Grim Dawn's Monster AI is really generic

Grim Dawn Ideas and Feedback
1

Hey all!

While I like Grim Dawn, this is one of the things that has been really bugging me. As the title states, I find the monster AI really generic. There’s nothing special to them. They all feel the same, making the entire experience… well really boring after a while.

Don’t really feel like writing heaps as I usually do, so I’ll keep it short with a few examples.

  • Almost any monster follow the same pattern: Move to the player until in range, attack. ranged or melee, doesn’t matter. For ranged characters, this makes kiting extremely effective(easy), and also extremely boring. The only way to make this types of monsters challenging is by upping their defences and offences. And that itself is bad game design(just look at Diablo 3 when it got released).
  • Monsters have similar attacks. And with that, I mean… Well, most monsters have the same types of attacks. For example, few of the ranged monsters, with special attacks, have a multishot, attacking with multiple projectiles at the same time. Or throw something on the ground from a distance, arcing, and sometimes leaving damaging ground on impact. Rarely, monsters(mostly champions and bosses)Summon some sort of a firestorm, multiple balls of an element that strike the ground in quick succession, or sometimes at the same time.
  • Monsters don’t have an identity. And that is a result of what I mentioned above. But monsters don’t seem really different from another. Yes, some shoot red bullets, other purple. Yes, some monsters inflict bleeds, others raw damage. The mechanics are interesting enough as they are now. Just the AI not. It all feels… So generic. I see new monster models, and see them using the same types of skills over again, or move in the same type of pattern as any other monster. I don’t have the feeling every monster has it’s own identity. More the other side of the extreme: they all feel related to another. A shame really, especially since this game has such a heavy emphasis on exploring(Love it!), encountering different kinds of monsters would make the experience so much more awesome.
  • Bosses are barely any different from the random champion encounters - in both AI, and skill setup. And this is a major offence. Bosses are supposed to be the big conclusion to an act, they have to feel strong, exciting. Do note: This doesn’t mean they need to be hard. But as it is now, GD’s bosses are… Well. Close to normal monsters. Especially they need to shine. Shine in AI, Shine in types of Abilities. I don’t care if you stay with the current animations - I’m sorry, but those are also… really terrible compared to your current competitors - just give them an edge. Make them feel a bit smarter! Don’t let that warden guy or Chronley chase you around the room mindlessly. Give them a proper AI. Let them react on you as a player, use the players location(No, the warden using his stomp when you are near him is not really the thing I am aiming at >.>), the skill they are trying to damage you with or whatever. Be innovative. But as they are now, I say the bosses are a major offence when it comes to proper game design. And that is bad guys!

Really bad, because right now, this is a good game. But that’s all. I play it, and just move on. That while in theory this should be one of the games I can sink many hours of time into it. As a theory crafter in similar games(Diablo 3, Path of Exile)I can tell you that this whole skill system is both unique and amazing. This game has found a fine line between being too complicated and too shallow. I just wish I had a more potent surrounding to test it on. Having monsters act exactly the same, just doesn’t do this system justice.

And no, I am not saying the game is too easy. I can’t even say that since I have’t fully explored all types of classes, skills and what’s not. I am just saying that the AI should get a boost. Do that, and I think this game will go from just good to amazing :slight_smile:

2

Good comment. I had noticed the same thing with a monster that explodes in fire. Very pretty but the damage from the explosion so small that you can ignore that ability of the monster.

Also agree that bosses not as interesting as in TQ.

3

Well these are some interesting thoughts and I think its pretty difficult to fix.

We all love grim dawn, It takes everything, which was good in the great hack slay ara (well diablo2, Titanquest)
put it in an appealing World with lots of charactercustomization. Thats good or is it?

Because all it does is basically take these old features and cement it
in the good old formular.

It works but it obviously limits what you can do with enemy patterns
aswell. There are no new cool dodge mechanics and while this is
fine in a hack and slay and loot enviroment you can also not put
as many cool movement requiring dodges in this game.

Basically you are right there are no really interesting fights, no cool
shoot patterns like missing good example, but diablo3 got actually
quite interesting bosses in it. Even though diablo3 is as generic as it
gets and I probably wont ever play it again, the enemy design from
a gameplay perspective is cool. So much stuff to dodge and you got
skills for it to use to dodge it.

Take a bullet hell game like Enter the dungeon.
Imagine enemies would shoot at you with similar patterns,
imagine you would got the abilities to dodge them all.
This would allow for more interesting opportunities to design bosses.

