The Rambling Quest for a “Good” Boss Encounter in Grim Dawn

Grim Dawn Ideas and Feedback
1

Download this feedback in PDF Form.

Introduction

Initially starting as a discord-based discussion on ‘wackiness’ in video games, this is a megapost of feedback on the simplistic-seriousness of Grim Dawn boss design, with the intent of communicating the point that more complex encounters can result in greater “shock and awe” for players, which all in all make them more engrossed in the game as a whole. While Grim Dawn largely tries to stay very down-to-earth with what players are exposed to, I feel like its bosses—especially those which are optional—could do with a substantially greater highlighting in terms of player involvement or attentiveness. Grim Dawn does little in the ‘wow-factor’ department outside of a boss’s initial introduction, such as seeing the Loghorrean or Theodin Marcell for the first time.

Before we get into the fine details of my shower-thoughts on the matter, there are a few things I feel are important to admit, first and foremost being that I don’t expect anything to change in existing Grim Dawn content as a result of writing this feedback. This feedback is not a suggestion to go back and redo everything, but is instead meant as something to reflect on for future development projects. Another thing worth admitting is that, like all feedback, this is purely opinion and speculation (albeit with some concrete research to back up some claims here and there), namely my opinion and my speculation. Other users likely have different takes on the matter, and if so I invite them to weigh in. As a final admittance, from a technical standpoint, I’m going to incorrectly use the term ‘AI’ throughout this feedback. What I’ll be discussing is much less to do with anything relating to AI programming, and much more to do with what players think an enemy is thinking by virtue of what’s happening in-game. With that out of the way, let’s get into things.

Suggested Reading/Viewing

I’ve been a hobbyist gamedev for a little over a year now, working on a small project in my spare time. Over the course of that year, I’ve read and watched hundreds of gaming-theoretic resources to better understand why developers do what they do. Relevant to this conversation on Boss AI in an ARPG, here are my recommendations on the matter; I’ll be making reference to some things made throughout the following resources:

Immersion and identity in video games, Yaman Terzioglu, 2015. This paper serves as a great background to the rest of our discussion if you have to ask “Why bother?” with respect to the purpose of this feedback. Insofar as we’re concerned, nothing past page 16 is really relative to us, but the case studies from page 17-on serve as good evidence supporting the first half of the paper, if you’re so inclined to read on.

What Makes Good AI? | Game Maker’s Toolkit, Mark Brown, 2017. An excellent video on all sorts of AI systems in all sorts of video game genres. The arguments I’m going to make in my feedback here rely heavily on concepts introduced toward the beginning and end of this video, but everything in between is a fascinating learning experience too. It also introduces our third and final recommended resource…

The Illusion of Intelligence: The Integration of AI and Level Design in Halo, Chris Butcher; Jaime Griesmer, 2002. In a PowerPoint presentation at GDC 2002, Bungie game developers lay out design decisions they made for the first Halo game pertinent toward engaging, ‘intelligent’ AI systems and how those systems interact with and influence their environment design. ‘Pages’ 6 through 8, and their accompanying notes, serve as a sufficient introduction to what I’ll discuss ahead.

Anyways, the feedback itself…

Grim Dawn bosses are boring. They run at you and slap you, or they summon things to run at you and slap you. And that’s essentially the full extent of their ability to exist. Bosses, while they may look unique, rarely feel so. To some extent Grim Dawn has done this by design, wanting boss fights that were not world-shattering and were instead more down-to-earth foes of political importance. I would argue that both could be accomplished via better mechanics, but the phrase ‘game mechanics’ is thrown around a lot and I’ve only rarely seen it well-defined. So before I go into suggestions of how to make ‘the perfect ARPG boss,’ I’d first like to set up a schema for player-boss interaction in the first place.

I suggest that, insofar as we can discuss game mechanics for an ARPG (or even just an RPG) boss, there are two levels of distinction to consider in terms of player involvement with those mechanics, and one preliminary definition worth identifying:

  1. Pacing of an encounter. A measure of how much a player is capable of ending an encounter in such a fashion that will not result in an unfavorable ending. In simple ARPG terms, how actively can a character attack an enemy at any given time without dying for having done so at that time? This definition need not apply strictly to raw-spreadsheet-dps/tankiness metrics, however; instead, suppose the question instead is “how soon can I jump over Bowser, hit that axe, and drop the King of the Koopa’s into a fiery demise?” A big part of (A)RPGs is in having and using some degree of well-earned power; how different encounters modulate your ability to do so is the topic of conversation insofar as we’re concerned here.
  2. One-way mechanics. An entity does something that potentially changes the state of the game, but there is not any further interaction. Example: an attack is thrown out at the player, and if it hits the player, the player takes damage. Provided the attack doesn’t kill the player, which would result in an unfavorable ending to the encounter, this is the full extent of the mechanic.
  3. Two-way mechanics. An entity does something that its opponent can interact with by choice, and choosing to or not to interact with the event has differing results on the game state. Example: a boss summons a pet that attacks the player, but if the pet is killed, the boss gets a damage mitigation buff; players can either try to endure the boss+pet combo and focus the boss down, or kill the pet but deal less dps against the boss afterward. Note that the player taking more damage due to the addition of the pet is not the mechanic at hand—likely that additional damage is dealt via a one-way mechanic from the pet. The mechanic at hand is singularly the choice of leaving the pet alive or not, with either option having a commensurate effect.
    [/ol]A YouTuber under the alias of Huizui makes ‘Mechanics Explained’ videos for various Path of Exile encounters. I don’t particularly enjoy this nomenclature because by and large, many of these videos feature boss fights that only feature “one-way” mechanics, or mechanics that I don’t really find all that mechanical; they are effects without a cause, so to speak. An example of such an encounter would be the Lunaris/Solaris fight, in which there are two bosses that tag each other in and out of fighting the player. But that’s really it. Players do not have any interactivity with the swapping of the bosses, and only really need to dodge (or power through) individual attacks, much like in Grim Dawn. Alternatively, consider Huizui’s video on the Doedre the Vile fight; as the battle draws on, the arena is flooded with various debuffing pools/bubbles to hamper the player’s efforts to survive. The player can actively choose when they want to ‘clean’ the arena by cycling it to a different set of debuffs while granting a buff to the boss. This is an excellent “two-way” mechanic that gives players total control over the state of the boss fight while simultaneously keeping things very dangerous. As Mark Brown points out in his Game Maker’s Toolkit video, the purpose of AI encounters is to present interesting gameplay. It isn’t necessarily to crush the player’s hopes and dreams, but simply to keep the player on their toes and engaged with what’s going on. Enabling players to interact with such an AI goes a long way toward fueling that engagement and alertness.

    In a discussion comparing Path of Exile and Grim Dawn, a friend impressed upon me the concept of ‘spreadsheeting over content.’ He argued that that was the extent of the scope of Grim Dawn gameplay, in which you reach a certain threshold of numbers in various areas and then you proceed to wade into combat until you inevitably win. And he’s right. This is not the case with Path of Exile, in which seemingly no threshold of numbers is attainable that enables you to seamlessly pierce through the game’s content. My friend made the argument that this inherently made PoE’s fights ‘better’ than GD’s, and I can’t say I agree…completely, at least. While, yes, PoE’s fights do involve greater player interaction, there is rarely any ‘choice’ involved with them, only the illusion thereof if you intend to emerge from such fights victoriously. Ignoring the mechanics of the aforementioned Doedre fight, for instance, will absolutely kill you. I personally don’t find this to be good game design. I don’t think anything should be forced down a player’s throat, or that players should ever need to do anything in a certain way. Ideally, I’d like to see both games reach a middle ground between each others’ extremes, where mechanics could exist to make the fight easier, but stats could still enable players to persevere in ignorance or indifference to such mechanics.

    In my now-ending sabbatical from Grim Dawn, I touched upon various other RPGs and MMOs…and some MMORPGs. One of these was Warframe, which by and large has not had a history for interactive or optional game mechanics. Warframe, in fact, is plagued by one-way mechanics that I refer to as “a loss of interaction.” Grim Dawn has some of these too, as does Path of Exile. These are mechanics that, instead of creating interaction between a boss and a player, the player is forcibly broken away from the flow of the game. This most commonly happens in Warframe with invulnerability phases, such as those belonging to Captain Vor or Lech Kril (in fact, Lech Kril spends most of his fight invulnerable, which is awful). Preventing a player from even engaging a boss at any period of time for any length of time is a massive turn-off for me. Grim Dawn’s Fabius/A.F.F. Bourbon does this with Blade Barrier, Cronley/Bourbon Clones do it (to a lesser extent) with an amp’d-up Maiven’s, and I’d argue that even the Mad Queen’s enrage is an example of this, because while not impossible to fight against, does make “manning up” somewhat ill-advised. I do bring Warframe up, however, because it does feature an end-game fight with a perfect balance between ‘spreadsheeting over enemies’ and in using mechanics to better one’s performance: The Eidolon Hydrolyst. Throughout the fight, the Hydrolyst will spawn giant orbs that create thunderstorms around the area of combat; these thunderstorms will, in turn, warp the ground into a swampy mess that inflicts heavy damage against players standing inside. The ‘giant orbs’ initially responsible for all of this can be focused down and destroyed as they’re spawned; however, because the Hydrolyst is constantly on the move (in a massive landscape) throughout the fight, advanced players may note that it will often arrive at safer lands untouched by its thunderstorms/swamps, and so may opt to look for continually better positioning as the fight drags on, using sufficient stat configurations to ensure they can reliably get from a dangerous point-A to a safer point-B. That was a mouthful, but in practice in results in a two-way gameplay mechanic that can be completely ignored either by player skill or stat configuration (or by playing Rhino, but if everyone takes Rhino to a Hydrolyst fight, you’re in for a bad time for other reasons).

