BGTSCC's Server Difficulty

[DaloLorn]

Fairly spot-on across the board, Calen, and I can find only one point of disagreement with you. Specifically, your list of stereotypical mobs was missing an entry:

The blaster mage: Has a spell selection guaranteed to inflict medium to high damage against any character that lacks the DPR to instantly kill it. (Which is a lot of characters, at the level they're intended to be fought, because they somehow retain most of the AC, HP, and saves of their nonmagical peers, and promptly supplement it with buffs.) Occasionally spawning in devastating groups of 2 or 3 at a time, and is only effective (despite having a fraction of the spell slots it should have) because it's not subject to attrition or sniping like a PC mage would be. In PnP, in singleplayer campaigns (like the NWN2 OC), or on PWs that try to more accurately emulate PnP, these would still exist, but they would be used more sparingly, more deliberately than they currently are.

[gedweyignasia]

Typical mob behavior

The mobs on this server are designed to be as annoying as possible with out much of variation, once you figured out the maps you know which areas are not worth leveling in. You'll often find yourself grinding location A while over leveling it ,just to avoid an unrewarding leveling experience in location B as the exp is worse and the loot is overall terrible either way. Personally I farm wyverns up to 21 and just end up getting the rest of my levels through roleplay/quests and occasionally joining loot runs, though solo play is off limits unless I powerbuild/Mule and approach the content with a meta/ooc mindset.

a few stereotypical BGTSCC mobs

- The dispeller [...]

- Death spell spammers [...]

- 'Da boss' [...]

- I hope you like DC's mob [...]

- bigby [...]

- Random spawn with no listen/spot check [...]

There is no cleverness in the design of the mobs or variety in having different type of casters with different spell books.

This is some very fair criticism, and a lot of it comes down to our spawner system. It's hard to do this right with random generation. We could set up a few bespoke encounters in specific locations, but players will run through those far faster than we can create them. (And once they're known, players will approach these encounters with the right buffs and strategies, meaning they'll either become boring and easy or they'll be a frustrating obstacle players want to avoid.)

This is a difficult design obstacle and it's not easy to do any better than what we have at present, but a few of us are experimenting with new approaches in some new areas. Most of these will probably work worse than what we've got, but maybe something will turn out better.

[DaloLorn]

Typical mob behavior

The mobs on this server are designed to be as annoying as possible with out much of variation, once you figured out the maps you know which areas are not worth leveling in. You'll often find yourself grinding location A while over leveling it ,just to avoid an unrewarding leveling experience in location B as the exp is worse and the loot is overall terrible either way. Personally I farm wyverns up to 21 and just end up getting the rest of my levels through roleplay/quests and occasionally joining loot runs, though solo play is off limits unless I powerbuild/Mule and approach the content with a meta/ooc mindset.

a few stereotypical BGTSCC mobs

- The dispeller [...]

- Death spell spammers [...]

- 'Da boss' [...]

- I hope you like DC's mob [...]

- bigby [...]

- Random spawn with no listen/spot check [...]

There is no cleverness in the design of the mobs or variety in having different type of casters with different spell books.

This is some very fair criticism, and a lot of it comes down to our spawner system. It's hard to do this right with random generation. We could set up a few bespoke encounters in specific locations, but players will run through those far faster than we can create them. (And once they're known, players will approach these encounters with the right buffs and strategies, meaning they'll either become boring and easy or they'll be a frustrating obstacle players want to avoid.)

This is a difficult design obstacle and it's not easy to do any better than what we have at present, but a few of us are experimenting with new approaches in some new areas. Most of these will probably work worse than what we've got, but maybe something will turn out better.

Oh, bloody Cloudflare ate my post.

I was saying that ditching the MMO spawner and more accurately mapping mob CRs to their threat level would be steps in the right direction. Right now, spellcasting mobs are typically treated as being effectively one level ahead of their peers, while remaining a threat long after said peers become harmless, and the MMO spawner guarantees a limitless supply of them.

[gedweyignasia]

I was saying that ditching the MMO spawner and more accurately mapping mob CRs to their threat level would be steps in the right direction. Right now, spellcasting mobs are typically treated as being effectively one level ahead of their peers, while remaining a threat long after said peers become harmless, and the MMO spawner guarantees a limitless supply of them.