But this is probably not what the majority of players expect from
an hack and slay game. Iam glad there is finally a great hack and slay
game which allow me to rotate my camera ^^

4

Keep in mind GD is not a big budget title, so something had to give. If enough of us buy this game, maybe the expansion will come with enhanced AI.

But I agree with you, the bosses are definitely meh (especially compared to TQ). To the point that many times you don’t even know you’re fighting a boss until you drop every enemy on the screen and notice one of them stubbornly refuses to die.

5

Any of you guys play path of exile? they went way out of there way to make most if not all monsters have some kind of really unique characteristics/attacks.

I HATE that game. I really want to like it, because it’s a similarly low budget title from my country which doesn’t have many or any other really noteworthy game releases, but the enemy behaviour just ruins it for me. Basically every monster has some characteristic/ability which just makes them feel really cheap or annoying to fight. I tried for a long time to like the game and i just can’t.

Now, i recognise that perhaps monsters could be developed in a way that doesn’t make them soul destroying to try to fight, and that PoE’s developers merely failed in that aspect, but it seems to me that any monster that doesn’t have some kind of behaviour which makes them difficult or annoying to fight is ultimately going to wind up forgettable. So it seems to me that it’s a dichotomy: would you prefer the game to be too easy or too hard?

6

Keep in mind, this is only my opinion here. But low budget or not, it’s not an excuse for bad AI imo. Crate already showed they have some really awesome people working on the game, as there are plenty of things I really admire in GD. Compared to all those things, I just think the monster AI is just… well really below the rest. And that’s a shame, because it really makes the game bland, and generic, while it shouldn’t have to be!

  • One thing I really like about GD is the fact that there aren’t any mass clear skills, like in Diablo or Path of Exile(skills that you cast a few times, that can hit almost any monster on the screen). It’s also hard to just simply clear monsters with just a single hit - in general you’re fighting the monster for longer amounts of time. I think this is a really amazing things to have in such a game since it allows for way better partying(the typical MMO tank/damage dealers/healers are actually viable in such conditions)and planning builds becomes to much more important when you are playing solo(Glass cannon build? Well. can work, but it requires skill).

Though. This only works is the AI is engaging enough. Because right now, I just grab my 2hander. And shoot > Walk back > Shoot > Walk back. There is nothing I need to anticipate. And the times I die, is simply because of the gear checks. You don’t have enough damage or survivability to survive the group of monsters? Well, nothing to do about, no matter your set up/skills(Okay, maybe I do exaggerate a bit, but it often feels like this way) - And that is due to the simple monster AI. You can not buff up the monster difficulty with ‘stupid’ AI, the only way to make monsters hard is just by simply tweaking the numbers. Something no game should ever do(I’m looking at you Diablo. Yes even now, nothing has changed(Tormex X is the same game only with 10x the monster HP/damage. yes. Fun.).).

And hard? Possibly. Impossible for a low budget game? No.

You don’t need quality artists doing many works, nor need 100 programmers to work 24/7 around the clock. A small group dedicated to AI, could easily rework the entire system. And with the game as it is now, I’d say that upgrading the AI should have it’s priority - it would definitely make the game look way more interesting, which could possibly make more people interested in the game(don’t forget, people often look up gameplay on youtube to see if the game is something(and let’s be honest, skill effects and the likes isn’t the strongest part of GD(but Will require a lot of money to do well))).

And GD has amazing ways to do this… really well!

Let’s take Cronley’s hideout for example. I think it’s a well designed place, level design wise. But. Most of the enemies there are… Well utterly boring.

They stand still. Hold out their gun, arm stretched. And shoot. That’s it. Sometimes they run away when you get too close, but. I don’t have the feeling I’m fighting something with remotely any intelligence. And those things are humans even! I mean, if it was a zombie doing so, sure. I’ll buy it. But not with those guys. How to fix?
Maybe they shoot one or two times in a row, try to re-position, and repeat. And the re- position location? Let’s say they try to stay 20 to 25 units away from the player(I’m making up the numbers as we go). They shoot in between the movement, so they actually do attack, in case of melee guys.
Let’s make it more engaging. There are melee humans as well. Don’t let them just chase you. The moment you target ranged mobs, they will locate not just next to you, but try to move in between you and the targeted range mobs. I mean, it would be chaos if the player would keep changing ranged targets, so let’s say they have a cooldown of 4 seconds before they try to move between you and his ally again. Suddenly, this encounter is way more interesting then it is right now. Hard to program? Possibly. I don’t know the exact code they are using, but impossible, no.

