Jump to content

Recommended Posts

Posted

One thing a few people have mentioned is increasing enemy sight range. While that's really a separate issue, I'd agree 100%. Enemies should have exactly the same sight range as the player, so if you reveal enemies while not sneaking, they instantly attack (as it was in the IE games). This MMO-esque/Diablo-esque "mob pulling" silliness does not belong in PoE.

 

Not that that's bad, mind you - it just doesn't belong in an IE successor. We already have the tools to never engage unintentionally via the stealth system - and this wouldn't change anything about that. The eyeball would just appear and the yellow pie would just start to fill as soon as you see enemies, instead of there being this magical fringe zone where you can see them and they can't see you.

 

Are you actually implying that people didn't mob pulled in old IE games? Dude they did even worse things, like exiting a building to allow buff to wear off and a lot abuse that ain't allowed in this game. I want you to know that i agree with the OP, but this you said is ridiculous.

  • Like 1
Posted

Enemy AI anti-300 door logic:

if (npc exists between doorFrameMarkerLeft and doorFrameMarkerRight) then
   if (npc armor rating > 10)
       enemynpc(switch to ranged attacks)
       enemynpcdialogueshout("don't get lured into their choke point, switch to ranged")
   endif
endif
  • Like 2

Having trouble with the games combat on POTD, Trial of Iron?

- Hurtin bomb droppin MONK - [MONK BUILD] - [CLICK HERE]

- Think Rangers suck? You're wrong - [RANGER BUILD] + Tactics/Strategies - [CLICK HERE]

- Fighter Heavy Tank - [FIGHTER BUILD] + Tactics/Strategies - [CLICK HERE]

Despite what I may post, I'm a huge fan of Pillars of Eternity, it's one of my favorite RPG's.

Anita Sarkeesian keeps Bioware's balls in a jar on her shelf.

Posted

 

Enemy AI anti-300 door logic:

if (npc exists between doorFrameMarkerLeft and doorFrameMarkerRight) then
   if (npc armor rating > 10)
       enemynpc(switch to ranged attacks)
       enemynpcdialogueshout("don't get lured into their choke point, switch to ranged")
   endif
endif

 

Stick Level 1 wizard in heavy armour in chokepoint, never worry about enemies' melee attacks and abilities again.

  • Like 5
Posted (edited)
The solution is simple: Program the AI not to fight in disadvantageous chokepoints if it can help it, and (where applicable) to attempt to use the same strategy.

 

If you are fighting the enemy in a chokepoint, then the enemy is also fighting you in a chokepoint.  In other words, if you choose to use this strategy then the AI is also using it, and if the AI were programmed to use this strategy then you would also be using it.

 

I do agree that a more advanced AI would be a wonderful improvement to these sorts of games, but there's always going to be a limitation to it that the human mind can exploit for advantage, at least with the computer hardware we are using to run games currently.  If you program the AI to fall back from chokepoints then someone is going to come up with a strategy that pelts the AI from range while toggling the arrangement of the front line to make the AI think that the path to the ranged attackers is alternatively blocked and open (resulting in the AI rushing forward and falling back in cycles, but never actually engaging or attacking) - or something else similar to that.  The AI's strategy is locked in, but the player is free to adapt to it and the player will do exactly that no matter how the AI is programmed.

Edited by Emptiness
Posted

I don't see the problem.  I feel like people just want to make problems with this game where there aren't any.

 

You're back again to fight against people that want a more challenging experience in some form? Is this your thing? Can't understand that some players want a more challenging experience without affecting the experience of lower difficulties? 

Posted

I would say that the only realistic solutions are:

 

  1. Design future maps to specifically minimize chokepoints and require the player to engage enemies in much more open spaces.
  2. Create hard AI leashes that drag enemies back to certain spawn locations if taken too close to doorways.
  3. Create more Shadow/caster-type creatures that can make fighting in chokepoints disadvantageous, either because they can make the chokepoint a deathtrap themselves, or because they can bypass the tank line entirely.
  • Like 1
Posted

 

People wouldn't do that if not for the dps/tank polarization + engagement system. 

What? Did you not exploit chokepoints in the original IE games? I sure as hell did.

 

 

Not as much as this game, but that's because the original IE games had so many god-mode solo builds.

Posted

Are you actually implying that people didn't mob pulled in old IE games? Dude they did even worse things, like exiting a building to allow buff to wear off and a lot abuse that ain't allowed in this game. I want you to know that i agree with the OP, but this you said is ridiculous.

 

Yeah, there is one particular Lich that is super easy to beat if you just go back and forth between his crypt and the area outside it, timing each exit to happen when he starts casting a spell.  His protections wear off and he runs out of spells and then he's about as dangerous as a peasant without a weapon.

