About the Poll
Poll time. Just to get a general feeling from players, I've added a poll at the top. Note that it's not asking if you think that another idea is better, it's asking how you feel about the existing idea. Here's why: we're not exactly looking for work at the moment. There's a ton of stuff we still need to get done, and if this works for people and accomplishes the needed goals, then -- for now -- we're done. It can be refined and extended and made more variant later. But if we try to do "every feature plus one" from the start, we'll never be able to get anything finished. So we start small.
And at the moment, if people are happy with what's here and it's not causing anyone aching misery where it shouldn't be, then this feature is going to stay as-is. Further changes or variants would be considered as an addition later. Our time is finite, please remember that. There are a thousand ways we could do anything, and the question here is whether there is enough active dislike of this one that it must be done differently immediately.
As much as some folks disliked control nodes, that wasn't exactly a crisis. They did their job and enabled new functionality until we later had the time (and revamped UI engine) to make the menus that replaced them. Plus we were collecting data on what those menus really needed to entail the entire time. And there were some people who lamented the loss of the control nodes when that happened, go figure.
Other points:
Providing intermediate goals is cool.
RCIX brought this up, and I think it's an excellent point. I'd been sort of thinking about this a bit when I implemented it, but not at a fully articulate level. There have been complaints from some quarters about how the lack of goals harms some people's ability to really get into the game. If we can provide a bit of structure for them without seriously impacting anyone else, that's a big win. And my read is that's what is happening now.
I don't buy the "I don't care about ARSes" argument.
Essentially "free" ships? As in, not costing knowledge? What could be more win? These are central to the game, and always have been. The fact that some players don't value them is... well, my read is that's been created by the culture of rushing and uber-low-AI-Progress that some advanced players have come by. If the AI progress were higher, you'd want all those ships. Put another way: if you just had to capture 8 more planets than you normally would choose to, I'd bet that 3-4 of those would be ARS planets in most cases.
I actually really do like the idea of AI Progress "gating" (that thing that Kemeno suggested), but I get the feeling you guys would have lynched me for introducing something like that.
Ideally players would be facing off against the home planet at around the 500 AI Progress mark or after. 400 at the very least. It does feel rather arbitrary if the players can just blow up 10 nukes to get to that AIP and then raid the AI early, though, I must say. That's one aspect I don't like.
Having to take certain planets does take away certain decisions, but it does also create other, new ones.
Some people rebel at the idea of being forced into take 4 out of the 5 ARS planets. It's seen as a reduction in choice. And, of course, in one sense it is. But it also opens up the following new choices once you have taken those planets:
1. Do you hold them or abandon them? They provide excellent income opportunities for metal/crystal, especially if you have higher-mark economic stuff unlocked. They could also provide tactical advantages, scouting launch points, and otherwise. Or not. But the choice is there for the player of "okay, now I have this planet: what do I do with it?"
2. So, now you have some random new ship type. What do you do with it? Your strategic milieu has just expanded in an unexpected and unpredictable fashion. Maybe you just throw the mark 1 ships into your mix of defensive ships. Or maybe this is a unit that changes everything about how you formulate your fleets on this map, based on what the AI has and what you are doing. It makes you play with ships you'd otherwise completely ignore, and -- surprise -- you might find some really new things to love while doing that. It's assumed that players will just pick from their stable of favorite bonus ship types, and thus they wind up restricting them to just a tiny percentage of ship types. With all the expansions, there are now 54 bonus ship types. The only way most people will explore most of those is by discovering them in an ARS. And so many of those ship types can be used in fascinating combinations people wouldn't casually think of from reading about them in the game lobby: again, something that only gets discovered via the ARS.
This is why I'm so fond of it making the players take the ARS planets, at least four of them. In 1.0 and thereabouts, you had NO HOPE of beating the AI home planets without taking 3-4 ARSes, and that was great. But as the game evolved, that's no longer true. And the game is better for it, in many ways (less grind, more interesting/powerful starships, etc). So I think of this as sort of a retroactive fix to bring back that one positive aspect of the 1.0-era design that has since been lost by the other many positive changes that have come since.
Of course these can't be optional.
For example, if AI Progress were optional, no one would turn that on. Who wants to be limited in how many planets they can take? Why not just steamroll every planet in sight like in every other RTS? That's way more fun, right? And originally in AI War, that's how it was. I wonder how many people would have quit the game, or threatened to do so, when I came up with that idea in alpha. I think more than would like to admit it. On paper it sounds like a terrible idea. But we all know that's one of the most interesting, central strengths of the game.
This is why I also insist that people actually try things. Unless you have a developer-level familiarity with all the intricacies of the game (and a few of our players come amazingly close, actually, but not many), it's hard to guage the sum effects of an idea like this. That doesn't mean you can't comment, or that your opinions have no value. But refusing to upgrade... well, I'm getting off on a tangent now. But my point is that some of the best ideas sound terrible at first, and that's why nobody else has done them. And then some of them really to turn out to be terrible, or at least just sort of bad. And others turn out to be bloody brilliant. That goes for things you guys suggest, too. Sometimes it takes me a while to really "get" an idea, and then months later I'm all "why didn't I think of this!?"
Being required to take a certain number of planets has some merit, but also many drawbacks.