Another example? Let’s give the monsters a little ‘handicap’, so they can’t do the same thing over and over again. The Cronley ranged shooters now have a magazine(intern, player don’t know about it)with six bullets. After the sixth shot, they need to reload, meaning, they will step back, while reloading. Suddenly, they do morethen just stand still and shoot.
Make it more awesome, let them use the surroundings as well. I mean, the map is pre rendered, so let them run to spots they are relatively save. Let’s just say the devs make a few spots, they can hide behind, just for reloading. Once done, they move out of the hiding spot, and start shooting again. Just mark many places on the map that should be relatively save, and add the following calculation: If the player(not from the camera’s perspective, but the ingame guy) has a direct line of sight on the ‘save’ spot, Cronley’s shooters are less inclined to use it, and prefer other save spots over it. This way they will actually look intelligent, and way more active. It’s also way easier to balance certain mobs, as you can tweak more than just the damage output, their HP and their movement speed.

And I made these two up at the spot. I’m pretty sure you guys have amazing idea’s on this subject as well, just like the devs working on this game.

So yeah, long story short: I think it will be healthy for the game if they started working on the AI, or at least make it a high priority :slight_smile:

Also, I agree that PoE has a lot of things going on when it comes to monster AI. I love PoE, and I’m only here because I’m waiting until the next league of PoE is coming. But I agree with you that there’s a lot going on. Once you are used to everything, it’s fine. But until then, PoE is really hard to get into.

The main problem PoE has, is not just the amount of different monsters in the game, but mostly the frequency how it’s introduced to the player. With all the random nemesis mods, strongboxes, bloodlines, Domination shrines etc etc etc. There is just so much coming at the player as once. As long as GD does not invite so much randomness in their monsters, I don’t think it’ll drop in the same pit.

Don’t forget - staying in one extreme is not what makes a good game. Finding a balance between extremes is :slight_smile:

7

corrolary to that and possibly easier thing to fix:

enemies “Detecting” you when your in absolute range not counting line of sight

then shooting AT you even though there is clearly a wall and no line of sight.

  1. the game knows you cant be hit because the wall blocks it
  2. the enemy just keeps shooting
  3. can the information from 1. be used to change 2. into a. path closer and check for shooting LOS again b. repeat until shooting LOS exists.
  4. i likle the idea of ranged enemies actively trying to keep their distance. maybe let them move a bit slower.
  5. some enemies who have “higher intelligence” should run away when their “pack” gets overwhelmed by you, and try to alert other packs of enemies to swarm you
8

Ah yes, forgot to mention that, but that is one of the reasons why I felt I needed to make this topic.

  1. the game knows you cant be hit because the wall blocks it
  2. the enemy just keeps shooting
  3. can the information from 1. be used to change 2. into a. path closer and check for shooting LOS again b. repeat until shooting LOS exists.

I think there is even an easier way. The game already has a proper collision detection system(That is why bullets are blocked by walls). That collision detection should make for easy Line of Sight 'calculation" Will I hit the player or not if I stand here?

  1. some enemies who have “higher intelligence” should run away when their “pack” gets overwhelmed by you, and try to alert other packs of enemies to swarm you

I really like this idea. The trigger could be something like “when main pack loses ~50% of collective HP”

You could even enable a morale system as well by doing this - let the enemies show some emotion, fear etc. Could make for some really interesting plays haha

9
  1. is the one that really got me. how come the enemies are fearless suicidal attackers

i guess it sortof makes sense. theres like 50 of them and one of you. they have no idea you’ve cleared the mines 200 times. for them its the first time :slight_smile:

but if we meta-it a little bit. it would be more fun to try to kill stragglers before you get a call to help. or just let them go and have them pull for you.

ah well. we all just want more strategic battles right.

when i played my first char. a 2h warder. i would stand by corners/beside doorways(not in them) and let enemies come to me and i could pick them off one by one. the ai needs more work

10

From a technical PoV, AI eats CPU for breakfast. And since the engine of the game is single threaded, adding smarter AI may be a tad too complicated.
Sure, smarter AI raises the value of a game a lot (coughHalflifecough), especially AI that learns and adapts. But today you get AAA titles that come with brain-dead AI (and no one notices/cares), so you can’t really fault Crate here.
Bottom line, I would love better AI, but I’ll take it as a bonus if it ever happens.

11

because nothing is more annoying then enemies that run away from you eg on low health or because they are ranged. You can make the battle strategic, they should not.

