Jump to content

Welcome to Obsidian Forum Community
Register now to gain access to all of our features. Once registered and logged in, you will be able to create topics, post replies to existing threads, give reputation to your fellow members, get your own private messenger, post status updates, manage your profile and so much more. If you already have an account, login here - otherwise create an account for free today!
Photo

Raz's Perspective: A Dissertation on Engagement mechanics


  • Please log in to reply
52 replies to this topic

Poll: Engagement Mechanics (57 member(s) have cast votes)

Do you believe this engagement system should be implemented into Pillars of Eternity?

  1. Yes (26 votes [45.61%])

    Percentage of vote: 45.61%

  2. No (31 votes [54.39%])

    Percentage of vote: 54.39%

Vote Guests cannot vote

#21
Kjaamor

Kjaamor

    (6) Magician

  • Members
  • 682 posts
  • Pillars of Eternity Backer

Why would 10 goblins storm past your Fighter to your Wizard? AI targeting never did that in the Infinity Engine games and it does not do it in Pillars of Eternity either.

 

It depends upon how the targeting clauses work in the finished version. If those particular goblins' top targeting clause was to attack spellcasters, as a clear and obvious example, they would run past the Fighter to the Wizard.

 

More practically, targeting clauses are likely to be based around the traditional manner of damage received from, although on this matter I would acknowledge part of Bester's post:

 

- Why would I even want to reach the casters at the back? They're not as powerful as they were in the IE games, where you HAD to deal with the casters first.

 

...which I agree is becoming a bit of a design problem, and Josh may have to accept that.

 

 

Here's how BG1 and BG2 Targeting clauses worked for Melee Enemies. It was very simple.

 

It is incredibly simple, to the point of being simplistic, and is easily exploitable and creates combat incongruous to the theme. Not everyone liked every aspect of the IE games, and not everyone thinks that copying them verbatim is a good design policy for PoE and Obsidian more generally.



#22
nipsen

nipsen

    (5) Thaumaturgist

  • Members
  • 550 posts
  • Location:Norway
  • Steam:retroquark

Not everyone liked every aspect of the IE games, and not everyone thinks that copying them verbatim is a good design policy for PoE and Obsidian more generally.

 

Good grief! Really? Whence comes this new mystical lore to the earthly realms!



#23
Sensuki

Sensuki

    Subway Apathist of the Obsidian Order

  • Members
  • 9965 posts
  • Location:Australia
  • Steam:STEAM_0:1:1229809
  • Pillars of Eternity Silver Backer
  • Kickstarter Backer

It is incredibly simple, to the point of being simplistic, and is easily exploitable and creates combat incongruous to the theme. Not everyone liked every aspect of the IE games, and not everyone thinks that copying them verbatim is a good design policy for PoE and Obsidian more generally.


I did not say copy how Baldur's Gate did it. I said (at least) copy how Icewind Dale: Heart of Winter did it. People were complaining that they couldn't force enemies in BG1 or BG2 to attack certain characters - there's a video for them to demonstrate how to do it.

Edited by Sensuki, 02 December 2014 - 08:06 AM.


#24
Doppelschwert

Doppelschwert

    Sword Enthusiast of the Obsidian Order

  • Members
  • 954 posts
  • Location:Germany
  • Pillars of Eternity Silver Backer
  • Kickstarter Backer
  • Deadfire Silver Backer
  • Fig Backer
  • Black Isle Bastard!

 

- snip -

I agree with your criticism of #3, Doppelschwert. However, I reckon, Raz is right in his ambition of including some player agency into it all.

On #2, I disagree. Raz's solution there wouldn't make Disengagement pointless in any manner or form. In fact, it would work beautifully and would vary baddies in interesting ways, including "bosses" and strategic ambushes etc.

 

 

Ok, look at the following paragraph:

 

 

Further expansion of the concept would involve engagement limits nullifying each other in a plus minus system. In this case, a fighter with an engagement limit of 2 and a boss with the same engagement limit would tie each other up. Party members that aren't the fighter would be able to engage in and leave engagement freely similar to how it is presently. If the fighter wished to disengage, then 2 party members with an engagement limit of 1 would have to engage in melee in order to occupy the boss enough for your fighter to retreat (thought he still might die to things that aren't disengagement attacks).

 

How would I ever get any disengagement attack when I can simply cycle through my party to occupy a boss monster? Eveytime someone is short of dying, send another guy in and retreat the wounded guy. Heal the wounded guy and repeat.

Either you have enough engagement limit to occupy the boss making you immune to disengagement attacks or you don't which means the mechanic is only available for trash where tactics should be less involved anyway.

You would never get a disengagement attack if you wanted to while the enemy would get several because AI either wouldn't be able to plan around it or could easily be forced to never disengage.

If you want to weaken it so considerably then you might as well remove it altogether in order to make it intutive at least. If you want them in the game, then you shouldn't be able to evade them with such ease.



#25
Shevek

Shevek

    (8) Warlock

  • Members
  • 1164 posts
  • Pillars of Eternity Backer

Why would 10 goblins storm past your Fighter to your Wizard? AI targeting never did that in the Infinity Engine games and it does not do it in Pillars of Eternity either.

 

Here's how BG1 and BG2 Targeting clauses worked for Melee Enemies. It was very simple. 

 



Will record an IWD video tomorrow or something, IWD:HoW has much more robust targeting and target re-acquisition and if anything, should be the model for PE's system.

I think I see what you are saying. I just think that kinda seems like achieving stickiness by implementing what seems like bad AI. Maybe I am missing somthing and if so, I apologize.


  • Chilloutman likes this

#26
IndiraLightfoot

IndiraLightfoot

    Apex of the Obsidian Order

  • Members
  • 4102 posts
  • Location:Over the Hills and Far Away
  • Pillars of Eternity Silver Backer
  • Kickstarter Backer
  • Deadfire Silver Backer
  • Fig Backer
  • Black Isle Bastard!

Doppelschwert: In part, you misunderstand his idea. There will never be any cycling going on in any easy manner. Also, with such a system in this pretty quick RTwP-CRPG with a full party, all members doing their stuff (some may even be blocked out, knocked out or wiped out), good luck even trying to achieve such easy shifts. In fact, even pulling them off would be more or less fringe behaviour, worthy of an achievement, almost (perhaps on some boss, it could work). :)



#27
Sensuki

Sensuki

    Subway Apathist of the Obsidian Order

  • Members
  • 9965 posts
  • Location:Australia
  • Steam:STEAM_0:1:1229809
  • Pillars of Eternity Silver Backer
  • Kickstarter Backer

I think I see what you are saying. I just think that kinda seems like achieving stickiness by implementing what seems like bad AI. Maybe I am missing somthing and if so, I apologize.


No, not really. I need to record some videos of Icewind Dale:HoW, because the targeting in that game is really good. I'll do it tomorrow. If you make a mistake and your Mage is in front and is being targeted, you can just swap formations manually around and usually that will fix it.

I'll do some examples where I do it properly and never let the enemy target my squishes, and I'll do some where I deliberately make a mistake, and show how to correct it simply by moving units around a little bit.

Edited by Sensuki, 02 December 2014 - 09:57 AM.


#28
Doppelschwert

Doppelschwert

    Sword Enthusiast of the Obsidian Order

  • Members
  • 954 posts
  • Location:Germany
  • Pillars of Eternity Silver Backer
  • Kickstarter Backer
  • Deadfire Silver Backer
  • Fig Backer
  • Black Isle Bastard!

Doppelschwert: In part, you misunderstand his idea. There will never be any cycling going on in any easy manner. Also, with such a system in this pretty quick RTwP-CRPG with a full party, all members doing their stuff (some may even be blocked out, knocked out or wiped out), good luck even trying to achieve such easy shifts. In fact, even pulling them off would be more or less fringe behaviour, worthy of an achievement, almost (perhaps on some boss, it could work). :)

 

I neither see why cycling through wouldn't be an easy manner in fights with less enemies than party members nor which part of the idea I misunderstood. Care to elaborate whats preventing me from doing this when I go all out with a party of 4 melee characters on a single enemy like the ogre?



#29
ISC

ISC

    (3) Conjurer

  • Members
  • 181 posts
  • Location:Sweden

 

I think I see what you are saying. I just think that kinda seems like achieving stickiness by implementing what seems like bad AI. Maybe I am missing somthing and if so, I apologize.

 

 

The thing is, I think, that bad AI (and it can apparently be done better) produces less additional problems and problem-solving than engagement (which doesn't have better AI anyway?).



#30
serenityangel

serenityangel

    (1) Prestidigitator

  • Members
  • 14 posts
  • Steam:serenatyangel
  • Deadfire Silver Backer
  • Fig Backer

I made a similar post under a different thread here: http://forums.obsidi...olutions/page-4 but didn't include a poll or anything.  Shevek's idea for a poll might be a good one if the idea is to reforge the engagement system.  However...

 

After Bester's insightful comments (at least in current game state) and my experience, as I feared in my thread, I am concerned that it grows too complex and cumbersome.  When a system becomes too complex, it becomes a hindrance more than a feature.  I start to think more strongly that they should just probably scrap engagement and focus on better reacquisition targeting scripts. They could add opportunity strike as a lvl up trait maybe, but not inherent to the game engine itself for everyone.


  • Sif and ISC like this

#31
Sensuki

Sensuki

    Subway Apathist of the Obsidian Order

  • Members
  • 9965 posts
  • Location:Australia
  • Steam:STEAM_0:1:1229809
  • Pillars of Eternity Silver Backer
  • Kickstarter Backer
I've recorded about 4-5 videos about Engagement tonight but I still need to do a few more, as there are quite a few problems with it. Might have the thread up in 36 hours or so.

I'm not posting them publicly until they're all done and I figure out how to format the post, but if anyone's interested in watching a couple, feel free to send me a PM. Some of them are pretty funny.

Edited by Sensuki, 02 December 2014 - 10:16 AM.


#32
Kjaamor

Kjaamor

    (6) Magician

  • Members
  • 682 posts
  • Pillars of Eternity Backer

The thing is, I think, that bad AI (and it can apparently be done better) produces less additional problems and problem-solving than engagement (which doesn't have better AI anyway?).

 

It's true, but again the issue comes down to whether or not PoE is the place to try and improve things.

 

The current engagement system, combined with the current AI, might not play better than the IE games. The question is to whether Obsidian should be attempting to improve such mechanics to make them better than the IE games, and I believe that they should.

 

The secondary issue, which Bester raised, is that while the traditionally glass cannon classes act as glass pea-shooters this might not necessarily be the case. I would argue that bow-wielding naked Ciphers and naked sneak-attacking rouges with an arbalest or arquebus have effectively replaced the wizard in this area. If Obsidian take steps to nerf those two builds, which arguably they should, then they could probably afford to lose engagement and just have bad AI.



#33
Sensuki

Sensuki

    Subway Apathist of the Obsidian Order

  • Members
  • 9965 posts
  • Location:Australia
  • Steam:STEAM_0:1:1229809
  • Pillars of Eternity Silver Backer
  • Kickstarter Backer
The naked sneak attacking Rogue with an Arbalest isn't anywhere near as good now, does like half the damage. Much better off dual wielding fast weapons.

#34
Kjaamor

Kjaamor

    (6) Magician

  • Members
  • 682 posts
  • Pillars of Eternity Backer

Good, because that was ridiculous and more than a little unthematic.

 

While on the tangent:

 

Rangers: Change the "my animal dies, I die" mechanic to turn them into an at least somewhat ranged class.

 

Ciphers:

 

They should merge the classes of Ciphers and Wizards into a new class, based around some manner of regenerating spell points and reuseable spells, and call it the "Wizard".

 

...but sadly that isn't going to happen.



#35
serenityangel

serenityangel

    (1) Prestidigitator

  • Members
  • 14 posts
  • Steam:serenatyangel
  • Deadfire Silver Backer
  • Fig Backer

Maybe they should just change wizards to "evokers" and beef their damages.  Back on point though, nerf the back line enough and will be no need to go back there which just super reinforces Bester's point of engagement system, why?



#36
Shevek

Shevek

    (8) Warlock

  • Members
  • 1164 posts
  • Pillars of Eternity Backer

I've recorded about 4-5 videos about Engagement tonight but I still need to do a few more, as there are quite a few problems with it. Might have the thread up in 36 hours or so.

I'm not posting them publicly until they're all done and I figure out how to format the post, but if anyone's interested in watching a couple, feel free to send me a PM. Some of them are pretty funny.

 

I think Sensuki deserves a lot of credit for finding these abuses. His efforts will, one way or another, improve the game. I do believe that AI can be adjusted to close in on Player Characters with greater care, however. Essentially, when giving chase, enemies could change targets to nearest PC as they close to avoid being eviscerated by engagement attacks (maybe even having close range lunging attacks on certain enemies to aid this). After such a fix, I am sure that a similar exploit could still be executed via shuffling units about but that will require significant player input and would be no more of an exploit than the MANY ways to trivialize combat in the IE games. 
 
Basically, if the exploit is tedious to pull off, I say let the player do it. If its a simple thing to do (positioning and a couple mouse clicks), then it needs to be curbed.
 
I will say that if AI cannot be altered to ensure enemies give chase in a smarter manner (or maybe even outfit more enemies with switchable ranged attacks), then I will join Sensuki in saying engagement should go. There is no way you should be able to so easily set up casual ambushes. 
 
I just don't think that is a difficult issue to fix however. I really can't wait till Sensuki posts his videos and the devs, hopefully, respond.

  • Captain Shrek and Sif like this

#37
Captain Shrek

Captain Shrek

    (5) Thaumaturgist

  • Members
  • 580 posts
  • Pillars of Eternity Backer

Alright.

 

My real issue with the combat is not concerned with the individual elements. But rather the observation that combat as a whole feels VERY boring. 

 

Now, I have to admit that this is because the encounter design in the beta and the demo twitch was probably just demonstrative of the mechanics. This might mean that the actual encounters in the main game will be interesting enough to overshadow the boring and clumsy feeling of the mechanics themselves. I strongly believe that the most important aspect of a single player game is finally content. The mechanics are important, yeah, but with good storytelling and smart positioning of enemies in interesting situations can totally save the game. Deus ex does it that way at least. 

 

But, looking back at Obsidian's previous efforts in the  encounters area, I think I am justified in having my doubts. 



#38
Razsius

Razsius

    (3) Conjurer

  • Members
  • 164 posts
  • Pillars of Eternity Silver Backer
  • Kickstarter Backer

Apologies in advance for my slow responses as i'm a slow typer and I was knocked out cold for a while there.

 

 

 

If you call this a dissertation you're probably not a real doctor. shifty.gif

Now that I made the bad joke I came here for, I guess I have to answer to the topic in order not to get off topic. I'll probably come off as harsh, but I appreciate the effort you put into it.

 

- I agree on the visual stuff for the most part.

 

- What you describe in 2) with your disengagement limit is basically a form of aggro mechanic and for the most part completely nullifies the whole mechanic, making it rather pointless to begin with.

 

- What you describe in 3) seems rather pointless as well. Why is there a distinction between an ability that disengages and one that nullifies engagement at all? The only reason you want to nullify engagement is in order to disengage, so the ability may make you disengage as well. If it allows you to nullify engagement, its just a break free of jail card you can occasionally play in specific situations. The abilities that allow you to disengage at the moment have way more uses and it's not trivial to decide what to use them for as they can also be used for different tactics.

 

In conclusion, I think the changes you propose make the disengagement system not trigger for the most part while being heavily penalized even if it ocassionally does. If you want to go with this kind of setup, you're probably better off removing it alltogether.

 

My engagement limit section is indeed proposed in order to allow the player a little more freedom of movement to do things but it does not completely nullify the engagement system entirely.  Both you the player and enemies mobs do not necessarily have to have an engagement limit of only 1.  It might be a limit of 2 or 3 hell it might even be 4 but it does allow you to "tie up" and interact with a set of mobs.  The neat thing is it gave would give the engagement limit system a little more "meat" to it.  Things like Hold the Line might actually be a viable talent for your off tank (pardon the mmo expression) to pick up.  Being able to hold off 5 units in melee between a fighter in defender mode and an off tank with Hold the line between the pair might be a good example to providing a safe or safer melee environment for otherwise squishy characters.  Sometimes it might only hold off a total of 2 enemies it really depends.  It does however, give the player more options.  I might be alone here but talents being more useful and more player options is something I will always consider a good thing.

 

If you notice the 2 examples I gave involving null-engage were both hobble abilities.  Currently, hobble abilities function as a relatively minor defense debuff, something that can proc a sneak attack and a movement penalty that doesn't matter after engagement.  Null-engage would make hobble abilities movement penalty actually matter.  If there was a big nasty spider all over your wizard eating his face off hitting it with a Crippling Strike would do the equivalent of almost nothing.  I mean sure yea now it moves slower but the minute your wizard moved it would basically be over for him due to the incoming disengagement attack.  So what then is the functional purpose of a hobble?  Is it a debuff only useful during the alpha strike phase of combat?  Even I consider that a little unforgiving and i'm certainly no casual player.  As above, it would change some of the landscape of abilities and spells.  For example, it would make the Wizard movement speed spell much better without adding a disengagement break on the spell itself.

 

 

 

Frankly, engagement itself has got a lot of undue attention from the wider issues of the UI being a complete cluster****, and a tanking Fighter being absolutely mandatory at this point.

 

UI issues aside (and why I have The Visuals as my first section) the fighter issue was something I was worried about 3 months back.  For what it's worth I did have ideas on how to make monks feel a lot more "monkish" while being able to fill the fighters tanking role.  To be honest the classes in PoE really fail to impress me for the most part.  Obsidian could've taken the ball and ran with it but they didn't.  They delivered the same stuff i've seen time and again.  I would've posted said ideas but i didn't feel like they wouldn't restructure any of the classes 3 months ago and I certainly feel the same way if not more so now.

 

As for crackwise's idea it is sound but does not necessarily address what I believe Obsidian wanted to be addressed.

 

Bester:

 

Ya I hear ya man.  The Wizard changes (see nerfs) were about the last thing I ever wanted to see implemented in PoE.  Wizards now have the same mundane, boring spells that every other game gives them.  Different colored balls to do different colored damage.  Doesn't help that anything a Wizard can do a Druid can do better and then some.  But that's a fight for another time (that likely won't be won).

 

In regards to AI:

 

I originally had an AI section in that giant arse post of mine but considered it a bit too opinionated.  The intention of the thread was to show players what engagement did and what it could do as well as to get feedback in those regards.  The changes were proposed as an attempt to add robustness as well as allow it to be easily understood and (somewhat) manipulated by the player.  I divorced AI from the contents of the original post thereafter.  Truthfully, I think the AI utterly sucks in the backer beta.  I've played C rank games with less "sticky" enemies.  Thief enemies that would flat out ignore your tanks and take out your squishies and the like.  I'd like similar dynamic AI in the games I play but apparently AI systems need to be built from the ground up so you get a "tricycle" level of AI vs. the "bike" level you hope is in the next installment.  I am not particularly happy about the 15 years of "progress" that's been made in that regard.

 

 

 

Doppelschwert: In part, you misunderstand his idea. There will never be any cycling going on in any easy manner. Also, with such a system in this pretty quick RTwP-CRPG with a full party, all members doing their stuff (some may even be blocked out, knocked out or wiped out), good luck even trying to achieve such easy shifts. In fact, even pulling them off would be more or less fringe behaviour, worthy of an achievement, almost (perhaps on some boss, it could work). :)

 

I neither see why cycling through wouldn't be an easy manner in fights with less enemies than party members nor which part of the idea I misunderstood. Care to elaborate whats preventing me from doing this when I go all out with a party of 4 melee characters on a single enemy like the ogre?

 

 

Does the ogre have an engagement limit of 1?  Does he have other abilities he can kill you with?  Does he have friends?  Is it bad that you can use 4 melee characters to do a damn good job of "holding a line"?  I thought people wanted more reasons to use melee characters ;).

 

As a note here is what Obsidian has currently prioritized in regards to my post:

 

The Visuals

 

1) The Zone of Control - Needed clarity (Sensuki video special)

2) Engagement “Arrows” - Being addressed by Obsidian

3) Disengagement Animation - Being addressed by Obsidian

 

The Mechanics

 

1) The Zone of Control - Needed fixes (Sensuki video special)

4) The Disengagement Attack - Needed fixes (Sensuki video special)

5) Targeting clauses - unaddressed so far

 

As a side note guys it looks like Sen is working on some (rather hilarious) but really exploitative videos in regards to melee disengagement (see engagement) attacks that should outline some of the reasons i've taken the time to address some of the things that I consider issues with the current system.


Edited by Razsius, 02 December 2014 - 12:16 PM.

  • IndiraLightfoot and Sif like this

#39
Shevek

Shevek

    (8) Warlock

  • Members
  • 1164 posts
  • Pillars of Eternity Backer

Alright.

 

My real issue with the combat is not concerned with the individual elements. But rather the observation that combat as a whole feels VERY boring. 

 

Now, I have to admit that this is because the encounter design in the beta and the demo twitch was probably just demonstrative of the mechanics. This might mean that the actual encounters in the main game will be interesting enough to overshadow the boring and clumsy feeling of the mechanics themselves. I strongly believe that the most important aspect of a single player game is finally content. The mechanics are important, yeah, but with good storytelling and smart positioning of enemies in interesting situations can totally save the game. Deus ex does it that way at least. 

 

But, looking back at Obsidian's previous efforts in the  encounters area, I think I am justified in having my doubts. 

 

@Captain Shrek 
 
I am of the opinion that encounters for a game like this should be tiered. Tier 1 is a standard trash encounter that may widdle down party resources (Health) but requires little more than SelectAll+Attack or similar such tactics with very minimal skill use if the party is built for passive play. Tier 2 may be either a large group of weak enemies or a smaller group of skilled enemies. Here some skill use may be needed against troublesome enemies or when the RNG just doesn't roll your way. These encounters would more dramatically impact party resources (Health + ability use). A Tier 3 encounter is a tough encounter. This could happen dynamically as a wandering group joins a static group mid combat or a set piece encounter at the end of a dungeon, etc. This is where you may consider resting prior to fighting (if possible) and painstakingly using a set of abilities, etc.
 
To make these encounters interesting, they really need to differentiate enemies. I think the engagement system affords an opportunity to help with that. Look at the design of the Monk class. With abilities like Flaggelants's Path, Rooting Pain, Stunning Blow and so on, that class is designed to disengage itself rather readily. What if select ENEMIES were designed this way. What if to differentiate encounters of various tiers select skilled enemies could be designed to have similar abilities. What if the elder wolves you meet in Dyrwood know down the tank and went for the lowest armored character in view? 
 
The key, I think, is to leave Tier 1 encounters nice and easy but pepper more Tier 2 encounters with a few enemies as described above more liberally throughout the map.
 
To make combat more fun in general, I think the disengagement abilities need to be a bit easier to pull off (instant, more effective/numerous, etc) and maybe look at adding a cooldown or some such to disengagement attacks themselves. These are not particularly difficult fixes, I think, but the devs would know better.

Edited by Shevek, 02 December 2014 - 12:12 PM.


#40
Captain Shrek

Captain Shrek

    (5) Thaumaturgist

  • Members
  • 580 posts
  • Pillars of Eternity Backer

Hmm.

 

In your tier system, I really do not see the need for the tier 1 to exist. Whittling down resources is bound to happen in a game where the resource economy is not broken.  In that case the encounters just become a nuisance that need to  be eliminated to proceed. 

 

Also, I measure the quality of an encounter by the presentation; take the IWD2 encounter near the Shaengarn bridge. There are two things there:

 

A totally infuriating trash mob and a conversation with the enemy commander regarding the bridge quest. I would ask you this: if the trash mob were NOT there, would you mind? I loved the dialogue and the encounter after that (Stop the trolls/pgres/giants before the destroy the bridge).  It could have been done better, but it was already pretty good. But you know what? The entire army of generic villains I had to face right after that battle nearby (or before depending on how you approach the bridge) left such a bad taste in my mouth that overall I marked down the game as having negative score.

 

That is the real problem. 

 

I think people respond better to encounter that make sense and have tactical scenarios, like in this case do it within a specific amount of time. Other such scenarios could involve using environmental resources to engage enemies or even holding control of certain areas etc. Instead, most RP games just throw trash at you. 

 

Simply having hard enemies is just not enough. The context in which they are presented creates the real "feel" of the combat. The rest is simply annoyance. 


  • Nakia and ISC like this




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users