    As good as such two-way gameplay interactions are, I would not argue that they are uniformly essential for any fight to be a good one. Of the three games I’ve mentioned so far in this feedback, all three of my favorite fights strictly have one-way mechanics—Theodin Marcell in Grim Dawn, Innocencein Path of Exile, and Ambulasin Warframe (though Ambulas has interactivity, the choice of ignoring this interactivity is mission-failure, so I wouldn’t consider it to be an actual choice at all). Grim Dawn has also had a tendency to want to make story-required bosses somewhat simpler and easier than its optional bosses, something Path of Exile neglects, and I am firmly of the mind that GD’s approach is superior—Warframe, meanwhile, makes all bosses—required and optional alike—relatively easy, which I don’t think is quite the right call. In any event, one-way mechanics can still make a fight interesting if they are more than just different sorts of melee attacks…

    …but one thing that Grim Dawn is sorely missing as opposed to its other RPG competitors is that of boss mobility and positioning. As I’ve alluded to earlier, bosses in GD usually just run up to and slap the player a bunch, with exceptions being ranged bosses—which exclusively use one or two very standard ranged attacks—or stationary bosses like Loghorrean or Queen Ravna (phase one) that tend to be summoners at range and melee bosses up close. Path of Exile and, hell, even Titan Quest had more complex movement mechanics on the side of bosses. TQ had the great fight against Bandarithat I’ll likely remember forever (even if I keep forgetting his name), an enemy that would continually teleport deeper and deeper into the arena. With Grim Dawn soon giving players more movement options with the upcoming augments for medals, I think it’d be the perfect time to consider giving bosses more randomized movement options and stationary attack patterns to complement them. There isn’t really a lot to say in this regard, as there are plenty of examples of other ARPGs having more mobile encounters while Grim Dawn is far more stagnant in its combat. I think this is one of the few grave mistakes of the TQ engine, though for its time, its competition was not terribly different.

    So…what do I want to see?

    Mark Brown summarizes the Bungie analysis into the premise of players finding enemies more ‘intelligent’ when they have more health and deal more damage. If we’re to assume that ‘intelligent’ enemies make gameplay more ‘interesting’—which to me seems like a reasonable assumption to make—then clearly this faux-intelligence is a worthwhile goal. But it’s arbitrary and frankly harrowing when simple enemy attacks just wind up oneshotting players (which is, and always will be, why I think PoE is far inferior to GD), so there really ought to be a middle ground. Dangerous attacks are something of thin ice in ARPGs, but dangerous mechanics are much more welcome, as a dangerous mechanic does not even need to directly be what might kill a player.

    Suppose, for instance, if the crystals that the Sentinel summoned were much more potent, but could be targeted and destroyed by players. This’d be a strong beginning of two-way mechanics being present in GD—players can choose to ignore or destroy the Sentinel’s crystals, or try to bait the Sentinel out of their radius as they do now. Presently, this premise of ‘baiting’ enemies around is seemingly GD’s primary attempt at developing more involved enemy mechanics, such as with Kubacabra’s blood pools, but I’d argue that so far it is a half measure relying on player movement that is not unique from encounter-to-encounter, as players moving around is something they do throughout the whole game anyways. Kiting isn’t especially mechanical or interactive, but forcibly nudging players to do so in moderation is not entirely a bad thing either. This is why I like the premise of enemy mobility, because if an enemy has unique movement patterns, then players would need to follow that enemy around in a likewise-unique fashion.