Yes, it's very clear that spellcasters and enemies with high AB are the most frustrating for players. Unfortunately, I'm not sure how to make combat challenging without making it frustrating. The parts that aren't easy for you to overcome fill both roles.

"Ditch the MMO spawner" is straightforward if you just want to remove enemies from those areas. "Ditch the MMO spawner and replace it with something better" is a tall order, however.

[DaloLorn]

I was saying that ditching the MMO spawner and more accurately mapping mob CRs to their threat level would be steps in the right direction. Right now, spellcasting mobs are typically treated as being effectively one level ahead of their peers, while remaining a threat long after said peers become harmless, and the MMO spawner guarantees a limitless supply of them.

Yes, it's very clear that spellcasters and enemies with high AB are the most frustrating for players. Unfortunately, I'm not sure how to make combat challenging without making it frustrating. The parts that aren't easy for you to overcome fill both roles.

"Ditch the MMO spawner" is straightforward if you just want to remove enemies from those areas. "Ditch the MMO spawner and replace it with something better" is a tall order, however.

IMO, challenging is okay if it's a challenge you're supposed to be able to overcome, or if the rewards are appropriate to the challenge.

You're not designing caster mobs to be killed/stunned in one round, and if you are, it's in response to a slowly growing percentage of people who are capable of achieving one of those two things due to assorted forms of power creep pushing high-end performance far above what the server is designed for with only a moderate effect on the average PC. Due to their near-guaranteed damage output, caster mobs are a challenge that can only be overcome through excessive amounts of HP/saves/DPR well above what you should have at the level the casters were designed for. They are also a challenge you usually can't opt out of, because half of the dungeons on the server have a caster mob in the mix.

As for the rewards, the spellcasters and high-AB enemies are generally not gatekeeping any loot nodes (the beholder in the gnoll cave is a noteworthy exception), nor do they innately generate significantly larger XP/loot compared to their peers (the warlocks in the Sharptooth cave are a minor exception to this one). You can get 40 XP for a nearly harmless gnoll marauder, and still barely hit 45 XP for a gnoll occultist that ate through half of your HP pool before it either died or ran out of offensive spells.

[Hoihe]

One stop-gap solution I can see for the caster issue is reducing time to kill.

Player spellcasters, unless fully buffed, go down like wet paper against level appropriate enemies.

NPC spellcasters on the other hand combine the endurance of a low-Cl gish (300+ HP, high enough AB to hit in melee for 30 damage a hit, DR) with the lethality of a pure caster.

If they had one of the following:

A) Less HP & DR
B) Had Imp invis/Displacement/Jaunt removed. Potentially elemental shield too.
C) Had kill/lose/autohit spells removed in favour of save or damage or touch AC spells

it'd already improve player experience.

Fighting frost/fire giant mages wouldn't be as big of an issue if the time to kill was not 3-4 rounds on a 42 AB 6 attacks per round, 20-38 damage per hit before DR class.

Like, 300 HP + Imp invis means you need to land 20 attacks on average to kill them. That's around 4 rounds where they'll hit you with IGMS/save or die spells.

[gedweyignasia]

You can get 40 XP for a nearly harmless gnoll marauder, and still barely hit 45 XP for a gnoll occultist that ate through half of your HP pool before it either died or ran out of offensive spells.

I believe that's the purpose of those harmless creatures; they're cannon fodder so that everyone gets to hit things, and they're free XP for you. Casters are an easy fix for balancing because that means that you're guaranteed a little attrition across a variety of builds and a small level range.

Player spellcasters, unless fully buffed, go down like wet paper against level appropriate enemies.

One-to-one, that shouldn't be the case. It's getting focused by multiple enemies that crushes players quickly in most cases.

[YYA]

Speaking of mob AC:

You can create mobs with high AC, but high AC means that a considerable portion of builds will only land a hit by rolling a natural 20. This was tried on the server before, and only Frenzied Weapon Masters and fully buffed up Favoured Soul/Blackguards could actually hit things with their first flurry of attacks. And moreover, the way you apply the high AC also matters. If the mob AC is dexterity based; sneaks will just pop out of stealth and get those one shot kills with their first flurry sneak attacks. But it will also manifest quite a number of either Shadowdancing Frenzied Weapon Masters or PTWF Fighter 12/Champion of Corellon/Weapon Master 5/Shadowdancer 3 builds Masters, who will simply collect a mob of monsters while running around, then hit HiPS, Whirlwind Attack against the flatfooted foes, and let their Cleave attacks cleave through everything around them. If you instead make the mob AC based on natural armor, all of a sudden all those characters who rely on sneak attack dice just cannot land a single hit, not when they pop out of Stealth, and not when they try to make use of Feint. Thus high AC mobs are not really a solution.

