Proposal: remove metal

[kasnavada]

Hmmm... it's difficult in the sense that all AI response times has to be balanced around it. Why not simply make everything tick four times as fast as classic Normal speed? There is a lot of leeway between  the existing AI War Classic timing windows of minutes to hours, and the limits of professional gaming (around 10~60-second in Starcraft 2, which we won't want).

This ties with the "just multiplying metal creates issues on its own" idea I stated before.

A "simple" example:current AI War, a cap of unit needs a few seconds to build, but, if boosted by enough constructors and metal ressources, metal usually runs out if sent to the front, and if they die "as much as they" live. With more metal, and nothing else, one could send fighters, and possibly other ships, blindlessly for hours as long as metal allows, which is always if metal is simply boosted. Because, metal, funnily, acts as a time-gate to prevent the game from being fast, and by forcing "complete" rebuilding / refleets to be long (if unable to stock enough of the stuff).

Hence, why I proposed a limitation of "XX time to respawn". Which becomes the time-gate and removes interference from metal.
I had tried a few set-up based on this "build until it falls" idea with neinzul ships which worked very well to overwhelm the AI's defenses, or to defend my worlds. I do consider those "border-line exploits"... as with "low" amount of metal I was able to overwhelm stuff I couldn't beat otherwise, by throwing dumbly things at it. The very point of that limitation is to leave (some) time to allow the AI to crush your stuff. Until you can come back and tell it that the party's over.

unclear

S***... I was unclear in my second proposal, and named "engineers" what I should have call "constructors", and mixed & matched. Damned. I'll rewrite that.

Quote
    Model 1: constructors limit max player fleet size, fixed refleeting time.
Meaning if a player has progressed to 10 planets and lose 7 in a tough fight, they can rebuild only 3 planets' worth of forces against 10-planets worth of continued AI aggression, until they retake more planets?
 If this is what you mean, it's a very-fine balance (to put it lightly). With metal, the players can stockpile reinforcements before they lose the 7 planets, and have an larger effective army than what they have left in terms of planets to make a comeback. Without some form of buffer, snowballing by AI is likely (downward spiral) as you will lose constructors and therefore fleet capacity. (BTW, most RTS/4X games show this phenomenon, since fleet size and upkeep *is* generally a function of empire size. Only one interesting battle per game.)

If you let the AI beat you enough so that your empire is 70% lost, then its next strike kills you. No comebacks.
In the particular concept of AI war, I think that each defense against a CPA & an EXO is an interesting battle, but not one that is supposed to leave you half-dead.
Finally, what you say is true of AI war now, since metal reserves are kept by command stations, and lost if the command station dies (if the reserve is high enough). My issue with it is that to make reserves, you have to do nothing for... a long time. Which is the point that bothers me. The waiting.

rest

Not sure I follow here.

What I'm meaning to convey is actually that I'd prefer a system where decisive attacks are more present than currently. And where "failed" attacks, on both sides, have lesser netflix time effects. By reducing the number of units that can be sent "until the planet falls", like in the current system, and preferring "bigger battles", in which the player or the AI, instead of dealing death of 2 thousands strikes, exchange planet-sized blows. That means that both side "repair" faster, so attrition "victories" are is less of a concern, but both sides can attack more often.

So, basically probably not a "liquid" war, not that I'm sure of what it means. Rather making the game "closer" to trading heavy blows, "turn-based". But with no turns.

[x4000]

I think that the amount of refleeting time should be irrelevant, but not because it doesn't exist or is shortened.  Make it not exist and you change the game too much to be recognizable.  Removing metal isn't something we could remotely contemplate because it would alienate a ton of existing players and it is just too fundamental to how many strategy gamers are used to seeing games work.  It's too radical for this sequel, though it is an interesting idea.  I could definitely see a game based around it, and I understand the design goals behind it.

My counter would be that, while you're refleeting, you shouldn't be completely focused on just that.  Aka, one of the interesting things about certain turn-based games is not that individual turns are short or that things take a short number of turns, but rather that you have several things going on at once.

