15 réponses à ce sujet

[PJ156]

Who has made high level mods and how did you get on with balancing the combat. I ask because I am having problems with my campaign with some of the final fights. Some players are finding them too tough, to the point of giving up.

My problem is I cannot test every combination of party and, by the time you get to lv 14, there are lots of combinations.

I understand about CR's but they don't work becuase of the engine and AI so there is no easy scaling to be able to balance the combat without serious playtesting.

I am embarking on a mod that will be lv 14 to 22 so I need to nail this. Otherwise I need to start agin at lv 1 ... that's easy to balance

PJ

[kamal_]

iforgotmypassword has some high level modules, beyond 14 anyway.

[Arkalezth]

Kamal too...

*SPOILERS*

PJ: Unless you changed it since the beta, one of the end fights of your module (the family in... Littlelake, is it?) featured some enemies casting Word of Faith, which is a no-save AoE spell which can disable the entire party.

I didn't find the next enemy too hard, but again, one of the spells in that fight (Vampiric Feast) could kill the entire party in the blink of an eye. IMO, that fight could stay as it is, because it's not really that hard (I actually found it easier than expected), and the spell makes sense when cast by that particular enemy.

The next fight featured another wizard who had countless casts per day of Create Greater Undead. Those summons are pretty though, and killing one just to see how another is summoned, rinse and repeat, was a bit frustrating at times. I managed to win this fight easily by controlling my wizard and disabling the enemy wizard, but that implied a very specific strategy with some very specific spells that may not be available to everyone.

I don't know what to say about general balance, and I'm not a module developer, but that's the most difficult part in your module if you ask me, and all three encounters feature overpowered spells... So I guess that may tell you something.

Modifié par Arkalezth, 12 juin 2013 - 10:56 .

[Dann-J]

In Isle of Shrines I had two approaches; I used encounters to spawn level-specific 'underlings' for which I created several blueprints (up to about level 15 in some instances), and I leveled up specific placed NPCs ('bosses') to match the average party level (potentially taking them up to level 30, depending on the PC).

The advantage of using encounters to spawn underlings is that you never know exactly what you'll get. I like to design modules around re-playablility, so I often introduce random elements to make each play-through unique.

The disadvantage of levelling up the bosses was that I couldn't change their CR. Instead, I created OnDeath scripts that gave you additional XP to make up the difference between their actual level and their blueprint CR. Examining them prior to choosing to make them hostile could still lull you into a false sense of security though.

I play-tested Isle of Shrines using various PC classes, with levels ranging from 1 to the mid 20s. At epic levels the PC easily eclipsed the underlings with their limited blueprint options, but the bosses became far more of a threat. Although you usually had a choice as to whether to fight them or not, as they mostly started off in a neutral faction.

Modifié par DannJ, 12 juin 2013 - 11:01 .

[kamal_]

Arkalezth wrote...

Kamal too...

A lot of my high level areas are encounters and I rely on the encounter triggers plus the massive set of blueprints painofdungeoneternal provided somewhere. My end boss area used npcs built as pc's might reasonably be, plus creatures from pain's blueprints.

If you really want to annoy players, golems with wizard levels. Choosing the package spell/feat options of the properties will stack the golem and wizard levels, giving them access to epic spells, tons of hp from the golem levels, and golem immunities. I had lots of comments that those guys + golem fighters were too tough.

I had a hard time coming up with reasonable enemies. Thayans (plus their thayan gnoll guards), Shadowvar, planar creatures, high level enemy adventurers, sarrukh (progenitors of the yuan ti, can use same creature models), packs of high level undead.

[Claudius33]

In 16 Cygni I have put options for hard boss fights thru a convo. If the player chooses intermediate I dynamically remove a couple of foes, if he chooses easy I remove some others.

In Sarmates! I have set up a 'prestige' concept. For hard fights the player may lower the difficulty at the price of a loss of prestige. Of course a low prestige is not a game breaker. In addition, if the player runs out of money he can get some, providing a loss of prestige. So prestige essentialy is an incentive and is implemented a dummy quest.
Except for bosses, enemies are usually 1 level below the expected PC level, with a less effective weaponry. Of course it's impossible to know if the player has done useful crafting in the mid to end game ...

The player usually ends at level 17/18, so I obviously can't test every combination, especially since the player can choose the classes/races of the companions and multi class them ... Therefore to fine tune the fights whatever the level, I play at normal level with no buff. I just allow myself healing potions and AI only casted buffs. As I use some tactics, if I get systematically decimated, I remove a couple of foes. In introductory convos I sometimes give tactical hints about the incoming fight.
So a player who is more story oriented playing at easy (nwn2) level and using buffs should make it. Nevertheless some players still find my mods hard ... 