And if you increase damage per hit, well all of a sudden one string of lucky crits could even Fugue a constitution based Barbarian 20 with Epic Rage, unless the barbarian took something like 10 Epic Toughness feats on top of all that constitution. If you instead raise the mob AB, well, you can easily render the AC most player characters can reach into an utterly irrelevant number: these character will be hit each round unless they have concealment from one source or another.

Thus when it comes to balancing monsters for all players of the server, while pursuing the nebulous goal of challenging content, you quite literally cannot do it without screwing over one character archetype or another. Thus, during those years now long past, high HP mobs were seen as the least of the evils, in order to meet the demands for server challenge by some players. Blaster casters and the mechanically unsound role-play builds got the sorry end of that stick.

Which is why I have suggested for the addition of these 'easy mode' areas. No chests to loot, and even if these monsters give less experience than their CR counter parts elsewhere, it would solve the server difficulty problems with almost a snap of fingers. You want to play mechanically unsound character? Well, you can, just head to these areas. You will probably find other players there too, and now you got a party to challenge some other areas you would never enter on your own.

Otherwise, the server just moves on as it always has, where the best advice I can offer to someone with a mechanically unsound build is to just do the weekly quests... which no longer even reset weekly.

Nothing can be done regarding the server's difficulty until the Admins finally implement some guidelines regarding the server mechanics. For example, 4-5 years ago the AC melee character were going for was 50, 50+. Nowadays, with all the PRCs that give AC and immunities left and right, with no restrictions, that number is 60, 60+.

The increase in AC does not mean that mobs have to be re-balanced -- because if the AC of 50+ was good enough then, going for AC of 60+ actually offers no additional advantage in PvE. A mob with 30 AB is just as likely to land a hit on a character with 50+ AC as it is to a character with 60+ AC. The concern here is largely related to PvP, which you can entirely avoid by taking RP out on the first sign of danger.

[gedweyignasia]

Speaking of mob AC:

You can create mobs with high AC, but high AC means that a considerable portion of builds will only land a hit by rolling a natural 20. This was tried on the server before, and only Frenzied Weapon Masters and fully buffed up Favoured Soul/Blackguards could actually hit things with their first flurry of attacks. And moreover, the way you apply the high AC also matters. If the mob AC is dexterity based; sneaks will just pop out of stealth and get those one shot kills with their first flurry sneak attacks. But it will also manifest quite a number of either Shadowdancing Frenzied Weapon Masters or PTWF Fighter 12/Champion of Corellon/Weapon Master 5/Shadowdancer 3 builds Masters, who will simply collect a mob of monsters while running around, then hit HiPS, Whirlwind Attack against the flatfooted foes, and let their Cleave attacks cleave through everything around them. If you instead make the mob AC based on natural armor, all of a sudden all those characters who rely on sneak attack dice just cannot land a single hit, not when they pop out of Stealth, and not when they try to make use of Feint. Thus high AC mobs are not really a solution.

And if you increase damage per hit, well all of a sudden one string of lucky crits could even Fugue a constitution based Barbarian 20 with Epic Rage, unless the barbarian took something like 10 Epic Toughness feats on top of all that constitution. If you instead raise the mob AB, well, you can easily render the AC most player characters can reach into an utterly irrelevant number: these character will be hit each round unless they have concealment from one source or another.

Thus when it comes to balancing monsters for all players of the server, while pursuing the nebulous goal of challenging content, you quite literally cannot do it without screwing over one character archetype or another. Thus, during those years now long past, high HP mobs were seen as the least of the evils, in order to meet the demands for server challenge by some players. Blaster casters and the mechanically unsound role-play builds got the sorry end of that stick.

I agree, and I think this is why we gravitated towards high-HP, low-damage builds. People don't like being hurt by the dice, and high-HP, low-damage means less statistical noise. I'd argue it's pushed things in a very grindy direction, though.