People recognized during the course of AI War Classic that metal and crystal are basically the same thing, and so they were condensed.  Fair enough.  But that actually gets at the point, I suppose: they are the same thing because there's only the one thing.

People like champions because it gives you a minigame to play while you're waiting for things like refleeting.  The fact that we added something like that is another sign that something is up.  I don't like that it is a separate minigame and something optional to the game.  That makes it too easy to ignore and to instead sit there in netflix time.

A true solution would, in my opinion, center around:
1. Leave metal and refleeting alone.  It's familiar and works.
2. Add in a second resource (crystal or whatever we call it) that is a 100% separate economy in every way from metal and the fleets themselves.
3. Add in some other mechanics that rely on crystal that can't at all help your main fleet-based activities, so there's no incentive to get those guys to help your main fleet battles (unlike champions).

Then we bounce back and forth between two overall things:
1. Attacking and defending with fleets, during which time crystal and metal accumulate.
2. Doing "stuff with crystal and the things crystal paid for," during which time refleeting happens.

And the game is based around these things all being a part of the core experience, not having one of those as a minigame or whatever.  My first thought is to make these crystal things be non-combat ships that work with minor and background factions in various ways.  Aka the way that you "secure background factions" and get new ships isn't metal-based, and doesn't involve your fleet.  You can focus on that when you're not fighting, and forget it when you are fighting.

That's just the germ of an idea, but I think it gets at what everyone is after here: familiarity, less boredom, and (in my case) not resulting to in-game minigames to avoid the netflix side of things.

[Pumpkin]

And the game is based around these things all being a part of the core experience, not having one of those as a minigame or whatever.  My first thought is to make these crystal things be non-combat ships that work with minor and background factions in various ways.  Aka the way that you "secure background factions" and get new ships isn't metal-based, and doesn't involve your fleet.  You can focus on that when you're not fighting, and forget it when you are fighting.

What about adding some planetary combat in the game? Something that would be handled a bit like TLF does. That would still be a minigame with few connections with the grand game, though. I imagine some sort of "guide troop transport to target planet". Something to do "when you have time" and cost no metal/refleet. Or something more "on the planet" with a dedicated UI and choices...

Eh, very raw idea, but if someone feels like picking it up and improving it...

[kasnavada]

Then we bounce back and forth between two overall things:
1. Attacking and defending with fleets, during which time crystal and metal accumulate.
2. Doing "stuff with crystal and the things crystal paid for," during which time refleeting happens.

I'm not sure I like the second ressource thingy, but... short of a mini-game, I'm not sure what can be proposed here. That said. There is a possible mini-game that could be envisioned thanks to you setting up minor factions.

How about crystal takes care of buildings (which, ultimately, there is no more things to build with it), and limited diplomacy options ? I'm stating limited because I don't think that "advanced" diplomacy is a good idea.

I'm thinking of adapting something that already exist and work, so average to low risk. http://wiki.starruler2.com/Diplomacy_and_Influence .

In a few words, diplomacy in that game came in the form of "cards", which you could call treaties in AI War 2, that can be bought and applied by paying influence (basically what you're calling the second ressource here). There, said treaties have different effects, from taking over a "neutral" system, getting ships, hiring mercenaries, metal, having them attack / defend particular areas instead of you, just plain reinforcing, paying a tribute so the player does not get attacked himself, and so on. All of those actions would be limited in time, so you'd have to rebuy them. Oh, and, compared to starruler 2, ditch the "vote" part.

Each encountered non-hostile minor faction could have a treaty holder with a few options that can be bought, to help you prepare the next assault and / or care about stuff you can't do.

[x4000]

I like the idea of planetary combat being something you have to pay for separately with this, and I also really like the idea of using a cards-style approach to getting said third parties to do things.  That would be a really fitting blend of AI War and TLF in some respects, and give players something to do during the fleet rebuild time that definitely isn't busywork (namely, finding and arranging the best deals to help gain third-party resources for the next fleet rebuild, shoring up defenses on planets that are embattled, etc.

