Putting the 'role' back in role-playing games since 2002.
Donate to Codex
Good Old Games
  • Welcome to rpgcodex.net, a site dedicated to discussing computer based role-playing games in a free and open fashion. We're less strict than other forums, but please refer to the rules.

    "This message is awaiting moderator approval": All new users must pass through our moderation queue before they will be able to post normally. Until your account has "passed" your posts will only be visible to yourself (and moderators) until they are approved. Give us a week to get around to approving / deleting / ignoring your mundane opinion on crap before hassling us about it. Once you have passed the moderation period (think of it as a test), you will be able to post normally, just like all the other retards.

Quickfire Systemic Criticism that contributes to banality of gameplay

Unwanted

Hatred

Unwanted
Joined
Apr 5, 2015
Messages
477
Location
Pit of Despair
Classes are defined by endurance numbers, health multipliers, starting accuracy and defense scores and their abilities on level up. We all know that Josh prefers classless systems so it's no surprise that his class-based system is kinda/sorta similar to classless in a way.
Wouldn't if have made more sense to actually have a classless system instead of this half ass half class system >.> PoE the job half done is how I will remember this game.
 

Sensuki

Arcane
Joined
Oct 26, 2012
Messages
9,800
Location
New North Korea
Codex 2014 Serpent in the Staglands Shadorwun: Hong Kong A Beautifully Desolate Campaign
Well they chose to use the Infinity Engine games to sell their product so they technically had to have classes, the game had to be real-time with pause and they also had to do a bunch of other shit that they promised, like their stretch goals.
 

sser

Arcane
Developer
Joined
Mar 10, 2011
Messages
1,866,688
I said real-time, not "real time strategy." But the concept is the exact same regardless of how the game operates - even moreso using "real-time with pause" which is a bastardized version of turn-based if you play on non-cuddly difficulty levels. Fast-paced RTS games aren't going to have disengagement/ZOC mechanics put into the code because the physical act of 'retreating' in an RTS game means you're going to be eating damage anyway. The combat-penalty for disengaging needs no simulacrum because the real (heh) consequences are abundantly obvious.
The Infinity Engine games are not real-time strategy, neither is Aarklash Legacy. Real time with pause is not a bastardization of turn-based, it's real-time that allows you to pause and issue commands to many characters without having to worry about being quick at micro. The difference between real time and turn-based is quite obvious - unit actions do not occur simultaneously in turn-based, which is one of the reasons why AoOs and Overwatch features exist. In real time, unit actions occur at the same time/units can act simultaneously.

Huh? The only reason I even discussed "RTS" in the first place was because you mentioned it. I did not say the Infinity Engine games as being RTS, or even mention Aarklash for that matter.

PilloE employs a system not dissimilar from, say, Jagged Alliance: Back in Action. The timed attacks - literally measured in seconds - are nothing more than glorified Initiative, granting certain persons faster attacks than others based on what weapons and gear they are wearing, frequently correlated to potential damage. This is the slickest way to blend real-time and turn-based. While some see this as 'real-time', I'm sitting there counting turns for what they are. I know whose turn is next because I can see the counter. I can rearrange who attacks first by changing the Initiative - ahem, the 'speed' of the weapons and the 'weight' of the armor. I then juggle in more pauses than an actual turn-based game. This is the very bastardization I speak of.



The AI in Pillars of Eternity is worse than the Infinity Engine games, but it has nothing to do with what you quoted. How should enemies act relative to Melee Engagement? Because if they didn't target you they'd suffer disengagement attacks, which means you can abuse the shit out of the system.

Don't kid yourself - the AI of all these games are virtually identical. To think there's a trickster algorithm hidden in either one is being a bit cheeky with your imagination. Most of the AI is concerned with casting the right spells, which I'd say they do well in both games; well enough to be decently challenging when the spellcasters are around, anyway.

As far as how they AI should behave... well, I'm not sure how exactly you want to discuss Disengagement. On one hand you say it's shite that you get wrecked for being in a bad position and disengaging, and on the other you say the AI is dumb and you abuse it. The latter is just bad AI which can be fixed and has nothing to do with systemic issues, the former is you being too stubborn to admit it's your own fault your bros got diced, instead blaming... the system. Although in this case I'd blame the AI :smug:





No. The AI in Icewind Dale Heart of Winter is very good. The best of all of the Infinity Engine games. Personally I don't think you have any idea what good AI is. What real-time game do you think has good AI? Your answer will probably be none. [later on...] And regarding AI, what exactly do you think would be good AI? Because the AI targeting in IWD:HoW is pretty fucking pristine, one thing that could be improved is ability/spell use.

The same AI that allows you to step away for a moment, and then come back to a reset target pathing? Yes, very good AI indeed. If you really think IWD has "very good" AI and isn't just a monster-mash loot-collecting game set in a pretty cool environment, well, that's your prerogative. All I saw was the exact same AI I always saw. Can't really change either point of view, really. The vast majority of games in this ilk are not all that concerned with AI. It just needs to be "not bad" and let the puzzle-esque nature of combat scenarios give the player a challenge. Wanna know how I know this? Because PilloE would not be any better if you gave it great AI. Why? Because the designs of the combat scenarios are almost universally lazy.





I do not think of games like real life, because they aren't. I played competitive games and one of the reasons why I was really good at the games I focused on is because I thought of things from the perspective of how they worked in the game, not how they worked in real life. These games are not realistic and they are not simulationist. The theory of zone of control isn't even realistic in the first place.

Zone of Control simulates the idea that if you are in combat with someone, and try and disengage, you will be at a disadvantage. If a swordfight is happening, Zone of Control says, "You know I'm gonna stab you in the back if you turn around, right?" It also simulates the fact that if you want to get from point A to Z, and there's a dude in your way, you can't just walk around them. There's a blocking scheme in American football that peels off big dudes to match them up with other big dudes, so that the smaller dude can squeeze through gaps. Wanna guess what this blocking scheme is called? Zone blocking. D'ohmygawd.

You think PilloE has real-time, but no dude in the game is going to have a procedurally reactive lunge at some guy turning around, or a procedurally reactive clothesline of someone trying to stupidly walk past him. They'll still be tied to their Initiative - I mean, their timed attacks. PilloE doesn't have the actual real-time combat to simulate you getting stabbed in the back, so the answer is it puts in Disengagement.

And I'm surprised that you think the broken AI chasing a character in circles is stupid (which it is, but it's fixable), and not the notion that, in your system, you could simply be chased in circles anyway, unless the AI was stupid a la IE, where your glorified repositioning makes the AI break off. You'd literally have to dumb down the AI to prevent the very situation you've already demonstrated as stupid. It's just a different side of the same shitty coin.



Everything else you said was mostly rehashed. What I'll say here is that I'm not happy with PilloE's combat for various reasons and I'm not happy with the Engagement system because it doesn't have enough proactive elements tied into it. But I like the fundamentals, because I've always liked Zone of Control or even its ugly cousin, MMO "threat". It makes the physical melee aspect important instead of being a slideshow to the fireworks displays that are the spellcasters. Yeah Zone of Control sucks balls if you wanna put on sunglasses and cast magic missile all day, but I for one kinda like the notion that if a big dude with a shield and hammer came down on your mage, you'd need your fighter to come save the day. Instead of hiking up your cloak and just Zoidberging out of there until the Big Bad gave up and turned around, ostensibly unaware that as soon as he did so you would too.
 

Athelas

Arcane
Joined
Jun 24, 2013
Messages
4,502
I think what bothers me the most about the system is that so many of the choices you can make across character builds are minor stat improvements that never really encroach upon the utility of anything. +5 deflection here, +6 accuracy there and I end up thinking maybe a nap would be an exciting alternative.

Most of the games I've played that took place in real-time were far more fun when you could acquire items or abilites that actually changed the way you played the game from moment to moment(and I think this is a point someone may have made on this website in the DotA 2 thread? not sure). I'm also pretty surprised they left stunlocking in the game.
What makes you think character building has any bearing on how the combat mechanically plays out? You can have the best combat ever made without allowing for any sort of customization (Frozen Synapse is a good example). Besides, character abilities are what define classes the most, not talents (although talents can certainly help create distinct builds). The issue with the current system is that it favors extreme specialization too much. This could be remedied by making attributes give larger bonuses in the mid-range, i.e. at the 8-14 range or something similar. Perception, Constitution and Resolve also need an overhaul.

This is in isolation of course. The game's encounter also design needs to be reconsidered, as well as systemic stuff like per-encounters and the engagement system.
 
Last edited:

Shannow

Waster of Time
Joined
Sep 15, 2006
Messages
6,386
Location
Finnegan's Wake
Inventory?
H4OLx3H.jpg
Shannow those are bugged stats. The 291 damage is probably a crit from scroll use.
Gb8WbG7.jpg

3rzDs3z.jpg

ATvSUs4.jpg

Q1iuc0i.jpg

lYxUTZ3.jpg

qjHIOIV.jpg

us7krad.jpg
 
Last edited:

Sensuki

Arcane
Joined
Oct 26, 2012
Messages
9,800
Location
New North Korea
Codex 2014 Serpent in the Staglands Shadorwun: Hong Kong A Beautifully Desolate Campaign
I said real-time, not "real time strategy." But the concept is the exact same regardless of how the game operates - even moreso using "real-time with pause" which is a bastardized version of turn-based if you play on non-cuddly difficulty levels. Fast-paced RTS games aren't going to have disengagement/ZOC mechanics put into the code because the physical act of 'retreating' in an RTS game means you're going to be eating damage anyway. The combat-penalty for disengaging needs no simulacrum because the real (heh) consequences are abundantly obvious.
The Infinity Engine games are not real-time strategy, neither is Aarklash Legacy. Real time with pause is not a bastardization of turn-based, it's real-time that allows you to pause and issue commands to many characters without having to worry about being quick at micro. The difference between real time and turn-based is quite obvious - unit actions do not occur simultaneously in turn-based, which is one of the reasons why AoOs and Overwatch features exist. In real time, unit actions occur at the same time/units can act simultaneously.

Huh? The only reason I even discussed "RTS" in the first place was because you mentioned it. I did not say the Infinity Engine games as being RTS, or even mention Aarklash for that matter.

PilloE employs a system not dissimilar from, say, Jagged Alliance: Back in Action. The timed attacks - literally measured in seconds - are nothing more than glorified Initiative, granting certain persons faster attacks than others based on what weapons and gear they are wearing, frequently correlated to potential damage. This is the slickest way to blend real-time and turn-based. While some see this as 'real-time', I'm sitting there counting turns for what they are. I know whose turn is next because I can see the counter. I can rearrange who attacks first by changing the Initiative - ahem, the 'speed' of the weapons and the 'weight' of the armor. I then juggle in more pauses than an actual turn-based game. This is the very bastardization I speak of.

Cooldowns between attacks is not a glorified initiative. The same system has been employed in real-time games for a fucking long time, dating all the way back to the 80s Arcade Games like Street Fighter, where there is a cooldown between when a player can perform an attack on their character but still move in between. It has nothing to do with turn-based and you are fucking delusional if you think it does. There are no turns. There are only actions. Just because they added a UI element to display a unit's recovery time is ticking down, that does not make it turn based, as it wouldn't if there was one shown in an arcade side scroller fighting game, or in Warcraft 3 (which uses a system almost identical to the Pillars of Eternity recovery time system).

You have also completely ignored the key difference between real-time and turn-based - that units perform actions simultaneously and units take turns in turn-based. In real time units can move at the same time and attack at the same time. You can move while your recovery timer is ticking down at the same time as other units are doing things. This is the problem that I think you and others have, is that you cannot fucking see the difference between real-time and turn-based and think that they must use the same mechanics or some shit.

Don't kid yourself - the AI of all these games are virtually identical. To think there's a trickster algorithm hidden in either one is being a bit cheeky with your imagination. Most of the AI is concerned with casting the right spells, which I'd say they do well in both games; well enough to be decently challenging when the spellcasters are around, anyway.

As far as how they AI should behave... well, I'm not sure how exactly you want to discuss Disengagement. On one hand you say it's shite that you get wrecked for being in a bad position and disengaging, and on the other you say the AI is dumb and you abuse it. The latter is just bad AI which can be fixed and has nothing to do with systemic issues, the former is you being too stubborn to admit it's your own fault your bros got diced, instead blaming... the system. Although in this case I'd blame the AI :smug:

There are targeting preferences differences, targeting loop differences between the Baldur's Gates and Icewind Dales and the latter titles in the series have more advanced AI that do stuff like use potions and whatnot. There is a difference between each title.

Most of the AI is not concerned with casting the right spells, the AI follows a script in the Infinity Engine games unless you play with an AI mod. Some of the key differences between the Infinity Engine AI and the Pillars of Eternity AI are as follows:

Unit targeting loops update slower in Pillars of Eternity when units are moving and when they already have a target and are stationary there is no running loop (which is a flaw) which is the reason why you see stupid shit like a line of spiders standing still doing nothing, because they can't find the path to their current target, are not moving (and thus not qualifying for the movement targeting loop) and thus will stand still doing nothing until they are engaged or the unit they are stuck behind moves or dies.

Targeting loops for enemies in the Infinity Engine games run every frame while units are moving, and when they're not moving they run on a loop every X seconds. The clauses are also more robust.

Enemy AI in Pillars of Eternity uses a weighted system which is pretty much the same as the stuff you do in IP routing. Josh Sawyer will give the AI programmer a 'struct' for each AI type, and they will assign a weight to each action in the struct (like a route might be assigned a weight in BGP routing). Enemy Rogue AI might have their Blinding Strike and Crippling Strike abilities set to a weight of 100, which means they'll open encounters with either one of those abilities and then fall back on their standard attack (which might be weight 50) and then other actions will have a lower weight and those weights will be raised or lowered if certain conditions are met.

The Infinity Engine AI follows a script. The Mustard Jelly dude in the area at the back of the Nashkell mines will always cast Invisibility followed by Lightning Bolt, and there is nothing that will vary this formula (and many other Wizards use the same script, such as Silke).

The Infinity Engine games have better targeting AI than Pillars of Eternity and better in combat Pathfinding. Pillars of Eternity has 'smarter' ability use AI but lacks things like enemy ability to use potions and consumables and stuff like that. By smarter I mean it's a more elegant system than the Infinity Engine single script, but the result isn't really much different.

]The same AI that allows you to step away for a moment, and then come back to a reset target pathing? Yes, very good AI indeed. If you really think IWD has "very good" AI and isn't just a monster-mash loot-collecting game set in a pretty cool environment, well, that's your prerogative.

Yes I do think IWD:HoW has very good targeting AI (but not AI for abilities), because if enemies don't change targets when you start moving away then the AI is ripe for abuse. It makes the most sense to have them switch targets in reaction to their current target's movement. As I previously stated there is an enemy AI loop that runs every so often, and this will cause enemies to re-think their targeting, which often means they will change targets in combat, and as a player you have to react to that.

I think that works very well, the targeting updates are snappy and robust.

The vast majority of games in this ilk are not all that concerned with AI. It just needs to be "not bad" and let the puzzle-esque nature of combat scenarios give the player a challenge. Wanna know how I know this? Because PilloE would not be any better if you gave it great AI. Why? Because the designs of the combat scenarios are almost universally lazy.

I know, I already said that.

Zone of Control simulates the idea that if you are in combat with someone, and try and disengage, you will be at a disadvantage. If a swordfight is happening, Zone of Control says, "You know I'm gonna stab you in the back if you turn around, right?" It also simulates the fact that if you want to get from point A to Z, and there's a dude in your way, you can't just walk around them. There's a blocking scheme in American football that peels off big dudes to match them up with other big dudes, so that the smaller dude can squeeze through gaps. Wanna guess what this blocking scheme is called? Zone blocking. D'ohmygawd.

Except, combat in these games is not a one on one fight. It is a battlefield with multiple units on either side. Combat is also an abstraction. I find it strange that you and others seem to think that if a character model physically turns in game that it means that that character has dropped their guard, but think it's okay for units to get free invisible attacks against other units while they're currently performing another animation. If games had unlimited budgets, I'm sure that animators could create blended animations for characters to display that when they do move in combat that they still face their enemy and don't drop their guard when they're close to them, but games do have a budget and you can't spend it all on animation just to please simulationists. I don't see moving in combat as dropping one's guard because it's an abstraction and melee combat is not a stationary thing ... but I suppose ... for people who think it's turn based (like yourself) I can see how you lean on the stationary shit so much.

You think PilloE has real-time, but no dude in the game is going to have a procedurally reactive lunge at some guy turning around, or a procedurally reactive clothesline of someone trying to stupidly walk past him. They'll still be tied to their Initiative - I mean, their timed attacks. PilloE doesn't have the actual real-time combat to simulate you getting stabbed in the back, so the answer is it puts in Disengagement.

Yes it does, units act simultaneously. As previously stated recovery time has nothing to do with turn-based, this is not a turn-based game and units can move during their recovery time anwyay, they just have to wait to perform their next non-movement action.

And I'm surprised that you think the broken AI chasing a character in circles is stupid (which it is, but it's fixable), and not the notion that, in your system, you could simply be chased in circles anyway, unless the AI was stupid a la IE, where your glorified repositioning makes the AI break off. You'd literally have to dumb down the AI to prevent the very situation you've already demonstrated as stupid. It's just a different side of the same shitty coin.

No. Units that don't change targets can be kited forever by ranged units (or melee units, with the absense of the engagement AI clause) and that is worse than units that do change targets if they lose their current target and are forced to move. In isometric real-time games where there are ranged units and the party has superior numbers and enough pathing space to move around there will be kiting. The best targeting implementation to deal with that is to have units change targets, rather than always target the same guy.

Everything else you said was mostly rehashed. What I'll say here is that I'm not happy with PilloE's combat for various reasons and I'm not happy with the Engagement system because it doesn't have enough proactive elements tied into it. But I like the fundamentals, because I've always liked Zone of Control or even its ugly cousin, MMO "threat". It makes the physical melee aspect important instead of being a slideshow to the fireworks displays that are the spellcasters.

I dunno, buffed up Fighters and whatnot are pretty scary in the 2E IE games. I enjoy playing a Kensai and I wreck shit with dual katanas zipping around the battlefield like a boss.

Yeah Zone of Control sucks balls if you wanna put on sunglasses and cast magic missile all day, but I for one kinda like the notion that if a big dude with a shield and hammer came down on your mage, you'd need your fighter to come save the day. Instead of hiking up your cloak and just Zoidberging out of there until the Big Bad gave up and turned around, ostensibly unaware that as soon as he did so you would too.

You don't need a zone of control to fix this issue. Your Fighter has a fucking knockdown ability - problem solved. Disables are the key to controlling melee combat in these types of games, something which melee characters in the Infinity Engine games lacked but other games like MOBAs make good use of stuff like stuns and holds. The Engagement system is unnecessary with the presence of melee disables like this. A systemic zone of control solution is unnecessary and active solutions like giving units the ability to disable is a much better, tactical and fluid solution.
 
Last edited:

hell bovine

Arcane
Joined
Sep 9, 2013
Messages
2,711
Location
Secret Level
Don't kid yourself - the AI of all these games are virtually identical. To think there's a trickster algorithm hidden in either one is being a bit cheeky with your imagination. Most of the AI is concerned with casting the right spells, which I'd say they do well in both games; well enough to be decently challenging when the spellcasters are around, anyway.

As far as how they AI should behave... well, I'm not sure how exactly you want to discuss Disengagement. On one hand you say it's shite that you get wrecked for being in a bad position and disengaging, and on the other you say the AI is dumb and you abuse it. The latter is just bad AI which can be fixed and has nothing to do with systemic issues, the former is you being too stubborn to admit it's your own fault your bros got diced, instead blaming... the system. Although in this case I'd blame the AI :smug:
The AI can't deal with the engagement system either, that's the funny thing (or would be funny, if it didn't make the battles so repetitive). If the enemy breaks engagement, they will get hit for it (with few exceptions like shadows) and you can use it, if it doesn't, it will effectively tie them down in place and you can use it too. The majority of enemy critters will continue to fight against an opponent that cannot even hit them in melee, but at the same time will completely ignore the ranged squad standing two meters away shooting them full of holes. No narrow corridors necessary, this works in the open field. And playing on PotD, when there are too many critters and not everyone can get to my chanter (who doubles as a tank, because even if you can't tell the hilt from the sharp end, you can still effectively engage an enemy in melee in PoE) the rest will run around them in circles, also ignoring said ranged squad.
 

Sensuki

Arcane
Joined
Oct 26, 2012
Messages
9,800
Location
New North Korea
Codex 2014 Serpent in the Staglands Shadorwun: Hong Kong A Beautifully Desolate Campaign
That's due to a combination of the Engagement system, bad pathfinding and the fact that a targeting loop doesn't run when enemies aren't moving.

I reported the targeting loop issue as a bug in BB v485, issue ignored or put on the backlog.
 

ArchAngel

Arcane
Joined
Mar 16, 2015
Messages
20,059
most points I agree, not the one about armor though.
Having armor that only raises AC has always been one of the major downsides of D&D. A lot of other systems have damage reduction for heavy armor with some kind of downside to it, and that is for the best. It is more diverse, fun and, for those who care (me) realistic.
That they didn't handle it very well is another matter. While I don't have any particular issues with the system as it is now (in fact, atm I'm using a wide variety of armors with my chars and doing very well), I can see why it could use improvements in the future. But it is not worse than D&D, no way.
There is nothing more realistic about armor absorbing damage in these systems. It is more realistic in system where there are different hit spots on body and where one good stab maims or kills your character like in real life.

For D&D like heroic game armor that reduces chance to hit is as realistic as armor that reduces damage, it is just a math simulation of your heroic character being able to take many blows and keep fighting like nothing happened but unlike D&D in PoE light fast and low damage weapons are crap vs armor as damage reduction which is exact reason why it is not used in D&D. There is a optional rule in D&D of using DR of armors instead of bonus to AC and there it also warns players about this change probably making many weapons useless.

Knives should be useless against plate armor unless the motherfucker is pinned to the ground by two other guys and you're stabbing through his eye holes. I know more leeway is (and should be) given for fantasy games with physical abilities bordering on, if not being well into the supernatural, but there is no need to make that a part of the armor system proper.
Daggers are not just knives. And these armor system make more than daggers useless. And of course 2 average guys needed to get the armored guy on ground to stab him, that is all part of the attack system and description system of D&D. You roll attack and hit with your dagger vs full plate guy in normal D&D and describe it exactly how you did, that is actually how you are supposed to do it in D&D. Tnx for proving my point.
 

Ninjerk

Arcane
Joined
Jul 10, 2013
Messages
14,323
I think what bothers me the most about the system is that so many of the choices you can make across character builds are minor stat improvements that never really encroach upon the utility of anything. +5 deflection here, +6 accuracy there and I end up thinking maybe a nap would be an exciting alternative.

Most of the games I've played that took place in real-time were far more fun when you could acquire items or abilites that actually changed the way you played the game from moment to moment(and I think this is a point someone may have made on this website in the DotA 2 thread? not sure). I'm also pretty surprised they left stunlocking in the game.
What makes you think character building has any bearing on how the combat mechanically plays out? You can have the best combat ever made without allowing for any sort of customization (Frozen Synapse is a good example). Besides, character abilities are what define classes the most, not talents (although talents can certainly help create distinct builds). The issue with the current system is that it favors extreme specialization too much. This could be remedied by making attributes give larger bonuses in the mid-range, i.e. at the 8-14 range or something similar. Perception, Constitution and Resolve also need an overhaul.

This is in isolation of course. The game's encounter also design needs to be reconsidered, as well as systemic stuff like per-encounters and the engagement system.
OK, but in the game we're actually talking about the designers clearly intended for you to make some choices every time you level up, whether with respect to class or general talents. My problem is that those choices (even on a class level) are boring and "move the needle" in barely perceptible ways.
 

Ninjerk

Arcane
Joined
Jul 10, 2013
Messages
14,323
Zone of Control simulates the idea that if you are in combat with someone, and try and disengage, you will be at a disadvantage. If a swordfight is happening, Zone of Control says, "You know I'm gonna stab you in the back if you turn around, right?" It also simulates the fact that if you want to get from point A to Z, and there's a dude in your way, you can't just walk around them. There's a blocking scheme in American football that peels off big dudes to match them up with other big dudes, so that the smaller dude can squeeze through gaps. Wanna guess what this blocking scheme is called? Zone blocking. D'ohmygawd.

Offensive linemen have to move to continue blocking people that try to walk around them. They can't just grab the back of their jersey and start punching them in the head for walking around them.
 

Perkel

Arcane
Joined
Mar 28, 2014
Messages
15,875
Like i said in some other thread.
I feels like combat system targeted toward PvP fights and in very small group (like 2-3 or less)
 

Athelas

Arcane
Joined
Jun 24, 2013
Messages
4,502
I think what bothers me the most about the system is that so many of the choices you can make across character builds are minor stat improvements that never really encroach upon the utility of anything. +5 deflection here, +6 accuracy there and I end up thinking maybe a nap would be an exciting alternative.

Most of the games I've played that took place in real-time were far more fun when you could acquire items or abilites that actually changed the way you played the game from moment to moment(and I think this is a point someone may have made on this website in the DotA 2 thread? not sure). I'm also pretty surprised they left stunlocking in the game.
What makes you think character building has any bearing on how the combat mechanically plays out? You can have the best combat ever made without allowing for any sort of customization (Frozen Synapse is a good example). Besides, character abilities are what define classes the most, not talents (although talents can certainly help create distinct builds). The issue with the current system is that it favors extreme specialization too much. This could be remedied by making attributes give larger bonuses in the mid-range, i.e. at the 8-14 range or something similar. Perception, Constitution and Resolve also need an overhaul.

This is in isolation of course. The game's encounter also design needs to be reconsidered, as well as systemic stuff like per-encounters and the engagement system.
OK, but in the game we're actually talking about the designers clearly intended for you to make some choices every time you level up, whether with respect to class or general talents. My problem is that those choices (even on a class level) are boring and "move the needle" in barely perceptible ways.
Of course class abilities affect how you play the game - you have to actually use them after all. For instance, if you pick a cipher's antipathetic field and cast it on an enemy, you have to be mindful of allies getting in the way of the corrosive beam. Unless you mean the game is too easy to warrant much thought about such things, but that's a separate issue.

As for the talent design, a single talent can triple the damage of a blunderbuss.:M Even something as seemingly minuscule as +6 accuracy can be a huge boon with the way the game's systems work.

I simply don't see the customization as the problem - the problem is the deeper systemic issues like the engagement system and the lack of impactfulness of status ailments.
 

Ninjerk

Arcane
Joined
Jul 10, 2013
Messages
14,323
I can do that too :nocountryforshitposters:

EDIT: Go look at solo PotD threads on Obsidian boards. Everyone wants to get Sanguine Plate or the retaliate shield from the ruins north of Dyrford, because it's one of the few items that can drastically change the damage output of a character who gets hit a lot. SP is a gamechanger, not the Fighter Class.

TCS pretender Ceranai makes a comment about his Chanter attempt: http://forums.obsidian.net/topic/76328-solo-play-tips-not-exploits-potd/page-3#entry1651222
Ceranai said:
lol the figurine makes it 1000x easier though. right now i only use my figurines to open up fights, then i let my summons take over. That and my sanguine plate does more damage than i do, it depends if i let myself get hit or not

Solo player MadDemiurg is running retaliation Cipher (not sure about difficulty or TCS status)
https://forums.obsidian.net/topic/76099-path-of-the-damned-solo-achieved/#entry1640657
MadDemiurg said:
Props! I'm currently act 3 with retaliation cipher, pretty sure I'll be able to finish the rest since L6 spells make things quite easy. Don't have that much time to play now.

Solo PotD Achievement holder "I am Catharsis" :)roll:)
https://forums.obsidian.net/topic/76099-path-of-the-damned-solo-achieved/page-2#entry1649587
I am Catharsis said:
Retaliation: This is what this build is all about. When you Parry / Dodge, you have 30% to hit back your opponent. With a high deflection, you will hit like crazy in big packs of mobs. You will get access to it level 5.
 
Last edited:

KK1001

Arbiter
Joined
Mar 30, 2015
Messages
621
The character customization is actually pretty good. And the idea that you should avoid trap builds and have, with the right tactics employed, every class be more-or-less effective is good in theory.

The main issue is, as Athelas pointed out, the soft counter nature of the game and the bad encounter design.

A comprehensive mod that reworked encounters from the ground up, tweaked the value of status ailments, and altered the AI would go a long way towards fixing many of the games problems.
 

Shevek

Arcane
Joined
Sep 20, 2003
Messages
1,570
Theres nothing fundamentally or ideologically wrong with soft counters or the core game systems. There are plenty of encounters that feel good when you hit them and arent over levelled. There are some issues steming from from a critical path that needs to be tuned up in difficulty and encounters that needs to be spiced up with better AI and enemy variety. This all should be moddable. But it seems like it would be a significant chore to edit every encounter in the game across each difficulty.
 

KK1001

Arbiter
Joined
Mar 30, 2015
Messages
621
There's nothing wrong with soft counters or the game system in practice, sure. But in practice, a lot of encounters - especially, as you say, along the critical path - lack variety and challenge. Much of the game feels like a first draft in terms of encounter design. Simple stat and AI tweaks will help, but there's a deeper problem. I shouldn't roll through POTD using the most obvious strategy for every encounter.

The engagement system is partly at fault here. Many encounters are determined, especially on POTD, by your initial positioning. If you can get the right enemies engaged in the right away, you've pretty much won. The AI isn't robust enough to handle an engagement system, and there aren't enough skills that allow the player to escape from engagement. Fights are determined before they've really gotten on and pretty much regardless of player action (besides casting damage spells and the same debuffs over and over again). This is exacerbated by the lack of hard counters, too.
 
Last edited:

Seari

Arcane
Patron
Joined
Nov 25, 2014
Messages
849
Pathfinder: Wrath

Shevek

Arcane
Joined
Sep 20, 2003
Messages
1,570
I won every fight before it started in the IE games as well. As long as you didnt place any artificial limits on yourself, strategy easily trumped tactics in those games. The issue I think is that PoE wussed out a bit on the crit path. If enemies were higher level and had higher defenses, you might need to use more than the same strat over and over. Similarly, soft counters are fine. Shit doesnt need to be immune to knockdown but things can be resistant (like Aumauns, etc). There arent enough resistant enemies or enemies that use as varied a skill set as the player can. Engagement isnt a problem. I enjoy the system. The issue is pretty much just with kinda lazy enemy and encounter design. Its better than BG1 but doesnt match Mid BG2/IWD or IWD2. We knew this would be an issue going in since the devs pretty much said that the AI wouldnt be as good as IWD:HoW since they are barely getting to know this engine.
 

Ninjerk

Arcane
Joined
Jul 10, 2013
Messages
14,323
There's nothing wrong with soft counters or the game system in practice, sure. But in practice, a lot of encounters - especially, as you say, along the critical path - lack variety and challenge. Much of the game feels like a first draft in terms of encounter design. Simple stat and AI tweaks will help, but there's a deeper problem. I shouldn't roll through POTD using the most obvious strategy for every encounter.

The engagement system is partly at fault here. Many encounters are determined, especially on POTD, by your initial positioning. If you can get the right enemies engaged in the right away, you've pretty much won. The AI isn't robust enough to handle an engagement system, and there aren't enough skills that allow the player to escape from engagement. Fights are determined before they've really gotten on and pretty much regardless of player action (besides casting damage spells and the same debuffs over and over again). This is exacerbated by the lack of hard counters, too.
I can't remember if I had this thought when I was reading about engagement before the game released or if it was upon initially playing the game, but the engagement system sounds ironically like something suited better for adversarial, real-time multplayer (likely with some tweaks on account of the almost "resourceless" nature of disengagement attacks). Unfortunately, double binds are far less effective with AI like PoE has.

EDIT:
Solo PotD Achievement holder "I am Catharsis" :)roll:)
https://forums.obsidian.net/topic/76099-path-of-the-damned-solo-achieved/page-2#entry1649587
I am Catharsis said:
Retaliation: This is what this build is all about. When you Parry / Dodge, you have 30% to hit back your opponent. With a high deflection, you will hit like crazy in big packs of mobs. You will get access to it level 5.
So I guess Retaliation isn't bad after all, lols

Did someone say it was? I don't remember saying that, but I could be wrong. I think Sanguine Plate and the other item (without remembering specifically looting it) are one of the few examples of good itemization, and I'm surprised they made the cut.
 

sser

Arcane
Developer
Joined
Mar 10, 2011
Messages
1,866,688
There are no turns. There are only actions. Just because they added a UI element to display a unit's recovery time is ticking down, that does not make it turn based, as it wouldn't if there was one shown in an arcade side scroller fighting game, or in Warcraft 3 (which uses a system almost identical to the Pillars of Eternity recovery time system).

In PilloE, I can put heavy armor and a big sword on a character, ensuring they act last. Or I can give them light armor and daggers, ensuring they act first. Not even taking into account a litany of other features, I'm not sure how this is comparable to a frame-by-frame fighting game, or Warcraft 3 for that matter.


You have also completely ignored the key difference between real-time and turn-based - that units perform actions simultaneously and units take turns in turn-based. In real time units can move at the same time and attack at the same time. You can move while your recovery timer is ticking down at the same time as other units are doing things. This is the problem that I think you and others have, is that you cannot fucking see the difference between real-time and turn-based and think that they must use the same mechanics or some shit.

Units can take turns simultaneously in turn-based. They can also go all one side, and then all the other. They can also go by initiative. They can also use a you-go-I-go system. There are phase-based turn-based games and there are clocked/timed turn-based games. Some have even blended player-skill into them. There are a lot of ways turn-based is handled.

The RTWP games typically borrow from many of them, because the whole purpose of RTWP from the get-go was to take a turn-based game and make it more animated, more fancy. Understanding the context of why BG was BG vis a vis its relationship to RTS games and Diablo is important here. The #1 problem was always how do you make an aesthetically-pleasing game with deep mechanics but not make it about the player's physical skills a la shooters or RTS games like Starcraft or clickfests like Diablo. The answer for years now has been to borrow turn-based elements and blend them into the game. It's very slick and designers are slowly getting better at it. I'm at a loss as to how one sees gameplay mechanics like Cooldowns or 'Per [X]' abilities or Initiative-counters, I mean Recovery Counters and not see the lineage of the turn-based genre.



[AI stuff.]

In terms of quality, I find the AI between these games largely indistinguishable (comparatively speaking). I think we have a good understanding of each others views here already.



Except, combat in these games is not a one on one fight. It is a battlefield with multiple units on either side. Combat is also an abstraction. I find it strange that you and others seem to think that if a character model physically turns in game that it means that that character has dropped their guard, but think it's okay for units to get free invisible attacks against other units while they're currently performing another animation.

Oh, this is a problem compared to what, exactly? I can already shoot a gun, draw swords, and turn around before the bullet has reached its target. When my spear critical-strikes a guy decked out in plate armor he explodes into chunks of meat. I dunno what's the big deal about animation issues. It's a game with gamey mechanics, surprise surprise.



No. Units that don't change targets can be kited forever by ranged units (or melee units, with the absense of the engagement AI clause) and that is worse than units that do change targets if they lose their current target and are forced to move. In isometric real-time games where there are ranged units and the party has superior numbers and enough pathing space to move around there will be kiting. The best targeting implementation to deal with that is to have units change targets, rather than always target the same guy.

See, like I said, your answer is to dumb down the AI to make positioning better. I can see why you like it, because it allows the battle to be fluid via juggling units back and forth. But to me this is a very poor solution and not very forward thinking, regardless of your opinion on Zone of Control schemes (maybe there's an even better way). What games with ZoC do is try to make positioning better by making it a gameplay mechanic in and of itself. In games like IE, or the system you're suggesting, you have to dumb down the AI.

You don't see your solution as dumb AI, you see it from your point of view which is rewarding you with 'smart' positioning summarized as, "I moved my character until I didn't have to." You also tend to reap the benefits of an AI that doesn't act in a similar manner, because if they did that would be annoying as fuck, right?




I dunno, buffed up Fighters and whatnot are pretty scary in the 2E IE games. I enjoy playing a Kensai and I wreck shit with dual katanas zipping around the battlefield like a boss.

It's personal opinion on whether or not you enjoy melee-characters in those games. In my own opinion I think designers have had a difficult time making the bruisers - particularly in the utility department - relevant when compared to their spell-slinging bros. After all, who buffs the Fighters? What class do people most often dual-spec the Kensai into? The fighter-classes just hit stuff. Which is cool. Doing damage is alright and killing baddies is pretty cool especially when they explode. But I think what designers really want to do is say, "What if fighters had utility that added to the benefit of the party that didn't just involve subtracting enemies?" Why does a soul-slinger like Aretha Franklin hire bodyguards? To hit stuff, or to keep her untouched? Hmmm.



You don't need a zone of control to fix this issue. Your Fighter has a fucking knockdown ability - problem solved. Disables are the key to controlling melee combat in these types of games, something which melee characters in the Infinity Engine games lacked but other games like MOBAs make good use of stuff like stuns and holds. The Engagement system is unnecessary with the presence of melee disables like this. A systemic zone of control solution is unnecessary and active solutions like giving units the ability to disable is a much better, tactical and fluid solution.

How do you knock down a swarm of enemies? What happens when they get back up? How do you keep your mage safe in this scenario? You use the 'walk away until the AI breaks' ability, apparently. Or you give the fighter utility in 'holding down' enemies so squishier characters can be safe via actual, legitimate positioning, not just you bandying about breaking AI leashes.
 

Sensuki

Arcane
Joined
Oct 26, 2012
Messages
9,800
Location
New North Korea
Codex 2014 Serpent in the Staglands Shadorwun: Hong Kong A Beautifully Desolate Campaign
In PilloE, I can put heavy armor and a big sword on a character, ensuring they act last. Or I can give them light armor and daggers, ensuring they act first. Not even taking into account a litany of other features, I'm not sure how this is comparable to a frame-by-frame fighting game, or Warcraft 3 for that matter.

No, they act at the same time. Their first attack will begin at the same time if executed at the same time, their recovery times will be different. The character with the faster weapons and lighter armor will act more often, but their initial actions will begin immediately. Simply because the Daggers / No armor = faster attack speed.

It is 100% a real-time system and has nothing to do with turn-based. It is exactly the same as a Meat Wagon and a Ghoul making an attack in Warcraft 3 but the Ghoul attacking quicker, or spamming fast attacks in a sidescroller fighting game producing more overall attacks than a slower attack being spammed.

Units can take turns simultaneously in turn-based. They can also go all one side, and then all the other. They can also go by initiative. They can also use a you-go-I-go system. There are phase-based turn-based games and there are clocked/timed turn-based games. Some have even blended player-skill into them. There are a lot of ways turn-based is handled.

The RTWP games typically borrow from many of them, because the whole purpose of RTWP from the get-go was to take a turn-based game and make it more animated, more fancy. Understanding the context of why BG was BG vis a vis its relationship to RTS games and Diablo is important here. The #1 problem was always how do you make an aesthetically-pleasing game with deep mechanics but not make it about the player's physical skills a la shooters or RTS games like Starcraft or clickfests like Diablo. The answer for years now has been to borrow turn-based elements and blend them into the game. It's very slick and designers are slowly getting better at it. I'm at a loss as to how one sees gameplay mechanics like Cooldowns or 'Per [X]' abilities or Initiative-counters, I mean Recovery Counters and not see the lineage of the turn-based genre.

Units do not take turns simultaneously in turn-based. Stop making shit up. It does not matter how the turn order is determined, one unit acts at a time. In phase-based, yes, units can act simultaneously. We aren't talking about phase-based though are we.

[As a side-note, the default combat style for AD&D 2E is actually phase-based, of which, RTwP could kinda 'feel like' phase-based, with auto-pause if all units began their round timer at the same time (which doesn't happen very often)].

I do not believe RTwP has anything to do with turn-based (it's original creation), Real-time games have in the past been pausable, you just have not been able to issue unit actions while paused. Allowing players to issue units commands while paused takes away the stress of performing micromanagement quickly and makes it more manageable to control multiple units at once. It doesn't have shit all to do with emulating turn-based, if anything the first game that tried to do that was Neverwinter Nights.

Recovery counters were simply added in as a UI feature because the way in which they chose to handle recovery time (using percentage based increases and decreases) made it confusing for the player to be able to know when units would next be able to perform their next non-movement action while also not having a separate animation for combat idle and recovery time.
The Infinity Engine games had a separate animation for combat idle and recovery time and units performed actions in a string over their individual 'combat rounds', which started on the first frame that they performed a non-movement action in combat and could be completely asynchronous with other unit's timers like in Pillars of Eternity, they just used a different method to calculate them. It was not confusing as to when a unit would perform their next action.

I dunno what's the big deal about animation issues. It's a game with gamey mechanics, surprise surprise.

You're the one who thinks that a unit moving in combat is letting their guard down.

See, like I said, your answer is to dumb down the AI to make positioning better. I can see why you like it, because it allows the battle to be fluid via juggling units back and forth. But to me this is a very poor solution and not very forward thinking, regardless of your opinion on Zone of Control schemes (maybe there's an even better way). What games with ZoC do is try to make positioning better by making it a gameplay mechanic in and of itself. In games like IE, or the system you're suggesting, you have to dumb down the AI.

It's not dumbing down AI. You honestly can't believe that having a unit target one unit specifically is better than having them switch targets if they lose their current target. It is a superior solution. The first situation is rife for abuse and pathing/pathing space issues. You tell me a better AI solution, if you think you've got a better one.

You don't see your solution as dumb AI, you see it from your point of view which is rewarding you with 'smart' positioning summarized as, "I moved my character until I didn't have to." You also tend to reap the benefits of an AI that doesn't act in a similar manner, because if they did that would be annoying as fuck, right?

No, you would just change how you approach the AI. If you know how the AI works, you will be able to use it to your advantage. In DotA 2, issuing an attack command on a hero increases the 'threat' of a unit which makes nearby creeps attack that unit performing that action, and with that knowledge you know that if you want to get to a hero you don't issue an attack command on them because it aggros nearby creeps, unless you're Axe and you wan't to proc spins. Stuff like that.

But I think what designers really want to do is say, "What if fighters had utility that added to the benefit of the party that didn't just involve subtracting enemies?" Why does a soul-slinger like Aretha Franklin hire bodyguards? To hit stuff, or to keep her untouched? Hmmm

Even in D&D you can get stuff like Improved Trip. In MOBA games, melee units often have mobility or crowd control abilities. In Pillars of Eternity the Fighter has several crowd control abilities - Knock Down and Clear Out, in particular (the latter is AoE). This is one area I agree that was lacking in the Infinity Engine games.

How do you knock down a swarm of enemies?

Use an AoE knockdown, or get another unit to do it for you (a caster).

What happens when they get back up?

Have a look at which units they're targeting, and then decide what to do.

How do you keep your mage safe in this scenario?

The same way you keep them safe in any game of this style, positioning and movement. Backline casters aren't often targeted by melee AI if they aren't into the fray early or if they're positioned behind other units. Against ranged units and casters well, that's a different story.

You use the 'walk away until the AI breaks' ability, apparently. Or you give the fighter utility in 'holding down' enemies so squishier characters can be safe via actual, legitimate positioning, not just you bandying about breaking AI leashes.

Walk away til the AI breaks still works in Pillars of Eternity, it's just that units don't change targets as fast as in the Infinity Engine games, there's a couple of seconds delay minimum. If Aloth is in the front and being targeted by a pack of Darguls but is not engaged, you can just simply move him away and the Darguls will target someone else. Controlling enemy aggro in games like this is done first through positioning and movement, and second by understanding what causes enemy AI to switch targets.

Engagement includes an AI clause, which is what gets enemies to switch targets. MMO threat based systems include targeting clauses, which gets enemies to switch targets. My point is that you do not NEED a system like Engagement or an MMO threat based system because it is already possible to control enemy AI targeting.
 
Last edited:

As an Amazon Associate, rpgcodex.net earns from qualifying purchases.
Back
Top Bottom