I also like the idea of mining stone from the mountain instead of from little patches. Corran is right when he says that harvesting wood consumes imp time more than anything else. But I don’t think that this is a bad thing, because time is a resource, too. So, I don’t think this would make stone “useless in terms of gameplay,” if the amount that one stone tile yields is made proportionately small.
Also, regarding imp time, I think that imps could collect more efficiently than they currently do. If they can carry infinite amounts of stuff, then they shouldn’t make a trip back for each thing they pick up.
Now having said all THAT, I could be completely wrong, and maybe an RTS system would be just fine.
In that case, I’d just suggest the following:
Therefore, if you want to move really fast, box your entire team, right click at some point on the map, and then just hold down the space key until they all get there. Or if there is a tree-hierarchy, just click the team leader and his minions will automatically follow him.
I’ll try to refine this idea some more…
And just some other points
Telepathic connections: Yes, I mean shared vision. But in general I mean shared detection, so that if some unit can sense some property of another unit, then that information is also shared.
Re: coming outside of your base to defend.
Okay, what if some stuff was outside your base, NOT defendable by boulders. Like maybe there are just mines sitting in the middle of the map instead of buried in the mountain. To get the minerals out of the ground, you have to erect a mining apparatus (buildable at the workshop maybe, maybe not). Anyway, the point is that this thing is really expensive (in terms of the time it took to build) and you’d rather not have it smashed. You CAN hide in your tunnels, behind your boulders, but it’ll cost you.
Also recommend that boulders completely smash everything they roll over, turning whatever equipment is being carried into component resources.
So corpses are brought to the ritual room, minions work on them, and only then you get mana? Certainly interesting. One problem with it is that it stops your progress if you don’t have any enemies to kill at the moment. It’s also harder to balance, which was the reason why I removed mana gain from simply killing enemies.
Well, the mana-generation comment was just a minor thing; the main thing I was suggesting was to use the ritual rooms as factories so that this factory motif would be repeated more. I was also going to suggest (but I didn’t) that corpses, when processed, could liberate secondary resources called “humours,” (i.e., sanguine, choleric, melancholic, phlegmatic biles). (These could possibly also be easily convertible to mana, but maybe not; I was even thinking about making “mana” simply a composite value of these four humours). Anyway, different rituals would require different humours in various quantities. (These humours could also be liberated from mushrooms). I wrote more about how rituals could work here: http://steamcommunity.com/app/329970/discussions/3/305509857557053068/
The real point of all this is to get different systems of the dungeon interacting with each other in sort of a network-like fashion, not to change how mana is produced. Macro-strategy-wise, the mana generation system is not that big of a deal; I just wanted it to be corpse dependent so that waiting around was not the best strategy.
=== Rant mode (only read if extremely bored) ===
Here’s my mana-related rant, since you triggered me by writing “game balance.” (I’m just talking about campaign mode here, don’t really care about single-map mode).
If the game is “hard to balance”, I think it would be because the starting conditions are so extremely variable. In some games you can start next to four tribes that can kill you at the blink of an eye, and in others you can kill everything in sight. It seems to me like a tall order to balance a rogue-like when the typical power curve has such a high standard deviation. (And this is compounded by trying to extend the initial conditions to include the single-map mode). It’s not as bad as it was a year ago, but I still think the game is basically stomp or be stomped, rather than winning by a long sequence of perfect decisions as a Chess grandmaster plays his game.
So, there are often long periods where you just sit and wait, because that’s all you can do when your enemies are powerful. Ultimately, I’d prefer it if the game forced me to do interesting stuff; if the base construction were more interesting, like a puzzle a la Rymdkapsel or if it REALLY taxed your resource management skills, then I’d not be so insistent on this, but the base construction is just busy-work so it gets boring after the third time you play the game.
tl;dr: So, if we started on the left hand side of the world map and tribes got progressively more difficult, then there would always be something fun to do (i.e., killing something).