This dovetails nicely with one of the conversations from yesterday about core fabs and related.

This could be... rather huge in a very good way.  The sort of thing that is comparably easy to program, that we have experience with thanks to TLF (and Star Ruler 2 as another example), and that would be ripe for modding.

[Tridus]

I think that the amount of refleeting time should be irrelevant, but not because it doesn't exist or is shortened.  Make it not exist and you change the game too much to be recognizable.  Removing metal isn't something we could remotely contemplate because it would alienate a ton of existing players and it is just too fundamental to how many strategy gamers are used to seeing games work.  It's too radical for this sequel, though it is an interesting idea.  I could definitely see a game based around it, and I understand the design goals behind it.

My counter would be that, while you're refleeting, you shouldn't be completely focused on just that.  Aka, one of the interesting things about certain turn-based games is not that individual turns are short or that things take a short number of turns, but rather that you have several things going on at once.

People recognized during the course of AI War Classic that metal and crystal are basically the same thing, and so they were condensed.  Fair enough.  But that actually gets at the point, I suppose: they are the same thing because there's only the one thing.

People like champions because it gives you a minigame to play while you're waiting for things like refleeting.  The fact that we added something like that is another sign that something is up.  I don't like that it is a separate minigame and something optional to the game.  That makes it too easy to ignore and to instead sit there in netflix time.

A true solution would, in my opinion, center around:
1. Leave metal and refleeting alone.  It's familiar and works.
2. Add in a second resource (crystal or whatever we call it) that is a 100% separate economy in every way from metal and the fleets themselves.
3. Add in some other mechanics that rely on crystal that can't at all help your main fleet-based activities, so there's no incentive to get those guys to help your main fleet battles (unlike champions).

Then we bounce back and forth between two overall things:
1. Attacking and defending with fleets, during which time crystal and metal accumulate.
2. Doing "stuff with crystal and the things crystal paid for," during which time refleeting happens.

And the game is based around these things all being a part of the core experience, not having one of those as a minigame or whatever.  My first thought is to make these crystal things be non-combat ships that work with minor and background factions in various ways.  Aka the way that you "secure background factions" and get new ships isn't metal-based, and doesn't involve your fleet.  You can focus on that when you're not fighting, and forget it when you are fighting.

That's just the germ of an idea, but I think it gets at what everyone is after here: familiarity, less boredom, and (in my case) not resulting to in-game minigames to avoid the netflix side of things.

This makes the most sense to me. Refleeting itself is at its core a good thing: if your entire fleet gets blown up, you *should* be vulnerable while you build it back up again, and the AI should exploit that. Granted, the AI doesn't do a great job of exploiting it in some cases.

Having other stuff to do that doesn't then get lumped into "add this to my fleet, and thus lose it during refleeting time" is a great idea, because it fills that time while still leaving you vulnerable.

[PokerChen]

(previous discussion:) okay, I understand it better now, there are a minority of games that I have played using purely time - based logistics, although they ended up as too-simplistic IMO. I brought up the case of 70% losses as comebacks from this point is generally possible in Classic (because of relative lack of AI finishing blows).

(New discussion on secondary resource:) I have some reservations about using a diplomacy resource, since the design doc elsewhere said that the AI will gobble up various backgrounds factions, which you are mostly unable to stop. If a given game is not going particularly well (and if you're refleeting frequently, this is likely the case,) how many factions will keep surviving for you to use this resource?

I could imagine using the existing hacking instead as a downtime activity, as an expression of covert operations that are fundamentally incompatible with fleet action. Say, for example, a number of more powerful hacking options require  both that you have a presence on & connection to the target planet, and that the AI planet is not alerted to your presence.  Then fleet action and covert action can play off each other: your first strike inserts a cloaked team deeper in AI territory, which you switch attention to while your main fleet regroups. A successful sabotage of AI defences make it easier for the next renewed offensive.
It's one of those things that I thought was a bit off with Classic. If the local AI is aware of enemy presence, should it not switch to secure band communications and more stringent firewalls?