Posted (edited)

 

 

Enemy AI anti-300 door logic:

if (npc exists between doorFrameMarkerLeft and doorFrameMarkerRight) then
   if (npc armor rating > 10)
       enemynpc(switch to ranged attacks)
       enemynpcdialogueshout("don't get lured into their choke point, switch to ranged")
   endif
endif

 

Stick Level 1 wizard in heavy armour in chokepoint, never worry about enemies' melee attacks and abilities again.

 

 

I don't think a wizard, even in heavy armor would be able to tank a barrage of ranged attacks, without that recover ability that fighters have.

Edited by luzarius

Having trouble with the games combat on POTD, Trial of Iron?

- Hurtin bomb droppin MONK - [MONK BUILD] - [CLICK HERE]

- Think Rangers suck? You're wrong - [RANGER BUILD] + Tactics/Strategies - [CLICK HERE]

- Fighter Heavy Tank - [FIGHTER BUILD] + Tactics/Strategies - [CLICK HERE]

Despite what I may post, I'm a huge fan of Pillars of Eternity, it's one of my favorite RPG's.

Anita Sarkeesian keeps Bioware's balls in a jar on her shelf.

Posted

Main issue I have with the chokepoint is that the AI is not reactive/adaptive in changing targetting preferences; for example I was in the Fampyr level of Od Nua and without fail every single fight they would attempt to charm my Cipher first; and would only swap to another target to Charm if I had her stand waaaaaaaaaaaay back. They would then swap to trying to Charm my Priest. Always.

So I could reliably trivialize the Fampyr encounters by abusing the omnicient stupidity of the Fampyr by putting my Cipher just around the corner out of LOS but close enough to stay on their near hardcoded aggro list while they ran into my tank wall trying to gain LOS on the Cipher and subsequently spending the first portion of the fight getting knocked prone and roflstomped. They never even "saw" my Cipher until they on the floor.

Anything 10+ on intellect shouldn't be quite that predictable. The AI scripting could use a touch more fuzz factor and less omniscience.

  • Like 1
Posted

 

Enemy AI anti-300 door logic:

if (npc exists between doorFrameMarkerLeft and doorFrameMarkerRight) then
   if (npc armor rating > 10)
       enemynpc(switch to ranged attacks)
       enemynpcdialogueshout("don't get lured into their choke point, switch to ranged")
   endif
endif

Off the top of my head, there are at least two obvious exploits. The first is the trick I mentioned above: the fighter stands in a position that's not quite blocking the choke point, but from which he can easily do so when the enemies come close enough. This is actually quite generic. The second is that one character in heavy armor blocks the door while the rest of the party engages the enemy (who by your logic are now obliged to use ranged weapons) in melee. :) You can of course try to refine the algorithm (at the very least check whether the rest are on the near or far side of the fighter blocking the door), but there will always be exploits. As I said, it's a difficult computational problem.

Posted

I'd like to see smarter AI responsiveness in combat as well. A tactic where foes hang back -- or fall back -- forcing you to come to them would make for very interesting tactical play. As an example, Total War AI does this to a certain extent to take advantage of high ground in Shogun 2.

Posted

I'd like to see smarter AI responsiveness in combat as well. A tactic where foes hang back -- or fall back -- forcing you to come to them would make for very interesting tactical play. As an example, Total War AI does this to a certain extent to take advantage of high ground in Shogun 2.

 

I fear this would mostly lead to what can best be called the "fidget strategy": attack them in melee until they back off, then switch to guns and move forwards just enough to shoot them until they decide to come back into melee, at which point you retreat back into the chokepoint.

Posted

Well, that's why Matt is right about the fog of war advantage as well. It should be evens. It would be even more interesting to have some foes *starting* from stealth if that were possible so that the party isn't necessarily able to see them to begin with. That could invert the first-mover advantage that the player normally has in a quite challenging way.

  • Like 1
Posted (edited)

I don't see the problem.  I feel like people just want to make problems with this game where there aren't any.

 

This is a silly reply. I'm not QQ-ing, I'm providing constructive criticism. No game is perfect, so to claim that there aren't any problems with the game is silly. I'm considerably on the "it's fine" or "they're not gonna change it so stop asking" end of the spectrum for most things, but I think it's absolutely worth discussing some of the more important issues. When there is one best strategy for most encounters, and its been discovered about a week after the game comes out, that's a serious problem for anyone looking to create a challenging experience. Unpredictability and requiring the player to adapt is how you keep your game from being solved and losing all replay-ability. Avoiding the creation of uber-dominant strategies is just good game design, and it's better for everyone. 

 

 

 

One thing a few people have mentioned is increasing enemy sight range. While that's really a separate issue, I'd agree 100%. Enemies should have exactly the same sight range as the player, so if you reveal enemies while not sneaking, they instantly attack (as it was in the IE games). This MMO-esque/Diablo-esque "mob pulling" silliness does not belong in PoE.

 

Not that that's bad, mind you - it just doesn't belong in an IE successor. We already have the tools to never engage unintentionally via the stealth system - and this wouldn't change anything about that. The eyeball would just appear and the yellow pie would just start to fill as soon as you see enemies, instead of there being this magical fringe zone where you can see them and they can't see you.

 

Are you actually implying that people didn't mob pulled in old IE games? Dude they did even worse things, like exiting a building to allow buff to wear off and a lot abuse that ain't allowed in this game. I want you to know that i agree with the OP, but this you said is ridiculous.

 

 

My bad - when I used the phrase "mob pulling", I was referring specifically to the MMO-esque way in which we can see enemies standing in visual range, but they don't attack unless we enter "aggro range". Bad terminology on my part. People certainly did do something that could also be referred to as "mob pulling" in the IE games (as well as all the other cheese you mentioned).

 

Cheese should stay on pizza and sandwiches - it makes for boring gameplay. I don't fault people for using it when a game is challenging, but a good game designer is going to be looking to eliminate as many exploitable strategies as possible. That was a fairly big priority in PoE if I understood Josh Sawyer's design goals correctly, so hopefully once it becomes apparent that this strategy is extremely dominant they'll do something about that. I wonder if Josh still reads these forums - probably not often. I don't know if I would. :p

 

 

Main issue I have with the chokepoint is that the AI is not reactive/adaptive in changing targetting preferences; for example I was in the Fampyr level of Od Nua and without fail every single fight they would attempt to charm my Cipher first; and would only swap to another target to Charm if I had her stand waaaaaaaaaaaay back. They would then swap to trying to Charm my Priest. Always.

 

So I could reliably trivialize the Fampyr encounters by abusing the omnicient stupidity of the Fampyr by putting my Cipher just around the corner out of LOS but close enough to stay on their near hardcoded aggro list while they ran into my tank wall trying to gain LOS on the Cipher and subsequently spending the first portion of the fight getting knocked prone and roflstomped. They never even "saw" my Cipher until they on the floor.

 

Anything 10+ on intellect shouldn't be quite that predictable. The AI scripting could use a touch more fuzz factor and less omniscience.

 

Absolutely. I'd be quite surprised if the AI doesn't already have some degree of probabilistic behavior built in. If X conditions exist, roll a die and pick your next action off of this weighted table. Etcetera. If their AI doesn't use something like that, that leads to some pretty exploitable stuff (as you mentioned). You might report that as a bug? More exposure for those kinds of exploits can never hurt.

Edited by Matt516
Posted (edited)

I used the choke point strategy like...  3 or 4 times in the game I don't see why it is so much of an issue. It's not like the game is forcing your hand.

Edited by Kimuji
Posted

I used the choke point strategy like...  3 or 4 times in the game I don't see why it is so much of an issue. It's not like the game is forcing your hand.

 

And it's not like you can do it everywhere.

Posted

I used the choke point strategy like...  3 or 4 times in the game I don't see why it is so much of an issue. It's not like the game is forcing your hand.

 

It is in PotD... Then again i've come to understand developers expected this behavior from people plaing PotD.

 

 

And it's not like you can do it everywhere.

 

 

Oh, but you can... Even in wilderness maps sometimes you have this little places that can screw the AI pathfinding and cause mobs to get stuck.

Posted (edited)

I used the choke point strategy like...  3 or 4 times in the game I don't see why it is so much of an issue. It's not like the game is forcing your hand.

 

It's not about what you did, it's about what the optimal thing to do is. And in any fight in which you can door block, you should door block (if you want to play "right"). That's just dictated by the current mechanics of the game and encounter design. Do you want to play a game in which you know exactly what to do whenever things get too hard? Or do you want to play a game in which you have to thoughtfully consider what you need to do in each encounter? The fact that the game doesn't "force your hand" (i.e. force you to play your best) even on the hardest difficulty is another problem entirely. But I digress...

 

I don't door-block much either, myself. But that's because I find it boring, not because I've found a better strategy. When door-blocking is possible, there is not a better strategy.

 

Same thing with the other thing I complain about all the time, the Estoc. The Estoc is the best 2H melee weapon in the game, and its only shortcoming (piercing damage) is made up for by the Poleaxe (its sister 2H weapon in the weapon focus). If you want to play a 2H frontline melee weapon character, and you want to do it well, the Estoc is the best choice. Period. And I hate that. I want meaningful choices, not the choice of "right" or "fun".

Edited by Matt516
  • Like 2
Posted (edited)

I used the choke point strategy like...  3 or 4 times in the game I don't see why it is so much of an issue. It's not like the game is forcing your hand.

 

 

Agree with this. Seems people are wanting to make a mountain out of a molehill for fights that you can door block not more than 10 times in the game.

 

 

I used the choke point strategy like...  3 or 4 times in the game I don't see why it is so much of an issue. It's not like the game is forcing your hand.

 

It's not about what you did, it's about what the optimal thing to do is. And in any fight in which you can door block, you should door block (if you want to play "right"). That's just dictated by the current mechanics of the game and encounter design. Do you want to play a game in which you know exactly what to do whenever things get too hard? Or do you want to play a game in which you have to thoughtfully consider what you need to do in each encounter? The fact that the game doesn't "force your hand" (i.e. force you to play your best) even on the hardest difficulty is another problem entirely. But I digress...

 

I don't door-block much either, myself. But that's because I find it boring, not because I've found a better strategy. When door-blocking is possible, there is not a better strategy.

 

Same thing with the other thing I complain about all the time, the Estoc. The Estoc is the best 2H melee weapon in the game, and its only shortcoming (piercing damage) is made up for by the Poleaxe (its sister 2H weapon in the weapon focus). If you want to play a 2H frontline melee weapon character, and you want to do it well, the Estoc is the best choice. Period. And I hate that. I want meaningful choices, not the choice of "right" or "fun".

 

 

 

As you said, since the game doesn't force our hands and you chose to digress, thereby fighting without the door block strategy, you let yourself play in the most optimal way (without door block strategy of course). You still get to play in the most optimal way. 

I personally am okay with it. I find it a legit strategy and I enjoy doing so. Door blocking in certain situations is the optimal way for me. I don't really see a problem. The fights aren't so very difficult we have to door block to win them.

 

If I see a bunch of ranged enemies, of course I'm going to adapt by doing a bit of kiting if the space allows. If there are tons of heavy hitting melee, of course I'm going to fight at a choke point if a choke point is available. Most importantly is that door blocking isn't 50% of the fights. It isn't even 20%. Its a non-issue imo.

Edited by Jerm
Posted

Ironically the developers presented the engagement system in an example of a doorway... Anyway, I have found that a good way of forcing me to be pro active is when enemies in the back have attacks that go through armor - like fire of drakes or magic - in these few cases where I had to strategically block the doorway to not get overwhelmed enemies. Teleporting foes can also mess up with my back line.

Posted

Make doorways more attractive Foe AoE targets - especially if it's, say, a raw DoT spell - for enemy AI as long as there's at least one partymember within its bounds?

  • Like 1
Posted

Make doorways more attractive Foe AoE targets - especially if it's, say, a raw DoT spell - for enemy AI as long as there's at least one partymember within its bounds?

Or hand out a few more "Line" and long cone attacks to mobs with ranged scripting (by ranged scripting I mean having a minimum range the mob tries to maintain from *Any* hostile, not just the one at the top of their hate list).

 

Then there will be at lease a few encounters where one or more mobs actually hang back outside of splash AE range and throw painful things through the tank plug into the squishies in the back.

  • Like 1
Posted (edited)

I don't see it as a problem, personally.

 

Pathing glitches aside (abusing pathing in an otherwise open area is a bit exploity imo) its not available everywhere and doorway choke points kind of feel like a CRPG and tabletop staple. I remember as a child late at night on my C64 coming up with the elite doorway strategy in Pool of Radiance to win that horde of orcs and hobgoblins fight on Sokal Keep.

 

Enemy casters should probably be programmed to AE through the doorway (just like we do) and pick off backline characters with archery a little more in this situation.

 

I remember one area after I dropped my Priest down to robes and archers started smashing him through the doorway. Made his positioning very difficult, was quite a challenging fight even with the doorway line holding. Unfortunately since then I stuck him in Scale and he hasn't been focused anywhere near as much :( Healers probably need some AI priority on Hard and above.

Edited by Hogfather
Posted (edited)

1. Push mechanics are in the game. Pull mechanics are in the game. The AI should be able to respond to players bottlenecking by actually MOVING them.

 

2. To give additional tools (for both monsters and players), the Prone status should allow others to walk over you. So there would be 3 tools then: push, pull, and prone.

 

3. AI should be rather heavily random. We understand it's difficult to write an AI script which consistently recognizes chokepointing behavior and counters it. Don't put all the eggs in that basket. Have the AI push and pull and prone and teleport and everything else at least pseudorandomly so it is always a potential threat.

Edited by scrotiemcb
  • Like 1

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...