Modifié par Claudius33, 13 juin 2013 - 07:23 .

[Claudius33]

[double post sorry]

Modifié par Claudius33, 13 juin 2013 - 07:21 .

[PJ156]

Thanks gents, there are some things to think about here. I have never been one for spawning but perhaps I will think about that. Too many of my scripted solutions fail in play as it is without bosses not turning up

I may repost here when the time comes. I have a bit to do before I start putting critters in. In the early part it is humans and that should be easier I guess.

I agree with Kamal_ though, about high level enemies. I build a low fantasy environment so devils and dragons etc I use very sparingly I dont want the later mods to be an outsider fest to keep the difficulty level up

PJ

Modifié par PJ156, 13 juin 2013 - 09:03 .

[Dann-J]

PJ156 wrote...

I agree with Kamal_ though, about high level enemies. I build a low fantasy environment so devils and dragons etc I use very sparingly I dont want the later mods to be an outsider fest to keep the difficulty level up  

'Special' creatures usually have known weaknesses and attack types, whereas humanoids can potentially be of any class or level and wield any sort of weapon or magic. That makes humanoid enemies far more dangerous in my book (and is probably why humanoids rule Faerun instead of dragons or demons).

[ColorsFade]

DannJ wrote...

....and is probably why humanoids rule Faerun instead of dragons or demons.

For the time being.... 

Goes back to modding... 

[ColorsFade]

PJ156 wrote...

My problem is I cannot test every combination of party and, by the time you get to lv 14, there are lots of combinations.

PJ

True - you cannot test every combination. But that is not the issue. The issue is encounter makeup. Encounters should be structured such that most parties can defeat them. 

Encounter creation is not about difficulty, but about the perception of difficulty. 

I test with a party that has one of every major class: fighter, wizard, rogue and divine caster. Anyone else is a bonus. The core group of four should be able to win the encounter.

I haven't nailed down my final decision on party size yet (right now I'm going with 5 or 6 - it will probably be six). There are going to be instances when I will spawn extra monsters if I check and see the party has an additional tank, or an additional wizard, etc. Depending on party makeup, I can spawn monsters that fit the scenario better. 

I think that's part of the trick - you can do some really neat stuff with scripting and it has the side benefit of making the encounter semi-unique to each party, so each time the person plays the mod, if they adjust their party, they'll get something slightly different. 

Example: A fight might spawn one big melee fighter (meant to be tanked by the party's one tank), and a couple secondary shamen plus a handful of archers. If you see the PC party has two tanks, then maybe you spawn two big bad melee fighters and fewer archers with the idea the PC party has to tank each one down. Or maybe you see the party makeup is rather squishy with multiple wizard types, so you spawn more archers instead, with the idea that the wizards will cast Protection from Arrows and then nuke away at the ranged adversaries. 

It makes sense to look at the party makeup and then adjust the encounter - when possible. 

And I'm talking mostly about big boss encounters. For the trivial stuff on the way... it should be easy, but present the perception of a challenge. 

Modifié par ColorsFade, 14 juin 2013 - 05:34 .

[Lugaid of the Red Stripes]

The way I see it, what makes high-level parties so difficult to balance for is that they represent the accumulation of many choices made by the player, not just character sheet builds, but all the extra xp and loot picked up by savvy players. Players who have made good choices thus have overpowered parties, and players who've made poor choices have weak parties.

The solution for me, then, seems not to be constantly adjusting the difficultly of the encounters, but making it so that there's less difference between a good choice and a poor choice. The easiest way to do that is just to make the module unpredictable, filled with wild cards that make it difficult to min-max a party to breeze through it. Disable magic, take away the party's equipment, overwhelm the tank with numbers, force the party to be stealthy, force the party to be diplomatic, just do anything out of the ordinary, and do it often enough that the savvy players are forced to hedge their bets and keep their parties balanced, and even the odd munchkin builds find themselves occasionally useful.

[bealzebub]

When I'm balancing combat, regardless of level, I usually start with "Neverwinter Nights 2 Persistent World Server Balance Guidelines"
http://nwvault.ign.c...s.Detail&id=103
and proceed from there. While not perfect, it gives me a good place to start.

[PJ156]

That's a useful document I will use that.

To me they seem a bit weak, maybe there in lies the problem

Thanks for all the other thoughts, I have much to think about for the new campaign.

PJ