Which is why I have suggested for the addition of these 'easy mode' areas. No chests to loot, and even if these monsters give less experience than their CR counter parts elsewhere, it would solve the server difficulty problems with almost a snap of fingers. You want to play mechanically unsound character? Well, you can, just head to these areas. You will probably find other players there too, and now you got a party to challenge some other areas you would never enter on your own.

Otherwise, the server just moves on as it always has, where the best advice I can offer to someone with a mechanically unsound build is to just do the weekly quests... which no longer even reset weekly.

This is part of why RP XP is huge compared to what it used to be. Even went out of the way to create Ulcaster Ruins as an RP dungeon.

[Snarfy]

This thread will just continue to go in circles unless we spit-ball more ideas on how to make things better, or expand upon YYA's suggestion, rather than continue to rattle off a litany of gripes(regardless if they're legitimate ones).

NPC spellcasters on the other hand combine the endurance of a low-Cl gish (300+ HP, high enough AB to hit in melee for 30 damage a hit, DR) with the lethality of a pure caster.

If they had one of the following:

A) Less HP & DR
B) Had Imp invis/Displacement/Jaunt removed. Potentially elemental shield too.
C) Had kill/lose/autohit spells removed in favour of save or damage or touch AC spells

it'd already improve player experience.

Of those three, I would lean towards A and C... the problem with the latter being that if the character has high SR or immunities, the NPC's might not cast anything at all.

There is no cleverness in the design of the mobs or variety in having different type of casters with different spell books.

This seems to be the consensus, so what ideas can we come up with to make things less predictable? (keeping in mind the biases/limitations of the AI)

I posted this earlier:

... rather than randomized spell books, there would probably need to be a few alternate preset types of variants created. For example, in the case of the goblin crusaders north of Beregost, they could be split off into goblin invokers, enchanters, shamans, novices, and so on.

This would lessen the frequency of encountering these types:

- The dispeller aka moneysink.

- Death spell spammers aka steadfast or die

I don't know how time consuming it might be to create, let's say... even three variations of caster types? Heck, apply them globally, scale to appropriate CR, and see how it goes.

- 'Da boss'

The bosses could even be fiddled with. For example, why don't the frost giants ever elect a barbarian/FB/whatever as their king? Even the addition of two other "types" of FGK would throw a curve ball at our metaknowledge of what to expect.

- DC's mob -

... I don't have much on this one, other than: tone it down.

- Random spawn'' No listen check, spot check or anything you just happened to get besieged by 3 mobs.

... and the MMO spawner guarantees a limitless supply of them.

"Ditch the MMO spawner" is straightforward if you just want to remove enemies from those areas...

You don't even need to remove it, once again: just tone it down. Having monsters constantly coming out of the walls and thin air isn't really necessary(or realistic, hello immersion!), maybe just slow down the spawn rate in some zones, and in other zones you could even have spawns simply not repopulate until players leave the area(say, for instance, sharptooth orc caves, or other smaller interiors).

Yes, it's very clear that spellcasters and enemies with high AB are the most frustrating for players. Unfortunately, I'm not sure how to make combat challenging without making it frustrating.

I think the challenge rating is fine in some zones, but a little over the top generally speaking. Maybe you don't need to find more ways to make it challenging, so much as it would be far more interesting/realistic by making dynamic tweaks.

Once again, heavily optimized builds are going to mulch through content regardless, so scaling the difficulty back, or even making things a little more dynamic, is going to have little to no impact on those builds whatsoever. If mobs are made easier to kill, or the XP/loot rewards aren't present, uber builds probably wont even bother visiting them, but at least sub-optimal/RP builds will have an easier go at it, and feel/seem less impotent while they're at it.

[Hoihe]

"Ditch the MMO spawner" is straightforward if you just want to remove enemies from those areas...

You don't even need to remove it, once again: just tone it down. Having monsters constantly coming out of the walls and thin air isn't really necessary(or realistic, hello immersion!), maybe just slow down the spawn rate in some zones, and in other zones you could even have spawns simply not repopulate until players leave the area(say, for instance, sharptooth orc caves, or other smaller interiors).

I want to echo the immersion/unrealistic nature of some spawners.

Even if you drag a spotter druid along, with True Sight and Track mode up...

Having enemies spawn literally on top of you despite strategizing to split enemies up beforehand kind of takes you out of things.

And some areas could really use toning down rates.


I ran into a group once outside frost giant fort.

Our interaction, summarized:

I indicated I am no threat. They indicated they weren't either.
I asked if they were going same way, they replied opposite.
We elected to split.

That's a very short interaction that makes sense if you run into strangers in a dangerous place, no?

In the span of the few minutes this all took, while we were standing in place... We've had frost giants respawn over, and over, and over again at least 3 times.


And please don't say "move to a safer area", because we didn't even know if the other was hostile or friendly, which was what we tried to identify in like 3-4 emotes back and forth.

[gedweyignasia]

You don't even need to remove it, once again: just tone it down. Having monsters constantly coming out of the walls and thin air isn't really necessary(or realistic, hello immersion!), maybe just slow down the spawn rate in some zones, and in other zones you could even have spawns simply not repopulate until players leave the area(say, for instance, sharptooth orc caves, or other smaller interiors).

I want to echo the immersion/unrealistic nature of some spawners.

Having enemies spawn literally on top of you despite strategizing to split enemies up beforehand kind of takes you out of things.

And some areas could really use toning down rates.

I think the issue was that having a ton of enemies in an area is draining on server resources, but since we've had Dae going around and fixing all the crappy code that was bogging down our server, this might be more feasible. It does entail writing a new spawner, however, so don't expect to see it happen overnight.

As a side note:
I'm very receptive to criticism and I don't get tired of explaining things to players, but that's not true for a lot of people who've worked really hard to make the dungeons and balance the spawns people are frustrated with. How our spawners work, why our spawns are balanced this way instead of that way, etc. is something a lot of players have ideas about, but they're not always right (even when they're close to being right). It's a little discouraging to hear people who don't understand your system complaining about why it's wrong and how something else would be better, especially when they might be suggesting things that have already been tried and haven't worked, or spawned their own threads very much like this one. Reading full a whole thread full of posts like that can be pretty taxing for someone who spends a lot of their free time trying to contribute to the server, so please, please, please consider your tone/the way your words might affect someone else, and try to be a little humble no matter how experienced you are. It will make devs more inclined to read threads like this beyond the first two posts.

[Steve]

...a lot of players have ideas about, but they're not always right (even when they're close to being right). It's a little discouraging to hear people who don't understand your system complaining about why it's wrong and how something else would be better, especially when they might be suggesting things that have already been tried and haven't worked....

Please do explain it then!

[gedweyignasia]

Please do explain it then!

I'm not sure how transparent the team wants to be about the spawners (lest people game the system more than they already do), and there are at least three different systems, which are sometimes combined in one area. Just understand that it's possible you've rolled a Listen or Spot check without knowing it, enemies might have been spawned in a position that was supposed to be out of sight of the person triggering the spawn but wasn't out of your sight, etc. You don't know how much you don't know.

[DaloLorn]

Please do explain it then!

I'm not sure how transparent the team wants to be about the spawners (lest people game the system more than they already do), and there are at least three different systems, which are sometimes combined in one area. Just understand that it's possible you've rolled a Listen or Spot check without knowing it, enemies might have been spawned in a position that was supposed to be out of sight of the person triggering the spawn but wasn't out of your sight, etc. You don't know how much you don't know.

Well, I have a general idea of how one of the spawners (the one used to spawn bosses in the Nashkel mines, among other things) works, gleaned from a combination of player experience and DM-side inspection. It's probably the least sophisticated and most predictable of the spawners, and I'd be surprised if there weren't a horde of players who hadn't intuited its mechanics already without the aid of the DM client. On this front, I would therefore challenge the assertion that we don't know how much we don't know.

The MMO spawner used on most of the server seems fairly predictable too, but even without access to the code, one can infer from its behavior that there's more going on than meets the eye. (Especially the process by which the spawns are selected, since it varies the number and combination of mobs every time it runs. It also seems to support one-per-map boss groups like the ones in the gnoll caves, but per the goblin mines north of Beregost, it is not required to always spawn them in the same location.) Here, I would agree that we don't have adequate information on how it works... but the unknowns are materially irrelevant to the issue at hand, which is that the MMO spawner should, at the very least, be considerably refined, if not replaced outright.

The third spawner that I'm aware of would be the "ambush" spawner that very rarely triggers on the Trade Way maps south of BG (plus the FAI map), which triggers so infrequently that it's impossible to glean any meaningful data on its behavior, except that it's apparently been updated to spawn level-appropriate mobs. I like it. That's all I know about it.