They are cannon fodder and should be, exceptions should be heroes (to a degree) and definitely bosses. It is not truly about AI here but about mechanics.

12

There are ways around it really. Raising individual AI might not be the best way to go about this, but changing the AI as a group shouldn’t be too much of a load(have one big guideline that changes how a group of monsters behave). The individual Ai might still stay simple this way, but at least it will feel a little bit more engaging.

Sure, smarter AI raises the value of a game a lot (coughHalflifecough), especially AI that learns and adapts. But today you get AAA titles that come with brain-dead AI (and no one notices/cares), so you can’t really fault Crate here.
Bottom line, I would love better AI, but I’ll take it as a bonus if it ever happens.

Well, the difference between most modern games and Grim Dawn is, that Grim Dawn has a bigger emphasis on the individual battles instead of the mass killings, as you have it in many triple a(and non triple a btw)games. The balance between monster HP and player damage output is simply different then most other games, as most similar games(Torchlight II, Path of Exile, Diablo III)prefer the players to have more overall power: If you want to wipe out entire groups, you can do it.
It seems Crate decided not to go that way. There are still builds that can wipe away groups of monsters, but they are far rarer, and only really seem to work on the lower

And that is fine. It makes Grim Dawn unique and fun in it’s own. But that approach also needs a more engaging AI as a simple AI will quickly feel really generic, way faster then other games.

See it like this. In PoE/Torchlight II and to an extend D3, monsters don’t have too much time to showcase their skills. They can often attack once, twice, maybe three times, but then it’s over, either for you, or for them. Due to this short encounter, a complicated AI is not necessary. The player won’t notice the stupid AI, and if they do, they won’t be bothered too much by it, as the fight itself won’t produce much repetivity.
GD however showcases longer fights. Yes, some mobs can be shot down in just one shot if you have a properly geared build. But most of the time, and especially on the ultimate difficulty, this is quite rare. And even if you can shoot down the cannon fodder, there are still the higher health enemies, that are not heros/bosses(things like those Skeleton Revenants etc). Your fight with them can sometimes take up to half a minute. And that can be really epic, you know those battles where you are at the edge of your seat. Grim Dawn has this possibility. Moments you want to put on youtube, because they are just so epic, things people watch a stream for. And this is unique to almost any big dungeon crawler out there. But having to kite, or repeat the same pattern over and over again for half a minute is not engaging, nor fun to do or watch and this is a missed chance. A big missed chance.

And let’s be honest. Grim Dawn, as it is now, isn’t the biggest competitor. Actually, It’s one of the smaller ones among some big titans. And it should not be, because it has unique idea’s, a fun and engaging skill system, a beautiful world. It just falls short with the combat phase.

You(and I btw)might be able to cope with how it is now, but many others won’t. And that will keep GD small. And I honestly think, this is the first step for them to really shine, even to the rest of the gamers, without this game losing what it currently makes so awesome.

So that is why I decided to do this topic, instead of silly things like the mediocre effects or whatever. I honestly believe, improving on the monster AI, no matter how difficult it might be, will certainly do this game much good :slight_smile:

because nothing is more annoying then enemies that run away from you eg on low health or because they are ranged. You can make the battle strategic, they should not.

They are cannon fodder and should be, exceptions should be heroes (to a degree) and definitely bosses. It is not truly about AI here but about mechanics.

Think outside the box. It doesn’t necessarily has to be just monsters running away on low life. It could even be things like, monsters reverting to their animal instincts(Running/Hiding/Fighting/Feigning Death), or trying to work together. I think the big message is, something could change when monsters are in ‘danger’. And what they do does not have to be annoying, it could be engaging even.

Looking from that perspective, it’s more about AI then just mechanics :slight_smile:

13

for most it just might be a “is it even worth it” cost. You talk about it would be easy if a few people worked on AI adjustments… Crate as a company is just a few people. I’d prefer expansions over AI tweaks.

In most cases trash dies so quickly it would be pointless to display low health AI tactics. I could see hero’s rallying additional monsters or going into rage at lower HP be more interesting. Even on ultimate with my half ass geared char kills bigger white trash very quickly.

There are far more interesting situations with complex mobs such as fodder, ranged, and healers all at once. The game offers a few instances of this where I’m like “shit I need to get those range and healers but I can’t because of everything else going on” Perhaps just tweaking what monsters can do in general so not everything in the game is 99% straight attacker.

Kiting is the most common method because it is a way to cope with the current AI. If you change things you do realize people will just adapt the same new strategies for the new AI and people will complain about how AI does the same crap. Not every battle in the game needs to be an epic struggle, I get maybe making certain areas and making hero’s more scary. I don’t need to spend an hour trying to clear on a zone because every battle is EPIC.

14

Pretty much this. There is no point for such behavior in this game. Nobody wants to spend extended amounts of time chasing down scattering trash in every single battle, which is exactly what would happen. Chances are they will die long before they have a chance to utilize these tactics. As for feigning death, that would be more retarded than them running away. In that case just have enemies lay down the moment they become aware of you and get it over with.

Exactly and goes with the above as well. While the thought of more intricate AI sounds cool at first the end result would often be over long fights with every single battle. The end result would be turning more people off than on.

GD is about building a kickass character that rips thru mobs and finding loot.

About the only thing that needs help with the AI is the ranged wall problem.

Sent from my SCH-I605 using Tapatalk

15

Content only last so long really. With the same monsters acting the same way in every act, it’s only a matter of time before people put this game away. And I never said every battle should be a epic struggle(if I did, wasn’t what I tried to convey). But there should be at least difference in battle - every monster encounter feeling almost similar to another is just… well… not really all that good.

Sure, trash mobs don’t need to have all the fancy things, but there are plenty of mobs that are common encounters, but do not die in a single blow.

And then there are champions that show the same boring behaviour as another. etc etc

I said it before, but GD isn’t one of the bigger competitors atm. Most people don’t even know the game, and that’s a shame. And it’s understandable. The combat looks just boring and dull, it’s just not appealing. And fancy pretty skill effects won’t just fix it. All you see if just monsters walk to you, and attack. The AI can’t get more generic.

So I say, yes, this is important, especially if GD wants to stay in competition with the rest(or keep earning enough to stay alive as a company).

And with what they have now, saying that they are small is no excuse. They’ve produced a great game, but it’s time for them to also invest time and efforts in the places they are lacking.

16

I feel bad saying this (because I can see that you’ve put a lot of thought into it) but I feel you’re taking this thread to really high extremes.

The reality is that the only thing wrong with monster/pet AI (and it’s been like that since TQ days) is that ranged enemies will continue trying to attack you through obstacles. Everything else that you mentioned is NOT required.

Grim Dawn combines groups of enemies so the player has to prioritise targets. The gameplay mechanic is simple to understand. Late game this doesn’t change, you’ll still focus on the stronger enemies but due to being heavily geared can easily take out the entire room in seconds until you run into one of those (3+ enemy heroes spawn) encounters.

Try picking a sword+board soldier and fighting melee picking defensive options and tell me what you notice from the AI when fighting Grobles on Normal Veteran.

A typical encounter looks like this.
Groble Champion (melee defense) - The strongest the Grobles have to offer. This guy has high HP and carries a shield. Will occasionally block your incoming attacks and hits like a truck early game. Though incoming attack is heavily telegraphed and as expected, they move slowly.
Groble Warrior (melee offense) - These guys are quick and will unleash a 3-hit combo on your character. They are fast and agile. Will quickly move in and try to surround you.
Groble Scavenger (melee trash) - These guys spend their time taunting you and running backwards and forwards and occasionally running in to hit you. They are simply put just trash.
Groble Tracker (ranged offense) - They fire a poisoned dart. Early game the DoT effect is noticeable. Will continue to hit you from a safe distance and as expected will retreat if approached and re-aim and fire.
Groble Witchdoctor (ranged support) - OK, these are the target priority. They heal all other Grobles in the vicinity and launch AoE poison projectiles from a safe distance.
Groble Hero Spawn (Surprise offense/defense/support) - Sometimes there’s only 1. They specialise in unique abilities. Have much higher HP and have increased stats. Occasionally the game will spawn 2-3 at a time. These guys make some of the games best moments and are usually the reason your characters just died.

Now here’s the thing. Taking these guys on melee focusing on defense is an uphill battle early game. It is for this reason that most players will take a big AoE skill such as forcewave and steam-roll through Act 1 and then re-spec to their desired character later on.

I see nothing wrong with the AI in terms of what it tries to offer through different enemies. But do feel a lot of it is over-shadowed by large AoE skills late game.

17

Maybe I’m taking it to the extremes. Heck, now I think about it, maybe it’s not just the AI I’m really bothered with.