Okay, now on to mana. The way I think mana could work is like “action points,” and I think it should only be acquired from going out and “doing stuff,” either killing enemies, stealing things, or maybe solving those little mario-party mini games like sokoban. To me, EVERYTHING in the game should be dependent on mana (in addition to whatever other resource): building items, rooms, even digging tunnels. When you do this, then mana doesn’t become hard to balance, it becomes the rate-limiting factor; the currency of balance, if you will. It helps to put a crude, time-independent upper bound on how powerful the player can grow. (I previously noted that gold sort of acts like this because it limits recruitment, but mana seems like it could do this for all systems in the game. It was also sort of why I’d suggested that a constant supply of wood should be required to keep the dungeon torches lit: progress requires economy, economy requires imps, imps require mana.
tl;dr: Make crude limit on maximum, time-independent player progression easy to balance by ultimately tying every operation to one scalar value. Arbitrarily choose mana to be this value, and very tightly control the maximum mana which can be liberated in each sector on the world map.
Basically the whole point of this suggestion was so we don’t have to play the waiting game as much. But, I guess it’s not a HUGE deal. Just my 2c.
Team tactics have been on my list forever. Probably not as complex as you describe, but for example assigning roles to members (for example healer, archer – they would try to stay behind), and simple orders like stay, attack, etc.
Thought about this topic again as I was playing today when an orc shaman healed me. I think this kind of automatic action defined by role synergizes with the rogue-like control scheme. As you say, whatever team-tactics there are should primarily be role-based.
Made me think, what if, in the team assignment screen, BEFORE the battle starts, instead of just assigning teams as a list of units with one leader, you could assign a team as as tree of units in a hierarchical chain of command.
Why?
– Trees allow you to issue orders recursively; during the battle, this means orders can be given QUICKLY, so that the game doesn’t drag. It should allow the player to affect the battle without giving him explicit, RTS-style control, which would violate the rogue-like paradigm (that I prefer).
– As a corollary, most of the “team strategy” is performed, declaratively, BEFORE you embark on your mission (or, if you’re defending, as soon as you notice that you’re being attacked).
– Lists of units have a complexity of O(n), where n is the number of units per team. Trees have an complexity of O(exp(n)). So, there’s potential for a LOT of strategy here, particularly if you code the game so that certain units work well together, or certain items (e.g., staff carried by a subordinate) have a synergistic effect on a certain unit (i.e., a team leader).
– It’s more fun to arrange your units in a hierarchy than to just dump them in a list.
How?
Rules, in decreasing importance. (Rule 1 is necessary, Rule 2 might not be necessary, but it’s more an implementation strategy to make units act naturally, and Rule 3 is not necessary if you don’t want to overload the player)
1. By default, a subordinate (i.e., child unit) will attempt to FOLLOW his IMMEDIATE superior (i.e., his parent unit), UNLESS the parent gives an order to move somewhere on the map.
Thus, if you are the keeper, and you have two legendary minions with their own squads of 3 greenskins, you can simply divide your forces in half with two attack orders. Otherwise, everyone will just follow you, because the rule to follow will cascade down the chain of command.
2. A unit will preferentially AID mates (i.e., heal, remove poison, cure burns, target attackers of the mate, attack targets of the mate) in the following (recursive) order:
– parent
– all siblings of parent
– all subchildren siblings of parent
– if parent is team-leader, then done, else recurse with parent of parent.
In other words, a little group of 3-4 sibling units will act as a “fire-team,” preferring to aid heal their captain and squaddies, but they will not neglect anyone else in the team. However, they WON’T abandon their immediate leader (rule 1) in order to aid someone else (e.g., if a team-mate in need is halfway across the map).
3. (Optional: this might overload the player) When you arrange your team, you precisely specify the roles. For example, a wolf could be the subordinate of an orc, and the wolf could have a role of scout, attack, and defend keeper, in that order.
(Optional: this just might not work well) The number of subordinates that a unit can control is dependent on its experience level (except for the keeper, who can control infinitely many units). The rule is as follows:
The sum of the experience levels of a unit’s subordinates must not exceed the experience level of the superior unit. However, with each level of indirection, the experience score that counts is halved. So, if an legendary humanoid controls an orc, and the orc controls a level 2 raven, then the raven counts 2 towards the orc, but only 1 towards the legendary humanoid.
