Quote:Based on what, exactly? As far as I'm aware, the queued movement behavior of strategy games doesn't have a consistent design pattern across different titles.
On common sense.
I mean if I tell the unit "go there and don't bother me until you arrived" and they wait for confirmation every turn anyway that defeats the purpose the automove system.
It's one thing in Heroes of Might and Magic where I'm at most allowed 8 heroes that can move and usually only have like 3 or 4 but in MoM that means I might have a hundred units waiting for an additional confirmation for no reason at all.
Even if I don't move that many stacks manually, if I set like 20 cities to Rally and each produces a unit per turn, then it's not that unlikely to have 40 or 60 or even more units on the move at any given time in separate 1 unit stacks, depending on how far the cities are from the rally destination.
Quote:This, on the other hand, is definitely an unexpected action. If I'm selecting a stack, I have a reason to select it, which may not be to re-affirm whatever I did last time. If the game immediately says "whatever, I'm doing the previously planned action" that's always going to be unexpected.
Except, the premise I was working with is that you CAN'T select the stack if it already has a valid going order because if it did then it already moved. It's not possible to issue two different orders at the same time and the one you do issue executes instantly, and orders from the last turn that carried over already executed before you get to do anything as well. So basically, the entire system wasn't designed for interrupting previous turn orders.
I don't feel like overhauling the entire overland movement UI and risk adding new bugs for a feature that, like you say, "has no consistent design pattern" so it's not like either of the solutions is strictly better than the other - both have their own flaws and it's subjective which flaw is greater. This isn't easy to add and would require splitting the "going" status to two different unit orders, "going now" and "going carried over from the previous turn". Which also means every part of the code that checks for "going" status needs changes with potential to break stuff, from AI to human movement.
Quote:I don't feel like overhauling the entire overland movement UI and risk adding new bugs for a feature that, like you say, "has no consistent design pattern" so it's not like either of the solutions is strictly better than the other -
The lack of consistent design pattern is in when, exactly, the units move. What's more consistent, at least in any game I can think of, to verify moves if conditions change (over a transition to the next turn, in this case). That's about all I can add, not wanting to mess with it is your prerogative.
(October 21st, 2020, 15:25)Seravy Wrote: It's a river tile. Should cities interrupt rivers the same way they do to roads? Or leave the river there but override the cost?
Does it matter from a gameplay perspective either way? They end up with 1 movement cost regardless, right? So I'd say whichever is easier to implement, if it's a purely graphical thing. It probably makes more sense to leave the river there and override cost though if there is a difference. In the future, we might even want to have something like water units able to go through rivers underneath the wall.
I am researching Dispelling Wave and it is not finishing despite it going to 0 RP left. I tried increasing the amount allocated to research before ending turn and that doesn't work either. It only finishes on the next turn.
In this game I'm seeing strange diplomacy between AI. Opposite alignment AIs are signing Pacts and even alliances with each, and only within a few turns of first coming into contact.
Sharee, Aggressive Militarist, 6 Chaos 6 Death signed pact with Kali, Ruthless Perfectionist 5 nature, 8 Sorcery on the very first turn their units met.
Sharee also signed Alliance with Raven, Peaceful Narcissist with 3 nature 8 sorcery 1 life. Not sure about timing, but they were allied by the time I met them both, which I think was around 1404.
Sharee also signed Pact with Ariel, Aggressive Theurgist, 6 Sorcery, 6 Life.
Shouldn't Sharee have very negative starting relations with these 3? And their personalities except for Raven aren't easy to deal with either.
Too many alliances
Another issue is that the peaceful AIs are allying way too many people. I believe I suggested before in one of the brainstorming/idea threads that allies of allies should have positive relations growth and acceptance rate, but enemies of allies should have penalties. Raven has been allied to me and Sharee even though we're both at war. I was tracking the relations closely before and there was no relations penalty for being at war with Sharee when they were allied to each other. Raven has 4 alliances that I can see already, and probably more with AIs on the other side that I haven't met yet.
Hostile Scouting and Pathfinding
The AI attack targeting behavior has been strange this game as well. Instead of attacking border cities, Sharee's doom stacks are making a beeline for whichever city has weaker defenses as if it's a neutral stack, and without having scouted any of them before hand. This is very annoying and feels unfair. They aren't using Volcano, but if they did, I would be quite upset in a real game that they don't need to scout. Even if pathfinding and volcano targeting is different, it's unfair that they can use attack target pathfinding with doomstacks to find my cities when there's a massive amount of ocean (like 20x20 squares) between us. It's just impossible to see them coming, and there's no way to deny "scouting" against a doomstack strong enough to take down a whole garrison. It's also annoying because I'm being forced to rush reinforcements to the targeted cities. While a human player would definitely target weaker cities, we still have to explore to find them.
Long Distance War
I didn't even have a chance to do any diplomacy as Sharee declared war within 2 turns of my unit first discovering her cities, and there was a 3 year period of nothing happening before their doomstacks suddenly arrived. This was probably a result of military rating disparity, but it feels ridiculous that an enemy more than 20 squares away on a faraway continent declares war so quickly then does nothing for years. When this happens it feels like human-player targeting, especially when combined with all those Pacts/Alliances Sharee formed with opposing alignment Wizards.
AI growth and AI vs AI
Finally, this new game since the update which fixed the double income/power issue seems to have evened out the growth of different AIs. However, I did notice that one AI steamrolled another one, conquering 5 cities easily because the defender's garrisons were bad, even though the overall power rating on the defender was significantly higher when the war started. Both Kali and Merlin owned approximately half the continent they were fighting over, and I owned a small piece of it. My contribution to the war was minuscule so it's not because I participated in it. But for whatever reason, even Merlin's cities had almost no ranged units or even summons, so Kali's sprites and a few odd cockatrices dominated. This happened even though Merlin also had 6 nature books to Kali's 5 and had both sprite and cockatrices too. Merlin just didn't prioritize improving its garrisons even though those cities were close to mine and Kali's cities, and neither of us had Pact with him.