EDIT: the cards thing from star ruler 2 is a worthy idea, but I think there'll be games where you run out of allies to play cards with.

[kasnavada]

@Zharmad
Or, tie all 3 systems.

At first, secondary ressources is used for buildings, mostly. Mid-game, there are still minor factions to play around diplomatically, so it's your "main expense" of secondary ressources. Late game, minor factions are gone / converted to either cause, or simply don't scale enough to matter anymore or something, and it's mainly used for hacking. You'd have to change hacking so that it's scaling "badly" at "lower" level and most useful for late game actions.

And / or, one action you could do with "loads" of the secondary ressource is "Revive a minor faction".

[PokerChen]

On that front, I'm reminded of the various 2-sided cards in a few blocks of Magic: The Gathering. Some of these can be used for either of its card's functions. Others are start on function-A and are converted to function-B depending on conditions.

Might be useful if the game did adopted card-based systems. Side-A: Communications network. Play on a neutral or friendly human background faction to increase its fleet size by 20%. Side-B: AI communications jamming. Decrease mobility and reinforcement of adjacent AI-owned system. However, this increases the AI's desirability to conquer the host human background faction.

[Captain Jack]

I don't like the thought of planetary combat because it seems unlikely to fix the problem of refleeting downtime in its current form. To make it work as a strategy game you have to create dedicated units and rules separate from the space game. This looks a lot like nebula scenarios 2.0. If you go light on mechanics and work in cards, we've introduced mobile game mechanics. Neither is something we should be doing without a coherent design in hand.

Paying the human remnant factions to do it for you is not a solution because you can pause the game, assign resources there, and watch it autoresolve while your fleet builds up. No reduction in netflix time.

[kasnavada]

Paying the human remnant factions to do it for you is not a solution because you can pause the game, assign resources there, and watch it autoresolve while your fleet builds up. No reduction in netflix time.

In Star Ruler II, they thought of that issue already.

1) You've got the choice of buying up to players + X actions (depending on upgrades in your colonies), the action that is "farthest left" is cheap, while others, more on the "right", are more expensive the farthest they are. Each time you buy something, it leaves the queue and one to "more" new cards show up at the time of the next "refill". If you don't buy something, it eventually makes it way "left" and it leaves the queue after the next "refill" (I don't remember the exact time, but 30 seconds or 1 minute).

2) What you're actually buying is future actions, not immediate actions. You're buying the right to do something later. That's when you need to pause, when you plan your actions on / with other people. Which you would have done anyway, if only to send your fleet.

Therefore, you "can" pause when browsing the shop, but it's pointless, so reduction in netflix time is on.


3) Deploying actions also cost influence. As you spend your actions before combat (or as they start a fight, possibly), you empty your reserves. That means that during "combat time", you will not have the ressources to buy cards anyway and don't need to pause and browse.

Edit : clarifications, and adding 3).

[Captain Jack]

Are you talking about handling the planetary defenses in person? That's not what the bit you quoted was about. I talk about why I doubt the planetary combat prior to that.

The proposal you seem to be making is that you can spend resources ahead of time to buy a planetary defense mini-game to do when you lose a system. Is that right? If it is, we've fallen into the issues I brought up in the first part of my post.

[kasnavada]

Hum, no, just the card / diplomacy idea I spoke of a bit before.


I have no clue about planetary engagements, nor how they would work. It's pumpkin's idea. I don't have an opinion on that one. I thought you were answering both.

[x4000]

I've been thinking about both planetary combat and the idea of dealing with the allies via this mechanic more, and I've kind of cooled on it.  There are a whole lot of potential moving parts to that, and visualization and organization is among the forefront of those.  But dealing with the factions that directly and that corely to the game also just doesn't feel right to me: some people definitely like the sense of isolation of AI War, and in a high-difficulty game I feel  like the potential for all the factions to get murdered should be available.  Plus playing without them in general I also think should be an option, for the same reasons.