It’s just that the monsters themself feel generic. I guess I’ve been too focused on the AI to see the bigger picture. But even so, I still think the AI is bland, and really needs some tweaking. I’ll try to explain going by your example. This is what I see:
Groble Champion (melee defense) - Walks towards the hero, attacks, has a lot of health.
Groble Warrior (melee offense) - Walks to the hero, attacks. Faster(I guess they try to swarm me, but only when I stand still. If I move around… they uhm. Well. See for yourself)
Groble Tracker (ranged offense) - Walk until in hero range, stand still, attack.
Groble Witchdoctor (ranged support) - Stand in range of their allies, cast their heal - (annoying little pricks hah)
Groble Hero Spawn - Walk towards the hero, attack the hero, cast their spells every few seconds.

With exception of the Witchdoctor, they all have the same basic pattern. Move in range, attack.

There is literally nothing else to it. And I would be fine with this, if not every monster pack does the same thing. Maybe there are a few monsters out there that differ from the packs, but there is barely any variety. After playing my second character through Elite, all I could think was - This is melee, that is range. I know exactly how to deal with them.

Yes, the mechanics behind their attacks is nice. I like how the DoT actually means something in this game, that there are plenty of different types of attacks(The three Elementals, Chaos, Vitality, Acid, etc etc etc). Everything behind the attacks is something I appreciate, something I like.

But the uniqueness behind those types are overshadowed by the bland AI.

I know I’ve been taking this very seriously, but I honestly think this is one of the biggest things that keeps this game… well small.

Just take a gander at the amount of users on several reddits.
Grim Dawn has around 8K users while Path of Exile 64K and Diablo 3 83K. Even the newer game Enter the Gungeon seems to have much more popularity if we look on Reddit(11K).

Same with the amount of views on Youtube, where both Path of Exile and Enter the Gungeon(Couldn’t find the D3 views as fast)had way over 500K, while Grim Dawn barely has any(12K)…

Don’t say this is because I am comparing them with way more popular games, or triple A companies, because Grim Dawn is doing things WAY better then Path of Exile and Diablo 3. Grim Dawn is a great game, but it is not shining as it should be. And that is simply because there are a few points that really drag down the game, by a lot. And one of the things is the monsters in general.

In really think that making monsters more unique, maybe not even by just improving the ai, maybe with better sounds, different types of animation, or mayby some of those useless taunt animations(Seriously, why would a monster taunt an overpowered thing that’s been slaughtering everything with a single hit).

Maybe I am being a bit too serious about it. But I really want this game to rank up on today’s market, love to see more people support this game. And I honestly think improving on the monsters is the first step to turn this into a great game :slight_smile:

18

Very well said! Last few days im playing Grim Quest mod, the one that adds the TQ Masteries to GD. Its a blast, because there isnt any “one hit kill them all” skill and that gives a whole new meaning where you position your char in the fight, what enemy type hits you, how and when you use your skills etc.

GD masteries are in need of an thorough and targeted overhaul so GD can show the full potential it has.

19

popularity doesn’t always mean = good… pop music for instance is complete garbage that is forgotten within the next year because the next thing comes along… D3 for the most part is garbage most of its career and by the time they got it right the game was far too old. Now its just a joke of power creep.

Those other games have also been around much longer or have a franchise basing its name from (dunno about Enter the Gungeon I’ve never heard of it).

If GD was out for years or had multiple releases it player base would be much larger, that is just simple fact.

Traditionally unique AI is not something expected out of an arpg and if Crate could offer more and be that special niche then great for them, but this is something I’d expect an attempt from a triple A company… not a group a few guys…

I think this game offers a lot and with more tweaking and expansions it will grow even better. I just don’t think specialised AI is what people are looking for… or not sure if its even possible yet (you also got to factor computer requirements). If you make a game too hard (essnetailly what smarter AI would do) less people will like the game and give it a good rating. You need derps to dump their money into the franchise and make it fun enough the first play through so people buy it.

The people who stick around in an arpg are the theory crafters, the people who enjoy the multiple classes and builds that exist. If you want a game that offers unique experience every time you fight something different, tbh go find a puzzle game.

I dumped about 1400 hours into D3… and that game was shit for an arpg. I guarantee with even the “shit bland AI” that you refer to (even though every other ARPG has the same type of AI) I will easily put more than 2K hours into this game because there is so much potential for builds to play.

I’m all for trying to make the game better, but from a resource perspective I really think the AI is fine. There are far more other things to work on before trying to establish ground breaking arpg AI from a indie company…

20

Congratulations! You just described the tactics of every fighter/soldier/army in history. It’s hard to get more realistic than that.