    As far as I know of the engine right now, such opportunity for suggestions are slim. The turret system used by Theodin/Gargabol and the damage-pillars used by the Amalgamation/Ellena are among the most recent technological advancements in enemy skill usage, and both of those technologies are fixed-in-place, leaving little room for creativity. Moreover, gating special skills behind certain %’s of HP on bosses is also a nontrivial task, especially for mobile bosses—enemies are only allocated one ‘enrage’ skill at one threshold of health. Loghorrean, for instance, is three individual enemies disguised as one to make the fight appear to be three distinct phases, but you do actually kill two different enemies when Logh’s healthbar hits 66% and 33% health, respectively. So realistically, there is not a whole lot of room to work with for, say, Forgotten Gods without greater engine work. I do, however, think that with the pathing system rework teased for the addition of movement runes, enemies like Valdaran could reasonably begin to target ground-based areas of the map that are neither swapping positions with a player or teleporting right next to a player. Picking some position within a certain radius of their starting position semi-randomly to move/leap/dash/etc. toward could go a long way to spicing up combat with otherwise-stationary foes.

    The vast majority of this feedback has focused on the premise of interactive, two-way mechanics, but unfortunately to the best of my knowledge I don’t really know how much is possible within the confines of Grim Dawn. The aforementioned Doedre fight, for instance, is not doable to the best of my knowledge, even with significant Lua scripting, and being as we’re so late into the lifetime of the game, it’s worth wondering how much work should really be put into making more advanced bosses in the first place. So while some of this feedback might maybe be worth considering for FG design, I mostly write this should Crate ever decide to do a Grim Dawn 2 with an improved engine system. So, to summarize everything:[ol]
  4. Players being able to interact with things, including the potency/power of their foes, is important.
  5. Enemies being able to kite around the player, instead of vice versa, can help to add a unique flavor to encounters intended to be unique.
  6. High danger, while capable of being a means of communicating what to remember and avoid in a certain encounter, is not the end-all-be-all of enemy design for ARPGs.
  7. Optional content being harder/more advanced than story content is most probably an exceptional idea.

inb4 flamed by Bananatie

Attachment: bosses and mechanics.zip

2

I absolutely loathe the Bandari fight. Needlessly dragging on the fight by running away like a pussy is not what i call a “good” boss fight, at least for melee builds.

One fight i like in another ARPG is Valdis in Dungeon Siege 2, where you have to perform some tasks to make him vulnerable. He also summons two clones of himself and starts healing and you have to hit the right one for that to stop. I guess some boss fights with more interactivity like this would be pretty cool.

3

In repetition it is a bit extreme, and the fact that its movement is entirely linear isn’t great either - I think that can be chocked against the level design - but mechanically speaking I enjoy this implementation of phase-splitting far more than an invulnerability phase or an enrage phase like Benn’jahr has.

4

I’ve discussed this with multiple friends, praetorians, random strangers I meet online, and while I have met some people who were ambivalent, I’ve yet to meet anyone who didn’t think the bosses could be made a lot more interesting to fight.

And if they were more interesting to fight, there’d be less harm in giving them more hp/survivability since the fight would be more enjoyable and use your brain more. I fight almost every single boss the same way I fight any other mob. Spam my skills. Few exceptions.

Zantarin for my characters that lack invuln skills? Dodge shotgun then do above.

Mad Queen? Don’t hit with projectiles while she’s red. If I have totems or something that auto shoots projectiles, run up hills when she turns red. That’s it.

That’s not to say I can’t see some effort put into making some of their attacks do a % of the player’s life and be threatening, but often it ends up being more of a gear check than a skill check at the levels they’re currently at. It’s not like telegraphed moves have to be about 1 or 2 shotting characters either. A telegraphed aoe could be for an extremely debilitating debuff, or long stun. Still there are even more complicated things that could be added to make bosses feel more distinct from each other as mentioned in the above post.

It’s worth noting that when you’re under-geared sometimes some bosses I don’t normally care about become more of a threat so I play them a little more cautiously, but not much different still. If bosses are made to be mechanically complex then it’d be okay if they can mess up both fully and undergeared characters.

5

Making most bosses dangerous for both undergeared and fully geared characters kind of defeats the point of an ARPG. Since the early days of this genre, gear has been pretty much the main thing in these games, with player skill being very minor compared to it. You get better gear to kill stuff faster and safer, with maybe an handful of bosses made specifically for fully geared characters, because you have to test your gear against something strong (like Nemeses and the celestials).

I’m all for making boss fights more interesting, but neteuring the main appeal of this genre is not a good idea in my opinion.

6

Not really, they’d still be more dangerous for undergeared players. Geared players would endure their less potent attacks more easily and kill the boss more quickly. I don’t know why you misconstrue my points to some extreme?
I’m not suggesting that it should be an identical experience no matter your gear. Seriously, neutering the main game? You got that from “If bosses are made to be mechanically complex then it’d be okay if they can mess up both fully and undergeared characters?” Did you think I mean every single ability in the bosses kit should do 80% of your hp as damage or something crazy? Cut me some slack dude :rolleyes:

How much more interesting can you really make a boss if nothing about them is a threat?

7

When it’s put like this, i can only get that this means all bosses should in some way be a threat to fully geared characters, which in my opinion they shouldn’t, at least not the mandatory ones. Maybe if you clarified more what you mean i wouldn’t be getting such conclusions.

8

“Mess up” is vague, I didn’t expect a vague line to be interpreted in an extreme and specific manner, (especially given the context of the thread is making bosses interesting via mechanics) but fair enough Norzan. I’ll clarify now: some bosses should have some mechanics that are a threat to geared and undergeared characters alike, so long as they’re fair and you can play around them. This shouldn’t ignore that the bosses should also have basic attacks, spells, and some mechanics where proper gear greatly decreases the need to even pay attention to said attacks, spells, and weaker mechanics. I’m also not saying all bosses should have both.

9

It was more the word “bosses” that threw me off, because just saying bosses makes me think you want all bosses in one way or another to be a threat to a full geared character. If you just meant some bosses, then my bad.

I do agree that some bosses should be a threat to fully geared characters.

10

Oh well now that makes sense to me, but at this point we’re derailing what I think is a very significant part of the game to be addressed.

Yeah I’m 100% not for main campaign bosses being overly lethal. I think they could still use more interesting mechanics to play around but given that only 5% of players have the “beat ultimate” achievement it seems like it’d be a bad idea to add super lethal moves to the likes of Krieg, Cronley, etc. We’re on the same page there. The part where I think we all agree on is it’s a bit of a shame that almost every single boss in the game is reduced to facetank and smash buttons… sometimes pop potions/heal/survival skill in earlier levels.

11

Grava’Thul and Aleksander do require dodging and in case of Grava, it’s pretty much required that you dodge his nullification projectile or else you lose your buffs or just die. Aleksander’s meteor requires a ton of tanking ability to survive and in case of a lot of builds, dodging is required. Or just time a Mirror or Blade Barrier, that works too.

But let’s just say these two aren’t exactly popular among some players. Not saying that is impossible to have enemies with moves that are pretty much mandatory to be dodged and have them be fun.

12

Seems to me you want more of an AI tweak for your better bosses. I’m not sure how feasible that is for time and design.

In a discussion comparing Path of Exile and Grim Dawn, a friend impressed upon me the concept of ‘spreadsheeting over content.’ He argued that that was the extent of the scope of Grim Dawn gameplay, in which you reach a certain threshold of numbers in various areas and then you proceed to wade into combat until you inevitably win. And he’s right. This is not the case with Path of Exile, in which seemingly no threshold of numbers is attainable that enables you to seamlessly pierce through the game’s content

While I do enjoy GD and appreciate its lower hardware requirements, I do agree with your friend.

GDs very nature is to give access to most every top item and you generally see the meta with every build: 3k OA/DA and many of the same devotions (fiend, viper, hawk, ghoul, dying god, etc). There is little compromise needed in achieving greatness. And little incentive for that much variety, even in skills.

In PoE with various map mods and much rarer desirable loot, the focus is more on capping resists, some defence, and maximum sreen dps.

However, even PoE has evolved to an area dps and clear speed meta (much of it off-screen) which results in much safer game play. I believe that it was shown that this play style is more appealing to the casual gamer…and logically they went for a larger market. Similarly, reflect mobs were reduced significantly. This was a move away from the early years where teamwork was pretty important to survive (at least on hardcore), even if you had the patterns down.

While GD only has a couple of multi-stage fights, each of these steps is effectively a bash-em fight with patterns. (PoE is similar)

The GD nemesis can be bypassed for the most part. After a bit of practice/study - the telegraphs and attacks are known. (I can remember many PoE leagues where certain heroes were bypassed on HC)

TLDR:

The industry trend has been to appeal to casual gamer who doesn’t have a lot of time to practice/read. The majority of the fights reflect that.

Patterned boss fights really only give the illusion of choice. Moving from patterns to actual AI is a huge leap in time and cost. I’d like to see this, but I can’t see it as economic for the producer.

It would make a great GD2!.

13

And the great bankaikiller has come up with the best solution for everyone here. Create a dlc that adds more depth to grimdawn bosses, giving them more attacks, as a game mode, or add on. If it cant be done, then crate can set up a poll stating what features they can add more to the game, and us players can vote on what kind of dlc we want put in for the game with our money. We may just have to wait for a new engine with grimdawn 2.

14

a friend impressed upon me the concept of ‘spreadsheeting over content.’

Well since that was me, I guess I’m honor bound to make a response.

A few things –


One-way mechanics. An entity does something that potentially changes the state of the game, but there is not any further interaction. Example: an attack is thrown out at the player, and if it hits the player, the player takes damage.
Two-way mechanics. An entity does something that its opponent can interact with by choice, and choosing to or not to interact with the event has differing results on the game state. Example: a boss summons a pet that attacks the player, but if the pet is killed, the boss gets a damage mitigation buff; players can either try to endure the boss+pet combo and focus the boss down, or kill the pet but deal less dps against the boss afterward.
Your definition of one-way mechanics seems to overlap with your definition of two-way mechanics. In the example you gave, when an attack is thrown at the player it must necessarily hit the player, otherwise if the possibility exists that it cannot, that attack would be an example of a two-way mechanic. The corner case of this interpretation is when an attack carries with it a probability function to hit. By that I basically mean ‘the attack is happening, and a roll must occur to see if the player gets hit.’ If the player cannot interact with the fact that the roll is taking place, I consider that attack to be a one-way mechanic regardless of the outcome of the roll. Note that that consideration is more relevant to the old school RPG style games. In layman’s terms, a one-way mechanic must be when the player does not have ANY control over what is happening.

Since the GD forum is not the best place for a discussion of PoE mechanics I will not respond further about them here, as most of your post is predicated on erroneous assumptions and lack of game knowledge. If you want to get at me bout PoE stuff, hit me up in a dm.

Having cleaned up the point about one/two-way mechanics, your description of the lunaris and solaris fight being a ‘one-way mechanic’ is absolutely false. As per the video you linked you can see that you, the player, gets to determine which one of them you fight first (and therefore what order you fight them in the entire fight).

Moving along, Innocence not only does not have ‘strictly one-way’ mechanics, but I would say has strictly two-way mechanics since every attack he has can be dodged. I appreciate that its your favourite boss fight because it really is an awesome boss fight, but really this whole analysis of PoE is pretty far off.

Grim Dawn has also had a tendency to want to make story-required bosses somewhat simpler and easier than its optional bosses, something Path of Exile neglects

This is also patently not true. Map bosses and endgame bosses in PoE are both optional and are WAY more punishing than anything in the main campaign. In fact, these can be SO punishing, that community members have actually made third party ‘boss fight simulators’ so that people can even have a chance at these encounters.

In terms of what you’re suggesting for Grim Dawn:

there are plenty of examples of other ARPGs having more mobile encounters while Grim Dawn is far more stagnant in its combat.

Wholly agree, in particular with respect to GD boss fights.

Players being able to interact with things, including the potency/power of their foes, is important.

This is probably a good suggestion. The thing is, players dont need to interact with much in GD because it is far too easy to simply ‘assemble a sick spreadsheet’ which obviates 99% of the boss fights in the game such that you can literally facetank them without any consideration of what you are doing.

Enemies being able to kite around the player, instead of vice versa, can help to add a unique flavor to encounters intended to be unique.

Similarly, most encounters dont feel very unique in grim dawn because of the impotency of bosses due to ‘sick spreadsheeting’

High danger, while capable of being a means of communicating what to remember and avoid in a certain encounter, is not the end-all-be-all of enemy design for ARPGs.

The vast majority of grim dawn encounters are low danger and in fact become trivialized with capped res. The encounters which are high danger or have 1-shot mechanics built in (like zant, grava, eketzul, etc) are complained about almost constantly it seems like by the community regardless of how avoidable their attacks are.

That seems to be your general complaint about PoE, that avoidable attacks are in fact 1-shots, except you dont really see those attacks as avoidable yet.

Other ‘high danger’ bosses in grim dawn, like mog, or ravager are similarly uninteresting questions of ‘is your spreadsheet tall enough to ride.’

