Designing challenging mobs

[BigErkyKid]

Hi,


The devs have an issue: either the mobs seem to easy or they one shot us. They don't seem to be able to come up with something that both challenges and entertains us. So as many have already said, the current end game (as of the latest lama) "challenges" us by forcing us to cheese the raids exploiting the ranged advantage. So how can they design mobs that target different kinds of builds? I guess that one answer is by choosing a variety of mobs that target specifically different weaknesses. Another is by coding more effects, but this has already been discussed.

For example, consider the following main types of defense:

1)To-miss: basically given by dodge, since AC is quite borked (although it does give something):
2)PRR/DR
3)MRR/evasion
4)HPs: not a defense per se, but has the same effect
5)Ranged/contact: how does the arquetype primarily deal damage?

We have builds that combine different kinds of "defenses". For example, barbarians are 2)4), rogues 1)3).

Now the question is what kind of attacks challenge differently every type. There are easy ones: magical damage favors 3). For instance, I was running the new Inn quest in heroics with a barb and I was getting vaporized by the abishais.

But what are the basic statistics of an attack and how do they interact with the different defenses? The stats are the following:

a)Speed/DBs (one and the other are quite similar)
b)Strength (as in Damage per Hit)
c)Source (magical, physical)
c)Type (ranged, contact)

Now for the sake of showing how the analysis would run, consider the case of two typical archetypes: i)is a dodgy rogue, ii)is a sturdy THF fighter. Is it possible to design different types of physical attacks that favor one over the other? Or is it all doom and the rogue will always be worse at defense? I think it is hard but possible.

As an extreme example, consider a rogue with 50% dodge and a fighter with 50% damage reduction (via PRR). Let the two have the same HPs (say 1000) and those be the only relevant stats for the combat. Now make them fight a mob that hits in DBs for 1000 a pop.

The outcome of the battle is that the fighter dies for sure (2000*0.5=1000) whereas the rogue dies in some scenarios, but not in others. The probability that the big mob does not kill the rogue is equal to the probability of not hitting in any case: .5*.5=.25. That is, 1/4 times the rogue will stay up.

Now if instead of this big hitting mob we used a swarm of critters that don't hit for much, things would turn around. Say we have 5 criters hitting for 200 a pop each. It would take 2 full rounds of attacks to kill the tanky class, but it is possible that they kill right away the dodgy class. If he/she misses every dodge roll, he/she will die.

Everything else equal, big slow great hits favor those with chances of survival via dodge, and small swarms force dodge types to roll a lot of dices and increase the odds of bad rolls and death. Whereas PRR types can keep damage down to manageable levels without sudden spikes.

Again, this is just an example of two extreme encounters that would challenge with differently two iconic archetypes. It would be possible to get a lot more detailed and craft more realistic stuff, but this is not my job. My point is that to design good challenge mobs must be of different types and significantly target different weaknesses / strengths. As we stand, the truly challenging mobs are just DBstriking amazingly hard hitting melee mobs that come in packs.

If you want to build good content, start by identifying properly what kind of mobs is good against what character and introduce a balanced combination in quests. The current BOOM BOOOM is just forcing as play ranged, but it is not welcome by us players.

[Qhualor]

ive talked about this in various threads before, but a few examples

1. mob DR

raise the DR on mobs so that you need to switch to a weapon that can bypass it to deal better damage than overpowering it with a main weapon. I sort of see this in the Amrath heroic quests when I solo them. I have to repair my Lit II after every quest because it starts to break near the end.

2. alignment damage

increase mob damage based on your alignment. it gives more meaning to what alignment you choose other than flavor.

3. mobs healing

give mobs better self healing or have an enemy healer in every pack of mobs. this would make players strategize better by trying to take out casters first.

4. mob awareness

ranging a mob from afar should alert neighboring baddies.

5. change mob AI

a. so that they attack what is considered to be the biggest threat in the group. that could be the barbarian that rushes through the door cleaving at mobs or the cleric/fvs/druid/bard/whatever in the back healing players or the caster in the back throwing down disintegrates. this may or may not be harder to code, but I think it was done like this in the past.

b. deception is too good. they need to make better saves against it. with good investment, ive seen mobs have their back turned 3/4 of the time before eventually dying.

c. at X hit points they know to heal or to run away. sometimes they get beaten down so fast they are almost dead before they heal or run away and than the ranged player gets the kill

6. design quests that utilize the lores that were given to classes. it could be like a choose your own adventure kind of thing where you could either fight or go down a different path in a quest. depending on what is chosen, xp could be built around it so the harder paths grant more xp and the easier path grants less xp.

