The other night I was musing over what I would've liked to have seen from the enemies in DMS - as much as I was disappointed in the general premise, I was doubly-disappointed to find that the enemies didn't bear any distinct new mechanics, and that as a result they felt like Vampires and Werewolves in name, only. sweatdrop
This may not be the most useful thread, but it'll essentially outline what I would change if I suddenly got the opportunity.
Vampires - Vampires aren't vampires unless they're drinking blood; how can we have vampire enemies that don't drain life? Additionally, they fell victim to the 'Everbody AoE!' method of making DMS 'harder', but I don't think the AoE attacks actually suited them. I'd be much more inclined to see them with high-damage extended DoT effects ('bleed', get it?) that self-heal (at a 1:1 ratio) as they deal damage. They wouldn't pose quite the same sweeping, immediate threat that Werewolves do, but once they stack a few bleed effects they become particularly hard to race - not to mention that they can continue to pose a threat after they're 'dealt with', making them feel like the more strategic of the basic enemies.
Werewolves - Unlike Vampires, I think you could reasonably justify Werewolves with AoEs; just look at those 'hands'! Indeed, it's hard to complain about big tough enemies that deal lots of damage not feeling like Werewolves, but within the context of zOMG! they're really just doing the same old thing; I'd like to see something that makes them particularly Werewolven. My first thought for this is a 'bloodrage' effect that increases the damage they deal inversely with their current health; that is, the more you hurt them, the harder they hit. Alongside the AoE, this can make 'weakened' 'wolves a significant threat if they're not handled quickly, forcing players to play around the effect by attacking in bursts. Alternatively, I considered a 'bloodlust' effect that would involve them dealing more damage to players as the target's health decreases - based on the 'bloodied' status from D&D - but that seems awfully redundant.
Night Frights - Now this is somewhere where AoE would fit on a Vampire; while it's not as obvious that these guys are bloodsuckers, I don't see any reason not to have them retain the vamps' lifedrain effect. I'd drop the DoT to replace it with AoE, since that already marks a distinct change in function. These would essentially require a tank, as it would be extremely difficult to keep up with AoE lifedrain when it's actually hitting multiple targets - let alone keeping up with healing those multiple targets. It seems a might simplistic, but I think the gameplay would be interesting enough without adding more bells-and-whistles.
Big Dog - Almost ironically, though, I did want some bells-and-whistles here; while 'wolves aren't known for being fancy, I wanted to try and fit the discarded 'bloodlust' concept from the Werewolves in an interesting way. I'd increase the damage considerably over normal 'wolves, but drop the AoE; the Big Dogs are the alphas, and they just want to fight the one player who's in charge. If they win that fight, though, your whole Crew's in trouble. The concept I was considering was having them automatically 'buff' themselves whenever a player they Hated dazes - the buff granting them AoE and footspeed boosts, among other things. They would probably retain the 'bloodrage' ability from basic Werewolves, meaning that if they took out your tank before you took out them, they'd probably just wipe the whole party right there.
Bloodbats/Counts - I list these together because I would've fused them together; I like the idea of capturing the vampire-turns-into-bat trope by having the Counts serve as the 'Aggro-form' of the Bloodbats. I would make the Bloodbats fast and dodgy - traditional goddamned bats - with high regen, and make them aggressive if-and-only-if their Health was over a certain threshold; then, I'd simply give Counts the ability to 'drop' Aggro when their Health got low enough, giving the effect of an enemy who retreats to a more defensive form when he's endangered. However, I'd re-use the daze trigger mechanic from Big Dogs for counts, to evolve the lifedrain effect from basic vampies; when a player Hated by the Count dazes, the Count regains all of his Health. Low-health Counts would be tempted to transform into 'bats, but similarly tempted to stay if their enemies were low on Health - if the AI were good enough, players may even be able to lure Bloodbats into transforming just for the possibility of killing them.
Kamila - Honestly, I don't know enough about first form Kamila to say anything; I know she takes a while, and something about coffins, but that implies that most of the fight is dealing with Counts, which I've already addressed. What I really want to talk about is 'Bloodlust' Kamila; she should be exciting, in a way that she's really not. She may be tough, but I get the impression that she's really just repetitive; not so much a puzzle to solve, as an interactive cutscene to sit through. What I'd like to see from Bloodlust is an enemy that isn't actually defeated primarily through attacking, and I think her concept could achieve that pretty effectively: simply make her attacks cost her Health. Thus, the battle becomes less about whittling away at her massive Health, and more about minimising the damage from the attacks she uses against you; you're aiming primarily to survive while she kills herself. I wouldn't increase her level so high that you couldn't damage her, but I don't think direct assault should be the most viable tactic. Assuming her attacks also retain the vampires' lifedrain quality, this makes each of her attacks a gamble for her, and therefore an opportunity for the player. If I recall correctly - though my recollection may be skewed by the concept images thread - she has three basic attacks.
Big Stabby Hand - This one's pretty straightforward, but I'd like to see it with a more lasting effect; you know I love my extended DoTs, and I'd love to see this one as the granddaddy-of-them-all. Seriously, like a 10-minute DoT, with lifedrain; I'd retain a slightly larger 'immediate' damage value (like Hack), but the massive part of the damage should be dealt over-Time. Interestingly, this makes it where you have to heal the affected player, but don't want to - because that healing just goes back to Kamila. Outside of a boss fight, this would be uncomfortable tension; inside a boss fight, the feeling of inevitability just fits. I've back-and-forthed on whether this should be single-target, or a line-AoE - currently, I'm favouring the latter.
Teleport-whatsit - I'd like to see this one as an circular AoE, that punishes players for trying to stay too close. Even dealing fairly moderate damage, AoE lifedrain effects can be really oppressive; this will force players to spread out a bit, where she can still hope to catch a couple in her line-AoE.
Bloodbat Summon - This is the most interesting of the three effects, because unlike the other two, she's not dealing damage - thus, no lifedrain. This actually makes it a significant weakness for her, because it's the only 'gamble' she can't win; unless her allies actually kill you, she's paying life for nothing. However, it can force players to change formation, making them susceptible to her other attacks. It would be used primarily when players are spaced too far for her circle-AoE, but not lined-up such that she could snag more than one in her line-AoE; with careful positioning, players could reasonably force her into using this, her 'worst bet'.
As I said, I was simply musing over possibilities - I'm sure there would be some additional considerations necessary for making them actually play in a cohesive manner. But what I'm really interested in seeing are enemies that present new puzzles for players to solve, rather than just necessitating 'better' versions of the same old solutions. I may consider doing more instance-reform thought experiments, as well, and naturally any commentary or suggestions are always welcomed...
