it’s a priority dispel for me.

Another one is the arcane spell, ‘Time Warp Bubble’. You cast it on a tile and then it has 2 effects. First, there’s 0 movement cost in the hex. Second, and more importantly, the deity owning the spell can cast 2 battle spells per turn rather than one. That second effect is very powerful and I’ve seen some people on the Steam forums bitterly complaining about the AI being able to cast twice per turn, not realizing it has this world spell in effect.

You may wonder what good it is if it only affects one hex. Well, if you cast it on your best city’s main hex, it makes defending the city from attacks a whole lot easier. You can also cast it multiple times (it cost 5 mana per turn to maintain for each iteration) so if you have tons of mana you could actually make a zero movement cost road with this spell.

There’s also the ‘Slow Time Bubble’ spell which is similar but reduces other deities ability to cast in combat to every other round. I wonder if you could stack the two? It would be pretty devastating.