7. expand mob tactics and spells

mobs tend to spam the same spells and tactics over and over. they need more variety and be more unpredictable.

these are just some ideas off the top of my head, but I think they would be a good start to replace increasing mob damage. increasing mob damage only leads to more power creep and players learning how to cheese content.

[Flavilandile]

They have the same issue every NWN module/world designer ended up having... the D&D rules tends to unravel beyond LVL 16-18ish... ( except for the D&D boxed set that had a clear set going upto 36 ).
This also involves the mob creation rules... To get a mob/encounter that will last more than 1/10th of a second against a party of LVL 30 characters, kitted do death, a CR 30 Mob/encounter will not cut it.... You'll have to use a CR 60+ mob/encounter.

Now to create a CR 60+ mob/encounter you can play with various variables, but the single variable that has the biggest impact is the Hit Dice ( +1 Hit Dice = +1CR ) while all the other way to raise the CR means liducrous numbers
( +1CR for +10 on all the stats, +2CR for a full Epic Item kit, +1 CR for each +20 to each save, +1 CR for each +100 Base HP ... { numbers are not DDO numbers for memories from NWN numbers... since DDO Mob is D&D 3.5 Based, the same rules should apply more or less } )

In the end, the easiest way to raise the CR of mob/encounter is to give heaps of HD to mob/encounter, and it also solve the issue of the mob/encounter lasting less than a tenth of a second... as lots of HP means you'll have to beat it down.
That's why we end up with punching balls. No more no less.

Right now, Challenging Mob is not an option anymore... When the Cap was 20, it was still ( more or less ) possible... but now, between the Destinies and the item power creep, it's definitely not an option... unless we just throw away the whole D&D ruleset and start again with another system that won't unravel once you reach high level ( GURPS, Champions, BRPS all tend to stay put at high level, Rolemaster has a tendency to unravel a bit over lvl 50ish, but it is still also a viable option because of all the optional rules available )

Anyway, whatever the system used, the fact that MMOs are in fact Monty Haul games, will put a heavy bias towards systems breaking before they would normally break.

[BigErkyKid]

They have the same issue every NWN module/world designer ended up having... the D&D rules tends to unravel beyond LVL 16-18ish... ( except for the D&D boxed set that had a clear set going upto 36 ).
This also involves the mob creation rules... To get a mob/encounter that will last more than 1/10th of a second against a party of LVL 30 characters, kitted do death, a CR 30 Mob/encounter will not cut it.... You'll have to use a CR 60+ mob/encounter.

Now to create a CR 60+ mob/encounter you can play with various variables, but the single variable that has the biggest impact is the Hit Dice ( +1 Hit Dice = +1CR ) while all the other way to raise the CR means liducrous numbers
( +1CR for +10 on all the stats, +2CR for a full Epic Item kit, +1 CR for each +20 to each save, +1 CR for each +100 Base HP ... { numbers are not DDO numbers for memories from NWN numbers... since DDO Mob is D&D 3.5 Based, the same rules should apply more or less } )

In the end, the easiest way to raise the CR of mob/encounter is to give heaps of HD to mob/encounter, and it also solve the issue of the mob/encounter lasting less than a tenth of a second... as lots of HP means you'll have to beat it down.
That's why we end up with punching balls. No more no less.

Right now, Challenging Mob is not an option anymore... When the Cap was 20, it was still ( more or less ) possible... but now, between the Destinies and the item power creep, it's definitely not an option... unless we just throw away the whole D&D ruleset and start again with another system that won't unravel once you reach high level ( GURPS, Champions, BRPS all tend to stay put at high level, Rolemaster has a tendency to unravel a bit over lvl 50ish, but it is still also a viable option because of all the optional rules available )

Anyway, whatever the system used, the fact that MMOs are in fact Monty Haul games, will put a heavy bias towards systems breaking before they would normally break.

The problem is that right until now they heavily relied on the automatic scaling system for mob creation (Self confessed).

They need to sit down and design the critters with those variables in mind and how they challenge the different styles.

It is not moving away from DnD by any means, it is just moving away from CR autoscaling. Give the mobs DnD stats by all means, just not automatically in this insane way.

[UurlockYgmeov]

players generally want to end an encounter fast.

This generally means brute force DPS.

Why is this a problem? Because brute force encounters are as interesting and fun as beating portals down. But this is what the players (at least the vocal ones) want.

Interesting encounters require teamwork and tactics. Example of tactics? CC (firewall + wall of fog + disco ball + web); shield blocking in a door way, pulling mobs, et al.

But using tactics require varied party structures and patience. Both require time, time which the player base (myself sometimes included) are not willing to spend.

Compounding this is the content being released now is Legendary. There has been much discussion about 'end game' (cap) game play since, well, the game initially came out. The stakes are high, and the dangers need to be just as great.

So in the end, part of the problem and therefore part of the solution is player expectations. There is always more than one way to skin a cat and win an encounter.

[CrackedIce]

Archers should tumble away from melee in opposite directions and use pin and whistler to stop characters in their tracks.
Archers should choose advantageous positions from afar. Using high ground, obstacles, having high jump, hiding behind mob melee
Archers should rotate targets based on need. Take down biggest threat such as cleaving barbarian and palidons builds. Characters that are low on health. Pin casters....etc

Mob melee should not bunch up and be slaughtered by cleaves and AOE. In general, mobs should stay out of persistent AOE while mob wizard dispells them. Against characters that use cleaves, mobs with high defenses and hard hits should go toe to toe. Lightly hp, fast attack melee mobs should mob rush characters such as rogues, bards.

Melee Rogue mobs should have hide in plain sight and assassinate and go back into sneak right after to choose new target...preferably casters and ranged characters.

Bard mobs should always stay invisible in sneak and never outright attack, but be singing crazy buff sings and adding utility by keeping mobs with greater heroism, hasted, fascinate and enthral characters. Dispelling character buffs.etc

Wizards should disable characters with sleet storm, over persistent AOE with very high damage per second, blind, slow, fear, enfeeble, paralyze, slay living, negative levels, Otto's irresistible dance, DOMINATE, flesh to stone. Mirror image should be available to mobs arcanes.

Sorceror mobs should nuke and nuke with large numbers. Should always be displaced and have high dodge.

Generally, melee mobs should protect ranged mobs to give casters and Archers chance to stay dangerous.

Barbarian and palidon (blackguard) mobs should always be hardest to take down. Blackgaurds have sky high saves, lay on hands, and many cleave attacks using defender stance. Barbarian mobs should have high hp and self heal while attacking. Can use cleaves as well. Fighter mobs have high tactic dcs to stun, knockdown, break fort...etc

Clerics should hang out in back and keep mob hp topped and raise fallen comrades. Casting buffs to save against debuffers or general welfare of its comrades.

That's it for now.

[sirgog]

Monsters - yes, even trash - need to be able to credibly threaten a player that is actively dodging their attacks.

I've been running some Heroic quests lately, and I'm reminded of two things I'd tried to achieve in the past.

1) The 'Running With The Devils Vampire Challenge' - back when vampire form meant 4x incoming light damage I used to try to run this on elite in vampire form, and strafing every Searing Light.

2) Bravery elite Reaver's Fate solo (back when the Disintegrate was a sure-fire oneshot, between Update 12.1 and MOTU).

Both involved aggressive dodging. Both were doable.


Mobs either need the ability to hit a moving player, or methods of slowing us down.

[BigErkyKid]

While I agree with many of the suggestions posted here, I want to clarify that I was giving the devs a framework of analysis to craft challenge with existing tools.

Sure, I'd love mobs doing fancy things. But it is important that they understand and implement designs that challenge all kinds of archetypes even with the basic attacks. As we stand this is NOT the case.

Designing challenging mobs for all archetypes requires some fine tuning, knowledge of statistics (not just expectations/means) and some IG experience.

Obviously they also need some grand plan: what is the design goal for a class? Is it achieving it? What kind of balance of encounters can we put in without completely requiring grouping but not relying only on DPS/PRR?

I am not sure the grand plan is there.

[Ayseifn]

Being able to outrun wolves is pretty silly and being able to outrun almost every monster was bad design. Fixing that wouldn't be too hard and would curtail the kiting somewhat, don't need to have every mob outrun us but we really need some that can.

The other thing to do is give almost all mobs better ranged attacks, if there's a quest filled with drow that throw daggers at a really slow rate for not much damage while their melee damage is high and can also doublestrike that's a problem. Hill giants in in Tor were brutal with that rock throw that'd knockdown and added some danger for ranged toons trying to solo, it only really mattered in that first fight but more things like that would be very welcome.

[Vanhooger]

Last 2 week, the forum was full of suggestion made by players.

Dev, ignored everything we said and they went for big damage / hp again.

I've got no hope they will do any work to bring us challenge other then the stupid oneshot, two shot.

They didn't event jump in any of the discussion and said "read by a dev".

I won't post anymore on game balance since they don't care.