Optional content being harder/more advanced than story content is most probably an exceptional idea.

Yeah, that suggestion seems to be something grim dawn already adheres to. Vanilla campaign bosses in the game now are mostly trivial simply with capped res. Optional bosses can require very specific gearing to be able to kill.

I generally agree that grim dawn bosses are fairly boring affairs once you have a certain (very low) amount of game knowledge (like the fact that you should cap your res). An improvement to the general bossing experience would be very welcome by me, but I think would be very unrealistic at this point.

For the record, I have given exactly 0 feedback about general case bossfights in GD to anybody at Crate up to this point. When I did give feedback to make content harder (specifically claiming that the 1.0.5 nerfs of crucible were a mistake) I was told to shut the fuck up by everyone who heard me (including Zantai). Judging from my experience with that, and the astounding number of people who complain about grava, zant, eketzul, et. al. seemingly constantly, both in the forums and elsewhere I would say that Crate’s current interpretation of boss fights is exactly what its community wants. The number of people who absolutely sobbed and made all sorts of nonsense bullshit noises when Crate nerfed their ‘be literally immortal spreadsheet mechanic’ of having 4-5k DA only further solidifies my view that the current state of GD bossfights is what its community wants. People who play this game want to be able to spreadsheet over everything.

I dont think that ‘spreadsheeting over everything’ is really the right, or most interesting approach to a game, and would welcome changes. But to expect anything like that to happen at this point in development is probably unrealistic, especially considering the resistance put up by Zantai recently towards feedback which suggests changes to the core mechanics of the game. I think the best we can do is hope that FG boss fights are more like the AoM bossfights than vanilla (which are generally, in my opinion way harder mechanically, more unique and much cooler than anything in vanilla – campaign kuba is a great example of a cool bossfight with interesting mechanics).

A more striking problem with GD (which is what I consider the ‘core’ problem with boss fights) is that, (particularly in ultimate) there is no incentive to actually participate in boss fights before your spreadsheet is ready. If there was sufficient incentive to actually go and fight Dreeg before you had capped/40% overcapped acid res, that fight would be WAY more interesting than the res-check that it is now because there would be a very high chance that you would die. You would have to actually dodge attacks, rather than just run up and hold down your mouse.

Finally, I think that blaming the (im)possibility of development on ‘the engine’ is getting a bit tired as a supposition. Crate have already proven that they can mold ‘the engine’ to their will, most recently with the revelation of the new movement skills and introduction of ‘Shattering’. I think it is much more of a question of what they think is a profitable expenditure of time and resources at this point. Frankly, from what I’ve seen of shattering/FG so far, I am much more excited for more information about their ‘unnamed’ project than FG’s release.

15

I’d definitely like to see some more interactive fights in GD but I recognize that a lot of the limitations are due to the engine.
But even so, some things should be possible: look at Lucius for example. He is one of the few bosses that has very visible cues for his attacks (his shotgun projectile and meteor drop to name some).

The difference between that boss and many others is that Lucius winds up for a moment before he attacks. Some other attacks such as Log’s poison breath, Zantarin’s shotgun and most of the Warden’s attacks are like that too.
Between preemptive and reactionary action I find the former much more preferable.
For many of the bosses, they do something and you either take massive damage or just take a few steps to get out of their spell effect, but the spells are used with very few visible cues, you just have to know their moveset from experience.

Or take the Anastasia/Moosi sky attacks where you can’t really predict where they drop, so your main defenses against them is: 1) being lucky 2) stacking resists.
That’s always been one of my gameplay peeves for GD, and I really wish abilities like that would have a visible cue so the player can walk out of the way to avoid damage through awareness.
It gives the player a feeling of satisfaction if you’re undergeared and manage to beat a boss through careful attention, but you can feel the opposite way if you die to an attack without a good visible cue.

That’s why I hope that in the future (whether it’s a patch, a new expansion or a whole different game), we can get some adjustment on boss cues and hopefully some new attack patterns too.

Some obvious patterns that I miss in GD are:
Bosses charging in a straight line after a windup, and player needs to move out of the way to avoid damage (could be an enemy version of the new player movement skills)
Ground pools that spawn in random patterns where the player needs to stand in the gaps to avoid most of the damage
Multiple projectile attacks with safe gaps in between (the sentinel’s wave attack is somewhat like that but I’d like to see the danger and safety zones be more pronounced)

GD doesn’t need to become full on Dark Souls / bullet hell since that’s a different kind of game, but having some more elements of strategy beyond what we have now, would make the fights immensely more interesting to me.

