Whew, okay:
1. Sorry to anyone I have offended here. I understand that my last post or two were pretty brief (on individual responses to points, not overall length), and may have come off as dismissive. That was not my intent, but if you see the latest topic today about the schedule and so on perhaps that will give some more insight into my own state of my mind.
2. At any rate my intent is not to stymie discussion and most
definitely not to hurt anyone's feelings, though I realize I may have done so. But we're passing a point where I think we are at "good enough for a first pass, and I bet things change during the alpha quite a bit, as is only natural."
3. When it comes to things like the leviathan class being only weak to other leviathans or whatnot, bear in mind that basically any class can defeat any other class... but it's dependent on how much you wind up throwing at it. If you have 5x as many bombers as a tac-sup group, you'll absolutely wreck the tac-sups by sheer numbers alone... and take some heavy casualties in the process. It wasn't efficient, but hey you got the job done. If you outnumber tac-sups 20:1 with bombers against them, you might have so much overwhelming firepower that you just demolish them before they have much time to even do anything.
What this chart represents is "like for like" combat. Think of this as "serving size" in food. You can get far more calories from lettuce than a Big Mac if you're willing to eat that much lettuce. But eating one serving of each leads to incredibly more calories from a Big Mac. If you are dead-set on eating only lettuce as your primary source of calories, then you could certainly do so, but it would be inefficient (and I'm obviously ignoring vitamins and whatnot that you'd need from elsewhere -- just talking raw calories).
4. The artificial bonuses are something that people have argued about back and forth for seven years now, and the conclusion for years has been that it doesn't work without them. Basically if you take away the artificial bonuses of types versus other types, then pretty much every ship suddenly has a relationship to every other type of ship. It gets to be a bit nuts, and is a balance nightmare, because even the most basic of tac-sup ships couldn't possibly be good against all the other things without also being too good against what it is supposed to be weak against, and/or having lots of strange exceptions. The natural non-bonus-based approach works super elegantly when the number of units is very small and can be polished to perfection, but in something like AI War (or many other RTS games) it gets to be a nightmare.
5. Star Wars as a great example of big ships being ineffective against little ones is a good point, when it comes to Death Stars, though that was largely a design flaw issue more than anything else. I was thinking more of Star Destroyers, and in particular how those are portrayed in some of the strategy and space sim games. Going a bit off the road of movie canon there I guess.
6. Trying to limit things too far down in terms of things like "all siege weapons are long range and {insert whatever general trait for them}" also artificially reduces variety. Going medieval, you have things like the trebuchet that is long range but has to be unpacked. You have battering rams that are incredibly short range and also slow, but powerful. You have catapults that are middle-range and middle speed. And so on. This is very interesting, because it's several different sub-ways to solve the same overall problem. I don't want any triangle to get in the way of the ability for these sorts of things to exist.
7. The limitations of things like special perks to just some classes also goes against that to a large degree. Keeping the "bag of perks" to a minimum is a good goal, I think, though. But there has to be a "miscellaneous" category, in my opinion. That may well disappear in alpha during testing, so I'd just leave it for now, because the proof will be in the pudding. I'll be plenty happy to be proved wrong if there's a better model, but overall people seem to be reacting pretty positively to this one, and so this one seems to be a good one to organize around for the moment.
8. I think that all the major arguments that the majority of people have had have been addressed, though there are still some specific arguments from specific people about some parts of it; with something like this that is still an on-paper issue, and that is under a time constraint, I think that's honestly the best I can hope for.
If there's anything incorrect in my perception there, please do let me know. To be clear, this is the model I'm referring to:
