Update #44: The Rules of (Melee) Engagement

[ddillon]

Somehow I missed this before:

We're not making a single-character MMO.  We're making a party-based RPG.  We're making it to appeal to the general tastes of audiences that have played D&D-based tactical party RPGs in the past.  Yes, when you play a single character, having that single character be locked down is annoying because your only character is prevented from moving.  You have a whole party to use.

Mind-boggling post...

 

Haven't played Dragon Age: Origins? Not even aware of it?

 

Go play DA:O now.

 

If it's not your cup of tea, fine, but playing on dislike of MMOs does *not* work as a dismissal of threat-based mechanics for a party-based game.

Edited March 4, 2013 by ddillon

[Osvir]

Great update, great video. Loving the Engagement mechanic

Thoughts on AI: IF the encounter is intelligent enough would they place themselves in the Engagement circle (which seems to be an unfavorable position for them) or is there another way to gain favor above the Player? Can the AI Rogue throw a "snare" (or something to stun or impair the Fighter) at the Fighter, making it easier for the Rogue to simply skip past him/her and engage the Wizard?

If you were the AI Rogue, would you jump straight into the engagement, hoping to get past the Player Fighter & then score a critical hit on the Wizard, or would you wait up by the doorway and try to sneak past the Fighter as he engages? Don't know if my point is coming across entirely, but would it be a wise choice for the AI Rogue to engage the Fighter in the doorway?

If intelligent enough of course, a mindless zombie or a dumb Fighter might not take positioning into account and just charge head on, but the more sophisticated classes. Perhaps the AI Cipher would Charm/Control the Fighter to fight for him, putting the Player Wizard in a very bad position.

How can the AI play against the Player and not into the Player's hand?

Player: "Yes! I got him now in my positioning! If he engages I'll win"
AI: "I think not" *ZAP*
Player: "Oh sh...!"

[Tamerlane]

 

Somehow I missed this before:

We're not making a single-character MMO.  We're making a party-based RPG.  We're making it to appeal to the general tastes of audiences that have played D&D-based tactical party RPGs in the past.  Yes, when you play a single character, having that single character be locked down is annoying because your only character is prevented from moving.  You have a whole party to use.

Mind-boggling post...

 

Haven't played Dragon Age: Origins? Not even aware of it?

 

Go play DA:O now.

 

If it's not your cup of tea, fine, but playing on dislike of MMOs does *not* work as a dismissal of threat-based mechanics for a party-based game.

But... that quote has nothing to do with threat mechanics?

[Monte Carlo]

The DA:O 'Threat' / Aggro mechanics kinda sucked... and I'm not an instinctive DA:O hater and kind of enjoyed the combat.

[rjshae]

The DA:O 'Threat' / Aggro mechanics kinda sucked... and I'm not an instinctive DA:O hater and kind of enjoyed the combat.

 

I didn't mind the Threat mechanics; it was the 'run around and skirmish' nature of combat that didn't work for me. There just wasn't much of a meaningful tactical flavor to it.

[IndiraLightfoot]

The DA:O 'Threat' / Aggro mechanics kinda sucked... and I'm not an instinctive DA:O hater and kind of enjoyed the combat.

 

Agreed, at least if we take a broad definition of threat here and simply conclude that combat in DAO wasn't very threatening at all. Sure, dragons were more difficult than in Skyrim, but it was a long clickfest for me. Recall a certain castle where hoardes of undead poured out of? The one where Rosemary's baby had grown up and turned into a Twilight-star-like demon dancing oedipally and prematurely on his ailing father's grave?

That misty encounter should have been epic, but alas no: it was so narrow that it turned into carpal tunnel instead.

[anek]

Hello All,

 

I got a related question on the topic of Melee Characters, even it not exactly related to the Melee Engagement Mechanic.

 

In most, if not all, CRPGs you end up with a strange power divide between the Melee Classes and the Mage Classes where the power starts in favour of the Melee Classes and then gradually shifts to the Mage Classes. Some of this might have to do with the fact that a lot of CRPGs are influenced by the DnD and other d20 games. Another part might simply be that most CRPGs try to keep it somewhat realistic and that in turn makes it easier to come up with jaw-dropping spells that would be cool to have but to compensate the mages need to be "hard to master" -- or simply hard pressed to survive long enough to learn the spells.

 

I was wondering what's Obsidian's oppinion on this for Project Eternity.

Are we going to see the classic power evolution where a mage starts out as relativly weak and then becomes powerfull were as the fighter goes the other way around?

Or are we going to see some mechanics put in place so that both classes evolve throughout the game without much, if any, power gap between them?

 

Sorry if this has been asked before -- but it hasn't been adressed in the QnA's and at least not in this thread so I figured I should go ahead and post it.

 

 

Thanks for the updates, always fun to check out how the game is progressing!

 

-TSD

 

 

Sawyer answered here: http://www.formspring.me/JESawyer/q/432792143375920971

[TheSadDragon]

@anek:

 

Thank you sir! Great to hear they are not looking to keep the old power dynamic

Finally I wont feel like im gimping myself just because i like em shiney swords :D

 

-TSD

[ddillon]

 

 

snip

But... that quote has nothing to do with threat mechanics?

The quoted post is Sawyer's attempt to sidestep n0mDePlume's posts about how threat-based mechanics accomplish the same aim in a fashion more appropriate to the genre. Unless I misread that string of posts.

 

---

 

@everyone: Have you played an IE or Aurora game lately?! How exactly is the combat any more 'tactical' than DA:O? Do you remember just how stupid the enemies are in those games? How boring combat is in general? DA:O *requires* tactical play and player involvement to win, and while perhaps not 100% 'realistic', enemies determine targets in a fashion much more believable than "saw u first", "last attacked by", "get the mage", etc (basic IE scripting, in other words).

 

(Yes, we all know that Forcefield is broken... I know someone wants to bring it up, but it isn't relevant to the mechanics in general).

 

Do you think Bioware looked to threat-based mechanics and cooldowns because MMOs are popular? Don't be foolish. Theat-based mechanics and cooldowns and mana address the same issues as engagement and grimoires. The very thing Sawyer is crying about in the video doesn't happen nearly as often in DA:O because (as I mentioned before) threat already solves this problem.

 

---

 

@IndiraLightfoot: Did you dislike threat-based mechanics and cooldowns, or was your real issue having so many activated abilities and being forced to take a much more active role in directing combat (in other words, pausing & clicking a lot, the 'clickfest' that you mentioned)? Nothing wrong with that if so, not everyone enjoys combat, etc, but recognize which mechanics are actually vexing you.

 

---

 

 

I'm done beating a dead horse for now. And I'm okay with PE not having threat, cooldown, etc. I've accepted that PE will likely be a step backwards mechanically; I'm in it for the world and story at this point. Anyway, be back Thurs or Fri. Looking forward to the next update.

[Lephys]

 

I'm done beating a dead horse for now. And I'm okay with PE not having threat, cooldown, etc. I've accepted that PE will likely be a step backwards mechanically; I'm in it for the world and story at this point. Anyway, be back Thurs or Fri. Looking forward to the next update.

I'm... not entirely sure P:E will be 100% devoid of anything resembling "threat, cooldown, etc."

 

I'm not sure what you're suggesting, either, with this whole IE/DA:O comparison. IE games were round-based, and P:E is real-time (wasn't DA:O real-time?), not to mention a variety of other changes and improvements in P:E as compared to the IE games of old. Besides... I don't know of many details, thus far, on how the AI will handle target acquisition and such. Why this "clearly if it's not copying DA:O, then there must be absolutely no reasonable means by which foes acquire targets, u_u" conclusion? Why the arbitrary dichotomy?

 

And, for what it's worth, Sawyer wasn't sidestepping. He was answering the question, using specifically MMO threat/engagement design as an example of the direction they aren't even facing with their mechanics. That doesn't mean "Obviously you meant you want us to design an MMO system, and we aren't doing that."

 

Lastly... DEAR LORD, THAT HORSE HAS A PTERODACTYL HEAD!!!!!

Edited March 5, 2013 by Lephys

[Monte Carlo]

 

And I'm okay with PE not having threat, cooldown, etc. I've accepted that PE will likely be a step backwards mechanically

 

Sorry, had to chuckle. This is a bit point. not. found.

 

PE is meant to be old-school, steeped in the values and feel of what went before. Most of us here think newer gaming models are a 'step backwards.' What we see with PE is a return to what's actually important - depth, tactics, gameplay, fun.

[IndiraLightfoot]

 

@IndiraLightfoot: Did you dislike threat-based mechanics and cooldowns, or was your real issue having so many activated abilities and being forced to take a much more active role in directing combat (in other words, pausing & clicking a lot, the 'clickfest' that you mentioned)? Nothing wrong with that if so, not everyone enjoys combat, etc, but recognize which mechanics are actually vexing you.

Oh nos, I certainly can appreciate threat-based mechanics and cooldowns. I like to play those games too. Despite its artistic shortcomings, the linearity and that constant online requirement, I actually enjoyed the combat part of Diablo 3, especially inferno, for instance. My main concern was that I never had to pause ever, and that is with an entire party (in Diablo 3, I often went solo). It was so easy and repetitive (including the poor selection of critters). I just clicked and clicked and clicked and hey presto, I had passed another encounter in the game. It's like Diablo 3's normal mode in the beginning, but without some of the encounter variety. You can still faceroll that game up till the skellie king, for real!

 

What I'm hoping for here is a strategic party-based rpg that actually is challenging from the get-go, a game where pause really means something else than afk and toilet break.

 

P.S. For what it's worth, I think DAO was a really nice game - I loved the epic scope of it and some of the story was great. But it wasn't great enough for me to go and buy the sequel DAO2 - it appeared to be some kind of machinima-game, so I just skipped it.

[Valorian]

Hullo!

While we're waiting for the update, I'll make an update of my own.

 

Mr. mechanics developer, I've been thinking... Engagement and "tanking" go hand in hand, so tanking is the topic of my update.

 

When you put someone at the doorway to block access and to engage enemies, you want that character to be a good tank. In PE that means high armor and high deflection. Since armor doesn't affect your ability to dodge at all, the best tank will be someone in heavy armor and with high defense (deflection).

 

I can see their flow of ideas: hmm, heavy armor is always the best option for tanking.. what if we added a huge penalty for some weapon vs armor combination so that heavy armor isn't always the best option for tanking?

 

Heavy/medium armor reduces attack speed, but it's irrelevant for tanking purposes, because you have 5 more characters attacking while you engage and tank.

Mr. Sawyer said he wants to support the possibility of making characters with high evasion so that's one of the reasons he added misses on the scale.

 

Long story short, why wouldn't they add a small penalty to deflection defense for medium and heavy armor on top of the attack speed reduction? Just a tiny penalty of like -4/-8 (we're talking about a d100), if nothing else, for style reasons. It also makes sense, if your attack speed is slower - because armor hinders your movements, you'd also have more trouble dodging.

 

--> If an enemy is hitting real hard, but isn't very accurate... you could consider going light armor instead of heavy, to avoid crits at least. On the other hand, if an enemy has high accuracy but low damage, you'd go heavy armor.

It accomplishes what they wanted to do with the 50% penalty, but in an elegant way.

 

Yes, you don't like the "redundancy" part, but I'd argue it's much less redundant than the 50% thing.

 

 

 

 

[Lephys]

That actually makes perfect sense (although, looking back, it seems a bit weird, in a way, that evasion and armor's force dispersal are both lumped into the same defense type... maybe we should split that between active -- parrying and dodging -- and passive -- static armor deflection?). Your heavy-armor tank is only going to be effective against physical melee attacks (Maybe unless it's against 3 stone golems with giant mace-like hands that deal loads of crushing damage? Guess we're still waiting on all the details for armor and such), so it's already not the ultimate defense in the universe or anything. For example, you probably wouldn't want to block a doorway against 3 enemy mages with a heavy-armor tank, because that deflection really isn't going to help you much against thunderbolts and lightning, very very frightening... 8P

 

Also, what if there were 2 stages of engagement? Partial, and full. So, for example, a Fighter in normal (non-Defender) mode can only offensively engage and "block" 1 enemy. BUT, maybe he can hold his own against 2 more enemies (as in they don't get any "flanking" bonuses to attacks against an already-engaged target), but they can disengage at any time without provoking an attack? But the original enemy could not. The fighter would be in full engagement with the original target, and in partial engagement with the other 2 attackers. Maybe Defender mode allows him to fully engage 3 targets (instead of 1) at the cost of his offensive capabilities (already a penalty in the current system)?

 

Just a thought.

[Sensuki]

There's no need to overcomplicate things, the theory works fine as it is.

 

If you're going to try and move past a fighter in close proximity, and he chooses to try and stop you, there's most likely gonna be repercussions.

 

The only problem I can see is with the movement cancel and when that occurs. If the coding isn't right it could be a bit buggy or jerky.

 

I *think* this is how it's going to work

 

Let's say the Fighter in your way, you have 10 metres of in game space to move through a corridor. Fighter's engagement area is 5 metres of ingame space.

 

You issue a move action to behind the fighter with a character.

 

Some considerations:
a) Pathfinding - if there is a gap between the moveable area and the fighter's engagement area - will your character automatically try and avoid the engagement area or will it just walk the most direct route through the corridor.

b) The Fighter's AI: Whether he moves to attack the nearest character.

 

If the AI fighter essentially "clicks on you" or engages the character that you issue the move command to, once the character enters the fighter's engagement area, that character's movement action is cancelled and as Sawyer said, there's some kind of connect between selection circles and an icon on the character portrait or something to signal that that character is now engaged in melee combat.

 

If you issue another move command to exit the engagement area, the fighter get's an attack of opportunity with a bonus against that character, and if it hit's it stops the movement action of that character and some kind of animation is triggered.

 

A question I can think of regarding that:

 

If you issue another move command shortly after, once the animation has finished, and the fighter is still engaging you, does he get another free attack ?

 

You could potentially use a rogue or a barbarian, who appear to be 'skilled' at breaking engagement with melee characters to bait a character away from a corridor so the rest of your party can move through, as long as you don't move another character nearby while doing so so that their target priority changes. Totally depends on encounter design though.

Edited March 6, 2013 by Sensuki

[Aedelric]

Great update guys, as a person who always plays as a melee character I like how you have approached the shortcomings of close combat.

[Lephys]

Great update guys, as a person who always plays as a melee character I like how you have approached the shortcomings of close combat.

 

 

... I see what you did there... 

[khango]

[Thumister]

Great update!  I'm glad you're addressing the engagement issue: it was hugely annoying in the Infinity Engine games.

[robfang]

The rules of (melee) engagement makes positioning a hard constraint in combat. That constraint, if not handled very carefully could make real-time combat too chaotic due to the unexpected movement from AI controlled characters. I don't think even pause and release type interaction would solve this problem which would lead to constant, annoying pausing. A natural way to handle this problem is to have turn based combat like in Temple of Elemental Evil computer game. 

 

Turn based combat systems are perhaps the most satisfying combat system for the tactical play. It removes annoying movement rate considerations in real time and the more annoying constant shift of attention required for the control of multiple units. However, turn based combats could be quite a nuisance if the combat itself is trivial. It is agonizing to have the game paused because you encountered a couple of skeletons with your 18th level party. One way to solve this problem is to make every encounter meaningful and difficult. Another way to handle it is to have the option of playing in real time with a not-so-bad AI. While the first option is always preferable, it could be difficult to have it so. In a semi-open world like Baldur's Gate series, some of the encounters are bound to be trivial. Thus, the option to go real time seems to be necessary in order to avoid frustration of trivial turn-based combat. 

 

The option to go real time could even be a measure of the party power from the player perspective. It would be like going auto-combat in a Heroes of Might and Magic game where some people might find satisfying. 

 

TLDR: Fully turn based combat with an option to go real time seems to be the best implementation for the tactical combat presented by the rules of (melee) engagement. 

[khango]

The rules of (melee) engagement makes positioning a hard constraint in combat. That constraint, if not handled very carefully could make real-time combat too chaotic due to the unexpected movement from AI controlled characters. I don't think even pause and release type interaction would solve this problem which would lead to constant, annoying pausing. A natural way to handle this problem is to have turn based combat like in Temple of Elemental Evil computer game. 

 

Turn based combat systems are perhaps the most satisfying combat system for the tactical play. It removes annoying movement rate considerations in real time and the more annoying constant shift of attention required for the control of multiple units. However, turn based combats could be quite a nuisance if the combat itself is trivial. It is agonizing to have the game paused because you encountered a couple of skeletons with your 18th level party. One way to solve this problem is to make every encounter meaningful and difficult. Another way to handle it is to have the option of playing in real time with a not-so-bad AI. While the first option is always preferable, it could be difficult to have it so. In a semi-open world like Baldur's Gate series, some of the encounters are bound to be trivial. Thus, the option to go real time seems to be necessary in order to avoid frustration of trivial turn-based combat. 

 

The option to go real time could even be a measure of the party power from the player perspective. It would be like going auto-combat in a Heroes of Might and Magic game where some people might find satisfying. 

 

TLDR: Fully turn based combat with an option to go real time seems to be the best implementation for the tactical combat presented by the rules of (melee) engagement. 

 

Turn based combat sucks. It's nothing like actually fighting. When was the last time you saw somebody say "Oh, it is not my turn. Please hit me sir." ? I can tell you the answer: never. Turn based  combat can indeed be somewhat strategic or tactical or whatnot. But it's not combat. It's a stupid little game of silliness. RTwP helps make things more realistic and more truly tactical/strategic. Combat SHOULD be chaotic...

[robfang]

Turn based combat sucks. 

I believe you have not played ToEE computer game. Otherwise you would not be as eager to say that turn based combat sucks.

 

It's nothing like actually fighting. When was the last time you saw somebody say "Oh, it is not my turn. Please hit me sir." ? I can tell you the answer: never. 

I don't think such arguments serve any purpose. No one says "alright sir I will wait until you decide a meaningful course of action" either.

 

Turn based  combat can indeed be somewhat strategic or tactical or whatnot. But it's not combat. It's a stupid little game of silliness. RTwP helps make things more realistic and more truly tactical/strategic. Combat SHOULD be chaotic...

In an RPG, chaos of the combat is already modeled by the randomness of the dice rolls. We are not talking about a single player FPS, or an action game where you control a single character. You are supposed to control 6 characters simultaneously in the combat. Controlling them in real-time is impossible for a normal person. That is why RTwP was invented. In the old Infinity Engine games where RTwP was used, there was not a rule of melee engagement as strict as in 3rd edition D&D games. Thus, RTwP would not interfere with the gameplay too much. On the other hand, the games where 3rd edition rules were used and had RTwP, like NWN, the real-time movement from AI controlled characters was only annoying. In order to control everything you had to pause and pause again after each sword swing and spell cast. If you had not, you get random attacks of opportunities and had your mage in clusters of monsters. Those kind of things happened only with 3 characters to control. Now think about 6 characters and melee engagement rules.

 

TLDR: Turning the game into a micro-control hell or pause-fest would only destroy the tactical combat without any additional benefits. 

[Lephys]

Turn based combat sucks. It's nothing like actually fighting. When was the last time you saw somebody say "Oh, it is not my turn. Please hit me sir."?

You're absolutely correct. Because, when it's not your turn, your character doesn't dodge or parry or counter. Wait.. they do? Well crap.

 

I don't think you comprehend the concept of abstraction in turn-based combat. When you're controlling multiple party members, real-time combat actually detriments you (which is why we invented a little thing called Command Queuing During Pause), because you're one person, trying to make multiple people's real-time decisions all at once. Therefore, it is necessary to either abstract that by allowing one action/phase to occur at a time, OR to simply automate all the characters save for one, completely defeating the purpose of that multi-character control and strategy in the first place.

 

As robfang said, the pausing in real-time combat and the turns of turn-based combat serve the same function. If you want a game with neither of those, then have fun with your arbitrary chaos. *shrug*

[Osvir]

I've had some thoughts on engagement and, specifically, the choice of standing a bit further away from the doorway. Now this might be a great tactical decision for 1, maybe 2 enemies coming at you, but when you're engaging several enemies (a mob of units) at the same time, standing away from the choke point might be a poor decision.

Here's some examples and concepts I thought of:

This is all good, the enemy Goblin will face off with the Fighter right away and he will be "locked" with the Fighter.

Next, a poor example:

The first Goblin gets "locked" in combat with the Fighter and thus none of the others can get in good position. This is bad, knowing that there's open ground both to the left and right of the Fighter, which should allow for more Goblins to approach.

Something like this should occur:

The Goblins gets past and now you've got problems (Okay, perhaps not all, probably 1 or 2 dies on the way but I'm trying to make a point)

That's why, I hope, you will be able to put a "plug" in that choke point and this will occur instead:


Well, that's all I wanted to convey, I hope my point comes across.

[Sensuki]

In light of that I hope there are knockback abilities available to prevent abuse of the mechanic in doorways.