So, this is sort of similar to the idea about having to hit an AIP threshold -- except that you can't shortcut it by just going nuclear. That could actually be a problem with some playstyles and some AI Types. And I should remind you that also my current implementation just requires that you briefly take the planet, not that you hold it.
So why do I feel this has drawbacks? Because if a player is used to taking just planets in a cluster of 8 behind a bottleneck, and we tell them they now have to take 16 planets, what will they do? They'll figure out a way to extend their cluster to 16, with one bottleneck. Or have two clusters of 8, each with a bottleneck, which is hardly better.
Referring back to my "this actually opens up some different kinds of choices" section, in the case of taking a distant ARS planet you have just incurred irreversible AIP, right? And you have a new ship for it. But now you also have this planet that may be in a weak position, that may require protection, and so on. It's not part of your core blob. You can just abandon it, if you like. But you paid for it. In an AIP sense, it's yours for free now. So are the resources worth it? Do you hold it? For how long? Do you abandon it and come back to hold it, briefly, later? You get none of this sort of interplay if you're always able to choose the planets you take.
And you know what? To me, this has always been a part of AI War. This is nothing new. From the start, you had to take ARS planets, and take and hold Advanced Factory planets, that would be out who-knows-where and possibly in terrible strategic positions. That's totally on purpose. Well, some players decided they didn't want to go through that hassle, and that they'd rather just skimp by on having a core network of planets; and the game evolved enough that became viable. That's something I see as actually being harmful to the game as it really reduces the potential for interesting situations.
Part of being an effective game designer is setting up those "wow" moments, and providing ways for them to come about naturally. Letting people just turtle in the corner drastically reduces the likelihood of those happening. The amount of choice that is sacrificed in service of this is sort of minimal, and it opens up other realms of choice. This seems like win all the way around, to me. And again: I think that if this mechanic had occurred to me in alpha, before people played the game without core shield generators, this would not have gotten any more comment than the AI Progress did. Granted that wasn't true for astro trains, but in this case I think it would have been.
More complex mechanics involving split-halves generators, or partial percentage-based protections, or so forth: yikes.
From a grognard's standpoint, which of course I am one, I see the appeal of those. Certainly it mirrors the complex shot damage mechanics that I myself had in the game until 4.0. But, yikes. Those are so complex, and in the case of the percentage-based ones require so much mental math.
If I can do 20% damage to the core guard posts, what does that really mean? Can I win? Certainly I know it will be much harder, maybe 5x harder (that's hard to judge, though I'll definitely do 5x less damage). But what does that mean in practical terms? I can risk attacking the AI and then them getting extra strong while their guys get stronger... but really, that's so high-risk I'll probably just go get the things that are reducing my damage. In short: I think it's a false choice. Almost no one would choose to attack early, and it makes it more complex for novice players to boot. I see the appeal and the thinking, but I just can't get behind that sort of thing at the moment.
In summary
If I had to choose a mechanic other than what's in there, it would probably be the "AI Progress gating" suggestion from Kemeno. But I really do think people would complain about that even more than this, and it doesn't really solve all the problems that the existing solution does.
And behind that, I think I'd put the "must capture X number of planets" suggestion from HitmanN. That's not a bad solution either, and it's certainly simple, but I feel like it only addresses part of the problem. And when it's just some random numeric goal, I think that feels way more arbitrary to people.
With something like a physical unit (the shield generators), it's a tangible thing that people can understand. Things A, B, C, D, and E protect thing 1. Kill those things, then thing 1. That has analogues in other games and in real life, and is understandable. And, heck: from a strategy game sense, it only makes sense. How many people here have multiple RAID arrays in their computer, and then still back up to maybe an iBook or similar external hard drive, and then possibly still back up to somewhere online? With the number of technical types here, I'd bet there are at least some (personally I just back up to online, but in several places).
Are you telling me that it seems unreasonable that an AI that has wiped out the entirety of humanity would have multiple layers of security? If I were Skynet, I'd never take any prisoners, knowing that a resistance comes up at some point. I'd send about 10 terminators back to get John and his mom, and then I'd send all the rest back to just before the start of the resistance to wipe out all humans in the slave camps and otherwise. Bam. Victory, with nothing lost from it. If I were the Empire in Star Wars, that Endor moon would have a base that was spanning miles and miles, with thousands of troops there guarding it -- and ideally, with a shield generator on some other moons or planets, too, or at least in multiple bases. If I were the buggers, and I saw Ender at my planet, I'd just fry every last ship of his as soon as I saw them, no matter what the cost. It wouldn't have been worth the risk with him being so close to my planet, even though what he did was "inconceivable." To be fair, those were aliens and not machines, so their way of thinking makes sense thematically -- I don't feel the book was flawed.
But in the context of these galaxy-spanning AIs in AI War, they don't have any thinking nuances like that, aside from thinking humans are pitiful, etc. And wouldn't some of that hubris come from having defenses that are SO multi-layered and far-flung that they should have no reason to fear anything? To me, that just makes sense.
Anyway, enough on the thematic stuff. Gameplay is king, here. I'm still open to other ideas, of course, and if there is just a universal unwillingness to accept my design I'll consider those by Kemeno and HitmanN. But hopefully this post clears up a lot of questions about why I find the various alternative designs to be not as effective as the current one. As with all things in AI War, it's many-layered and complex, and I probably left out some key points. But that's the bulk of it, anyway.
Poll