Not that this means that I want to downplay factions per se, but I also don't want the entire game to integrate them THAT heavily.  At core, this game is about a conflict between you and the AI.  It should be able to be narrowed down to just that, or have other things going on as well, at your option.

Exploration and discovery is certainly a core part of the game that isn't conflict-oriented, but to me that would be along the lines of derelicts.  Tying crystal into that would be fine with me, for example.

THAT said, I think that the core thing here is hacking that interests me.  Think about it: when you're talking about a foe like this, there really would be two battlefields in a lot of respects.  One is the physical, and the other is the information/digital.  If you haven't read The Last Firewall, I highly recommend that whole series (start at the start, though).

The thing that I largely don't like about hacking in the current version of the game is that it's so numbers-based in a way that is just another form of resource management, ultimately.  There's nothing wrong with that, but it doesn't say "hacking" to me.

I think what I'd like to do is actually include a secondary battle layer to the AI where you can actually see kind of a holographic view of the AI's systems and then mess with them.  I'm actually even MORE tempted to make it a command-line interface because that makes it feel waaaay more hacker-y.  But also possibly a lot less accessible.

At any rate, I could see crystal being used for things like scouting through the AI network (why send physical scouts?  That seems like a horrible idea.), gaining insight into what the AI is planning, and even reprogramming or rerouting waves in extreme cases.  Can you imagine an incoming wave being reprogrammed such that it actually comes out the other end as YOURS?  Talk about an AI War (*cackles*).  The command line interface is what excites me the most as a way to do this, but I'm not sure how others feel about it.  It's the most flexible by far.

Heck, we actually could probably build in some simply query language type stuff where the programmer-oriented type folks could run some automated checks against various data they have scouted, and plan strategies based on that sort of thing.

For that matter, potentially that's how you meet other factions, who presumably are somewhat in hiding after the early game if the AI takes over: through the network.  Having the network traffic open in a sidebar that you can open and close would be pretty slick.  I'm kinda worried that non-programmers would be too intimated by the command line to really make use of this part of the game, though, so making it central to the experience would potentially be risky.  I guess we'd need a visual interface for it at the very least.

[Captain Jack]

some people definitely like the sense of isolation of AI War

*raises hand*

At any rate, I could see crystal being used for things like scouting through the AI network (why send physical scouts?  That seems like a horrible idea.), gaining insight into what the AI is planning, and even reprogramming or rerouting waves in extreme cases.  Can you imagine an incoming wave being reprogrammed such that it actually comes out the other end as YOURS?  Talk about an AI War (*cackles*).  The command line interface is what excites me the most as a way to do this, but I'm not sure how others feel about it.  It's the most flexible by far.

Heck, we actually could probably build in some simply query language type stuff where the programmer-oriented type folks could run some automated checks against various data they have scouted, and plan strategies based on that sort of thing.

A strange game. The only way to win is not to play. 

But seriously, the command line is fine, just include a way that the players can have certain commands input for them. Buttons on the side or something.

At any rate, I could see crystal being used for things like scouting through the AI network (why send physical scouts?  That seems like a horrible idea.), gaining insight into what the AI is planning, and even reprogramming or rerouting waves in extreme cases.  Can you imagine an incoming wave being reprogrammed such that it actually comes out the other end as YOURS?  Talk about an AI War (*cackles*).  The command line interface is what excites me the most as a way to do this, but I'm not sure how others feel about it.  It's the most flexible by far.

This was one of the things I wanted to talk to you about. I'll throw up a dedicated topic or wait for our email, but basically a lot of the changes to the AI that have been suggested like turning off the AI's omniscience could be managed in-game. You hack into the AI and turn that off, and the AI has an initial freak out, then starts doing things differently. Taken even further we can build a hacking based victory condition where you seize control of the AI.