[Astoroth]

Right now, Challenging Mob is not an option anymore... When the Cap was 20, it was still ( more or less ) possible... but now, between the Destinies and the item power creep, it's definitely not an option... unless we just throw away the whole D&D ruleset and start again with another system that won't unravel once you reach high level ( GURPS, Champions, BRPS all tend to stay put at high level, Rolemaster has a tendency to unravel a bit over lvl 50ish, but it is still also a viable option because of all the optional rules available )

Anyway, whatever the system used, the fact that MMOs are in fact Monty Haul games, will put a heavy bias towards systems breaking before they would normally break.

They just need to throw in the towel on d20, and go d100 for epic levels. Saves, caster Dc, to hit all of it. All these +1 mods just overwhelm the range of a simple d20 and any effort to balance such a narrow range is going to be impossible. They just made up epic ddo and destinies anyway, no point in continuing to force d20 onto it.

[Qezuzu]

Devs need to at least start with adding defenses to mobs that's not just HP.

If it's just straight up HP, it's a dumb balancing act of "oh we need to give this class 10 more melee power or they're too far behind." Instead we can go "this class is better against this [defensive measure] than other classes, and [defensive measure] is common enough, so we're fine." People wouldn't care about DPS against Bruntsmash if every single mob in the game did not take roughly identical damage from all sources. Only exception being elemental resistances and fortification, of course, and the Epic Amrath DR.

We already know mobs can use Dodge at least, having had the occasional Hasted mob dodge a hit...

Varying mobs like this would be a good step. Add thematically appropriate defenses:

Big, armored/burly enemy: high PRR, but no evasion or MRR and low/zero elemental resistance, high fort saves => casters are better against them, but can't use insta-kill reliably

(Anti)Paladin mob: has high saves and carries a Harper Pin/similar object for good measure, as well as a deathblock item. Only has moderate PRR => melees are better against them, DC casters are bad against them

Nimble, rogueish enemy: high dodge, evasion, possibly deflect arrows, but low fort saves => casters with non-evade-able spells are better against them, reliably FoD'ed, ranged toons worst against them if they have deflect arrows

Caster enemies: low/no PRR or dodge, high MRR, possible spell absorption item (~5 charges on average, so reasonably penetrable, but it can't be FoD'ed on sight), possible Mantle of Invulnerability => melees are better against them, but must close the gap. Ranged toons best against them.

"Glass cannon" e.g. Abishai: no noteworthy defenses, but deal high damage.

Furthermore, add in some class abilities that allow different classes to be more effective different defenses. Like, a rogue would be reasonably more effective at fighting another rogue, so they can ignore some dodge. Monks can debuff PRR, and Improved Sunder can debuff it further, which helps tactician fighters. Sorcs can strip away some MRR but Wizards have higher DCs, etc. These are just examples.

Afterwards they can do more technical things like improving AI or giving mobs some specific class/Destiny abilties (or just brand new abilities).

[GlassCannon]

Ok, all these suggestions are fine and good but Turbine does not have a D-Wave Quantum Computer running AIs on each server. They have to balance AI intelligence against overall lag.

Folks don't realize that the more an AI thinks, judges and makes decisions, the more processor time is used up and the more lag it causes. That's in the event AI isn't so overwhelmed with choices that it literally engages in an infinite loop and just stands there and takes damage until it dies, thinking.

Now multiply that by the number of critters it takes to generate Dungeon Alert 4.

Now spread that lag across the server because you know at least 200 people are going to be doing something stupid at the same time (myself included, as I do stupid things very frequently).


You thought item effect stacking generating loops was bad and caused destructive game-breaking lag? AI can do better and crash servers outright if it's not conservative enough.


I do like the proposal to further spread out the differences between critters' general build stats. I'd like to also see CR balanced correctly: Not the actual critter level, but the level of the character that it challenges, assuming it has the rock-paper-scissors advantage against that character. Seeing level 64 critters run around at level cap 30 is just ridiculous.


Side note: Wolves are already fast enough, thanks. Their Trip ability is famously (and excessively) overpowered and Striding is the only thing that keeps us just barely out of their reach until we can finally land a bolt or spell on them because they are behind us when they are right at our feet and nothing hits them until we stop and risk the Ultimate Trip of Makes You Fall Down And Become Paralyzed No Matter What.


That raises another issue... Knockdown Immunity is not preventing Knockdown from Trip effects. Please fix it.

And another: A wizard can cast spells on his back. A sorcerer can cast spells while upside down. Relative direction to "down" is irrelevant when considering spells. Knockdown should not ever cause Spell Failure, but a Concentration Check instead. I trust you'll start giving monsters Concentration and forcing them to use it?