16

OP’s post can generally be summarized as:

“I want to hold LMB and do nothing else until boss dies. If I can’t do that, I don’t like the game.”

At no point does any valid argument for a game’s mechanics being poorly designed manifest. All the OP’s bloviating manages to prove is he can’t separate subjective opinion from fact.

Not liking the fact that you are allowed to dodge an attack does not mean it is bad game design. It only means that you do not like the fact that you can’t facetank everything until it dies. Note that this does not mean that the game’s mechanics are bad.

This means you’re bad at games that require actual complex mechanical interactions.

P.S. You don’t like to dodge attacks in PoE, but you want Grim Dawn bosses to attack you in a manner that requires you to dodge.

rick_and_morty_get_your_shit_together.meme

17

So I guess the gist of the argument is that bosses should invoke more active events the player must respond to while having enough unavoidable-but-tankable offence and raw passive defence to serve as a gear/build check as well.

18

It is actually rather impressive just how incorrectly you interpreted Ceno’s post.

19

Following a lengthy (and productive) argument with a man with too many “u”'s in his name, the two definitions at the beginning of the post have been expanded upon and now number three. This change has been reflected in both the PDF document and the forum post itself (and there was also a short line in the forum post which was formerly missing from the PDF; that has since been added in).

Weird. It’s almost like I said exactly that within the feedback.

I wouldn’t disagree with what you’re saying in a literal sense. I think, certainly, if you try to appeal to a wider audience you’ll certainly find yourself with wider wallets in turn. But I, in my total-lack of market analytics, can’t help but feel that there’s a large enough audience for the difficult/hardcore games to make enough of a buck on to serve as a worthwhile, even if not optimal, investment for developers. Again, not necessarily something to go back and change everything for for a pay-once-to-play-forever game that’s been out for 2.5 years, but worth considering for future endeavors.

Well we’ve been over this in DMs (as wisely suggested by you) and clarifications/amendments have been made in turn. Maybe the lens through which you read the more-formalized OP will enable you to reexamine things in this regard as I had originally intended. Get back to me, here or otherwise.

I believe I admitted in the DMs that a lot of one-shots in PoE do seem perfectly dodgeable to me. Finally (after ahem 40ish tries) beat the Shaper/Elder fight in the simulation you linked, and rarely was ever oneshot past the tenth attempt or so. But if it wasn’t already painfully obvious within those DMs, allow me to make it even more obvious here:
I will never, under any circumstances, find a 1-shot ‘mechanic’ to be a good idea. You could give a player two-minutes of warning and if you killed them after that because they fucked up despite your warning, I’d still call that head-in-ass game design.
You’re free to try to convince me otherwise, but the last 11 or so years of playing ARPGs hasn’t really done so, so good luck. Yes, I would be a firm advocate for the removal of any existing guaranteed one-shot mechanics in GD; the only reason I’m not complaining about the Zantarin shotgun, for instance, is because it’s not guaranteed and I’ve survived it on many a fuck-up.

To your other points, this is why I find being able to spreadsheet a good thing, and if we can’t convince each other otherwise, so be it. But I am adamant that being able to attain “statistical god mode” for particular encounters is perfectly OK in an ARPG. You and I will agree, however, that it is far too easily accomplished within GD to do so on a universal, content-encompassing-scale, and that the path to that level of stats does not necessitate any significantly challenging content, to which I’ll throw PoE a bone and admit their concept is sound - even if we may bicker on the overall execution.

TL;DR: I agree that spreadsheeting over everything is not sound design. However, I think not being able to spreadsheet over something, no matter the costs (in time/farming, preparation, etc.) makes that something poorly designed.

I think this is both fair and unfair. Sure, if Crate says something is simply just not possible, I’d think that’s fairly ludicrous. But if they were to say it’s just not viable to do for how (in)significant they’d assess the cost/rewards to be, even if I may disagree with that assessment (which I probably wouldn’t, because I don’t have their knowledge of their financial situation), I’d still find that claim reasonable…

…that’s basically a more wordy version of what you said two lines later, I think.

Wow, you managed to miss something literally in the first two paragraphs of the post, before the feedback even started. Congrats.

But hey, as long as we’re making personal attacks against one another, allow me to appreciate that you, Mr. “I spent more money on PoE than I want to admit,” chose to make your first post on the Grim Dawn forums in such a way that so very clearly demonstrated your own ability to separate subjective opinions from facts.

20

Yes we would :wink: