Goblinworks Game Designer
|
| 8 people marked this as a favorite. |
I'm starting a new thread to try to have a productive discussion on this, as the other thread is already pretty heated. Let's please stop arguing about whether or not the system is leashing and talk about the pros and cons of potential solutions to what you don't like about what's going on right now.
The threat system we have right now is very simple. As Ryan pointed out, better AI has not previously been a priority, so, like a lot of our systems, only what we considered minimum viable AI was implemented.
There are a lot of things that seem simple to describe, but would be days or weeks of work on the part of our AI programmer. Our AI programmer wears several other programming hats, so asking for something hard to program means some other area of functionality that he's working on gets delayed. So I reserve the right to note that proposed ideas are much harder than you think, such that if you really, really want to push for them to be crowdforged, something else would probably get delayed. That's just the reality right now.
There are also variables in the system that I can change by just adjusting a spreadsheet, and I'll note those by bolding them. These are the things that, if something seems popular and well reasoned, I can just change and you'll see it in a build soon.
There is a much more complex AI system written up in draft form, but it hasn't been vetted by programming, so some ideas might just get "that's very similar to what we want to do, so I'll convey its popularity to the rest of the staff." :)
So the actual rules creatures are following right now:
- Creatures notice you when you get within 20 meters (shortened by being in stealth), and this immediately gives them 50 threat towards you. They immediately share this with all other creatures within 15 meters. (Known issue that one-shotting them at range keeps them from ever sharing this initial threat.) I'm not sure whether they get this initial threat on spotting you if you damage them from more than 20m, or whether they just share the threat generated by the damage you dealt.
- Damage you deal to the creature and damage dealt by the creature to you is converted to threat at a 50% rate (e.g., if you do 100 damage, it gets 50 threat). This is the only way to add threat currently other than the initial perception range.
- Creatures lose threat at a rate of 15 per round (actually tracked as .25 per tenth of a second). In combination with the 50% damage to threat rate, this should mean that if you're doing less than 30 damage per round (aka 5 dps) to the creature, and it's not hitting you, it will eventually give up and go home even if you're significantly injuring it over time. If they just got the initial threat, and have never hit you or been hit, they should give up after 10 seconds. (If you're positive you're doing more than 5 damage per second on average, and it's still giving up, please give us a write up on exactly what kind of enemies and situation that's occurring in, as that's a bug. Also, if you're sure something got initial threat and gave up sooner than 10 seconds, that's also a bug.)
That's pretty much the whole of the system right now.
We did try a simple method of sharing threat regularly with the group rather than just on initial aggro. It resulted in an unfun situation where in an actual group, all the monsters would always focus fire on whoever had the misfortune to pull aggro on one creature. In the design, they're meant to share when they hit certain milestones (like hitting half HP or less) rather than constantly, but that was complicated enough to implement that it didn't make it into the priority list yet.
So please give us your thoughts, and please keep it constructive. I'd consider posts phrased something like "I suggest X. As I understand it the ramifications of that would be Y. I think it's worth it because or Z." to be the most helpful. ;)
|
Goblin Squad Member
|
| 3 people marked this as a favorite. |
That is very basic, and hard to modify without adding new variables.
One of the biggest issues is probably what they do when threat zeros out. If any damage taken while they're on their way back to base could reset threat, that would remove some issues. Can it easily be reset before they return to "base?" Could base position be made larger, so that as soon as the creature re-enters the camp it recovers threat, even if it hasn't reached its original spot? If all creatures in the camp could share the entire camp as their "Threat base" that would help a lot, in getting all of them to aggro as soon as one is hit, and recover agro more quickly if there is still a problem.
|
Goblin Squad Member
|
Thanks, Stephen. That should help to focus the proposed solutions. I suspect that tweaking the spreadsheet values might be all that's needed to satisfy a lot of players.
My first-pass idea would be to leave the radius alone, slow down threat decay from 15 per round to 12 per round, and bump up the threat conversion rate to 55 or 60%. They're not huge changes, but they seem like they should interact to produce significantly more chase behavior from the monsters.
I suggest leaving the radius alone because hexes without much player activity are already pretty densely populated with monster camps. I'd like to retain the possibility of crossing hexes like that without drawing aggro from every monster in sight.
|
Goblin Squad Member
|
| 1 person marked this as a favorite. |
I think some of the issue comes with Kiting, and in this we really need the stamina situation to be fixed to fully see the ramifications of any changes that are made to the aggro system.
That being said, one the the major issues is that most of the group "leashes" back to their destination. I suggest that a small % of the threat generated by PC attacks only, not when the NPC attacks, be shared with the group fully. This should allow for the group to stay in combat and chase PCs longer.
Furthermore, while any PC is within 10 meters of an NPC I suggest that they generate threat. This will allow you to eventually outrun them, but if you go toe to toe in combat you are potentially generating more threat than those at Range. This would eventually allow for pseudo tanking, but not in the traditional since.
I think this would help combat a bit, and NPCs would seem a bit more intelligent, running after someone a bit longer, or actually turning and attacking the guy stabbing him in the back.
|
Goblin Squad Member
|
First thing: Monsters are just moving harvesting nodes in my book. Personally i see no reason changing anything.
Since this seems an important issue to many fellow players/testers, i will describe my observation with red skeletons. Before i start, i would suggest looking into their aggro mechanic. I had situtions where i hit one (no group, only solitary) with overdraw, it starts moving, and immediately resets. Sometimes they started rushing at me at second or third shot only.
1)Pull aggro with overdraw. Mob starts chasing.
2)Turning around running, using attacks that can be used on the run.
3)When my stamina gets low, my dps drops to a point you were describing resulting them in turning back.
4)Me chasing them, hitting.
5)Rinse and repeat.
Suggestion: Raise the aggro gain with each consecutive hit. This could offset the aggro loss because of low stamina. Making the "kiting race" at least longer, possibly not viable anymore.
Just my 2 coppers...
Ah ha. First thought, if you kill an enemy, threat of the whole group increases mucho.
Maybe even make the threat related to the amount of HP the creature had.
Having creatures converge on one player is no good, but having them switch target infinitely is worse. Is it possible to allow them to judge damage and seek out, collectively, the highest dps player? And, if smart, you would have to have the tank aggro until the others could unleash everyone for fear of breaking aggro and having them just chase the ranged. I've seen that in other games, but I don't know how it is implemented.
Can you have a collective group threat that tracks the amount of damage done to the group per player and go after that player?
There needs to be something between going after the first player and not locking in.
I still think a big issue is allowing melee to 'hit'. Even if melee at high level has some sort of rage shout half-stun that stuns you for .5 seconds if you are within longbow+ range, I feel them having made contact would result in stickier aggro.
I don't see this as an easy fix because at the very least, we want the mobs to have a plan of attack. Once we come up with that plan of attack, it needs to be implemented. It is going to require more advanced AI, altho not rocket science. Just more variables.
We need to punish groups for having all ranged. How? If we could guarantee that each group was properly diverse... which melee and all, one could easily make the melee threat based on range moreso than damage done and the ranged based on damage so that, but that may lead to range being separated. Could also make it so that ranged would never go past their own melee units. Or stay behind them in a certain range.
Could make them have a formation and push forward towards the attackers without assessing threat until they get within a certain distance.
I also think giving high level mobs a speed boost and more snares would help. If you engage a mob of ogres, you should have to, for the most part, fight.
Just throwing out ideas...
Edit: @ nodes. I still think there should be much of pve that is a node, but there should be high level pve... both guarding areas where there are valuable resources and allowing for one to get better stuff. Like if you wanna get a tier 3 recipe, you should have to kill a dragon or something, IMO.
|
Goblin Squad Member
|
| 2 people marked this as a favorite. |
I am completely fine with the AI as it stands at this stage of production. That said, if we do feel that mobs need to be better at pursuing their attackers, I would suggest:
A) Adding the ability for a mob to regain Threat immediately upon reaching 0. From what I can tell, they are currently unable to register threat until they return to their spawning point. This should prevent the Pull-Leash-Chase tactic from being possible, so the mob stays engaged as long as it receives more than 5 DPS.
B) Share 50% of gained Threat with creatures within 15 meters on each hit. (So, 25% of total damage.) This should make it easier for other players to draw aggro (in comparison to the previous experiences where the mobs focused down the initial player) while preventing a player from focusing on a single mob and having the remainder of the enemies return to camp, ignoring the fight. The remainder of the group should still return if the player is doing low damage, just above 5 DPS, but if they were doing significant damage then the entire group would respond to them as the greatest threat. This seems like it would be more challenging behavior to cope with, especially in reaction to attacks like Whirlwind.
C) Increase the Threat gained from 50% to 60% (for example) to prevent Killing-At-Range by Attrition. I have been able to repeatedly fire Distant Shot at an enemy (Risen Fighter Adventurers are my best example) and have them consistently lose Threat at a greater rate than I can increase it - therefore typically they will move about 5 meters before reaching 0 Threat and returning to their spawn point. As long as I continue attacking I will eventually be able to kill them without them ever coming to me, because (presumably) my DPS is lower than 5, but still great enough to prevent them from regenerating health. This would prevent those situations, if set to the correct percentage.
I do not know how difficult these would be to implement, but this is what comes to mind.
Edit: Also, thank you for giving us something else to talk about while waiting for the patch. (At least, that's what I'm doing...)
One of the things that makes this discussion very tricky, is that different people are trying to solve different problems with suggesting changes to the aggro mechanic.
If the issue is that kiting is currently too easy, shouldn't the change announced to ranged combat solve that issue? Also, we've been told a number of times that there are changes to ranged combat that will come in time, so again if kiting is the issue, there are lots of things in alpha that are unbalanced or not optimal that will be addressed over time.
If the issue isn't kiting per say, but that soloing large groups of mobs is too easy (picking them off one at a time in a way that causes the other mobs grouped with the one you're killing to lose aggro), then solving that problem has some other implications, for example making it considerably harder to solo at all because you will have to deal with multiple mobs. Is that the desired outcome?
Understanding exactly what problem we're trying to fix here seems to be the first step in trying to figure out what solution would be best.
|
Goblin Squad Member
|
I am not a programmer, but if this were real life (it's not) the group getting attacked by the adventurer should react in a realistic way.
1) The whole group latching on to one attacked is a hard nut. It makes sense, but it sounds like the big groups would be impossible so the current radius sounds like a reasonable alternative.
2) That said, if one of the group dies, the killed should get a big agro gain, since "That guy just killed john! Get him!" Humanoids would respond, but undead and animals would not care.
3) They should remember you for longer, and be keeping an eye out. After the initial fight, there should be a period where the NPC party "posts a watch" and remembers the attack. The agro radius would double for a time (maybe 3 minutes), or triple if an NPC died (for maybe 6 minutes?). That way a player could kill a wolf, run away, return and kill an archer, run away, etc... If the original instigator returned within 10 minutes he/she would get all agro. If the killer of a mob returned within 20 minutes, he/she would get all aggro.
This makes more sense for a humanoid group (humans, ogres, gobbos) than for a bunch of undead or wolves. Those groups might have a different (or the original) rule set.
|
Goblin Squad Member
|
Is there a way to change the initial threat the creatures get when you come within 20 meters to something that gets rechecked over time or is that tied to them being at their starting position? If a creature could gain aggro again after the first time just because you are within 20 meters then they won't behave like they want to run home, as it kind of feels they do now.
My concern isn't too easy vs too hard but just that the creatures are acting so dumb it 'feels' wrong.
|
Goblin Squad Member
|
| 5 people marked this as a favorite. |
Honestly, the AI should still not be a priority. Mostly the AI is there for something to grind and gain achievements, and for gear. Past that they have little purpose.
I would suggest, and take it as you will, that you work on the "other" features that go along with ranged combat.
Leave ranged attacks as they are, and when you get the other things implemented we can give you feedback on the changes.
That being (casts summon Being) said, if you want to make an easy number change that will affect PVE, increase the initial threat amount or decrease the threat loss.
Dont put much effort into this though. Any changes you make could burn new characters into the ground.
|
Goblin Squad Member
|
| 1 person marked this as a favorite. |
Also, if you're sure something got initial threat and gave up sooner than 10 seconds, that's also a bug.
I've seen this happen frequently. Most often, it seems to happen when there are significant z-axis differences between the mob and the player, or when there are significant deformations in the terrain between them. I'll find a spot and demonstrate it on twitch tonight if possible.
Creatures lose threat at a rate of 15 per round...
Can you help us understand the intent in having threat decay at all?
So please give us your thoughts...
I would like to consider the impact of the following changes:
- Change Notice Distance to 40 meters;
- Change Initial Threat to 15;
- Change Threat Decay to 5 per round.
The increased Notice Distance will make it more cumbersome to pick and choose targets, and make it impossible to one-shot one mob in a camp without the rest of the camp noticing. The decreased Initial Threat will make it easier for other players to pull aggro off the initial puller, which should make it more attractive to be in a group. The decreased Threat Decay means that mobs will be much less likely to quit chasing you while you're damaging them; and in combination with the decreased Initial Threat means that mobs will chase you for a minimum of 3 Rounds, which isn't terribly long, but is long enough to make it more difficult to run away from them in a crowded area.
Building off of Cheatle's post, we don't have the stamina stuff in right now. A minor tweak to the threat tables today could be a major unbalance to the game when Stamina goes in. Likewise for armor penalties to speed. Right now it is possible to run away from a mob if the player needs to; but later running away may be impossible if threat changes are loosened too much. So perhaps, right now, we should just think about possible changes within the context of the current, incomplete system?
That said, where running away is always a possibility (assuming the Bombers don't get you or the wolves don't gank you!), then what Caldeathe was saying in his first post is probably the thing to focus on. That and the way mobs will turn back after only coming halfway.
Mobs turning back early: I see a lot of this with level 1 mobs when I pull a larger member of the group from 35m range. Based on the tables, the simplest solution to this would be to either a) increase their initial agro meter, b) decrease their decay rate, or c) a mix of both. To figure out the correct rate, the math would basically be: how much time does it take for skeleton/goblin/oger/bandit/wolf to move 60 m (35m range + 15m for radius + 10m for the palyer backing away after the attack)? Set threat variables so that decay time to 0 threat = time to traverse 60m.
I don't see any way to fix a creature turning back while still being attacked just by messing with the tables, other than to increase the aggro conversion rate. That brings up a big question: Are the aggro conversion rates for player damage to the target and target damage to the player the same variable? Does changing one automatically necessitate changing the other?
Thank you for stepping in, Stephen.
|
Goblin Squad Member
|
| 2 people marked this as a favorite. |
I would like to consider the impact of the following changes:
- Change Notice Distance to 40 meters;
- Change Initial Threat to 15;
- Change Threat Decay to 5 per round.
this would make it pretty hard to move around without constantly catching the attention of the mobs. Many busy areas there isn't 80m between separate camps and you'd have no choice but to fight your way through the hex.
I would, however, like to see the sound of an attack going off instantly increase the notice distance to 40m.
Can you help us understand the intent in having threat decay at all?
The alternative would be to have a mob (or group) you walked too close to follow you 8 hexes until you got back to a town where the guards would take care of them for you.
At some point, it makes sense for a creature to 'give up' because it's not worth it. Now if you did considerable damage to a creature and killed its friend, it would retain that aggro much longer and chase you for some time (hopefully).
|
Goblin Squad Member
|
Creatures lose threat at a rate of 15 per round...Can you help us understand the intent in having threat decay at all?
So please give us your thoughts...I would like to consider the impact of the following changes:
[list]Change Notice Distance to 40 meters;
I believe threat decay is there so they do not chase you forever. New characters will have serious problems if they chase you across the map. (I have noticed that guards do not help you lol, pulled a mob while standing next to a guard)
The notice Distance to 40 will also make things difficult for new characters. They will get multiple mob aggro a lot. People would be forced to drag off archers from their spawn area, or they would need to deal with multiple mob aggro as well.
Then of course is the pure gatherers, which will not be able to do that without fighting mobs... but that doesnt bother me.
So please give us your thoughts...I would like to consider the impact of the following changes:
- Change Notice Distance to 40 meters;
The increased Notice Distance will make it more cumbersome to pick and choose targets, and make it impossible to one-shot one mob in a camp without the rest of the camp noticing.
I think you are going for a thought experiment here, but I'll toss this out there anyways.
KarlBob highlighted the downside of changing the notice distance: traveling through the terrain will be a lot more difficult, especially in areas where mobs are closely packed. Players will routinely find themselves drawing 2 or 3 mobs at once if they are not careful or the mob spacing isn't increased.
Honestly, the AI should still not be a priority.
I agree but for different reasons. Personally, I don't want to see yall putting any more attention on this issue right now than maybe screwing with the variables in the table. There is a world of other things the programmers have to get done that are far more vital to the success of this game and its status as MVP.
|
Goblin Squad Member
|
I would, however, like to see the sound of an attack going off instantly increase the notice distance to 40m.
Actually, that should be a penalty to the attacker, not a bonus to the attacked, since my shooting a mob should probably not cause the rogue hiding on the other side to suddenly lose their stealth.
|
Goblin Squad Member
|
I think increasing the notice distance to 40m might be too much. Right now, you can get past two camps of monsters as long as there is more than 40m between them. Increase that safe passing distance to 80m, and hexes without many adventurers will become perpetual combat zones. Gathering will become a two step process: kill all the monsters, then send in the gatherers.
I know the River Kingdoms are dangerous, but in my opinion, one should be able to walk more than a few yards without being jumped by a dozen monsters.
|
Goblin Squad Member
|
First, I was specifically limiting my suggestions to the variables that Stephen explained he could easily change.
Second, I was hoping that Stephen or another dev would expound on the more nuanced aspects of their intentions with Threat Decay. I understand that some decay is necessary if they don't have a distance-based leash.
Having it be more dangerous to run through mob infested areas does not strike me as a downside. The world is supposed to be dangerous, and groups acting cohesively are, I believe, intended to be more successful than solo gatherers.
|
Goblin Squad Member
|
Can you help us understand the intent in having threat decay at all?
The alternative would be to have a mob (or group) you walked too close to follow you 8 hexes until you got back to a town where the guards would take care of them for you.
At some point, it makes sense for a creature to 'give up' because it's not worth it. Now if you did considerable damage to a creature and killed its friend, it would retain that aggro much longer and chase you for some time (hopefully).
This sounds like what I would like to see different. If you run past mobs ,they chase you and give up , that's fine . But the same group of mobs seem to give up as fast even if you kill one of them from range and then run. Killing one should get you a lot more aggro somehow.
However if our ability to run far is going to change soon then any changes should wait for the stamina changes to be tested.
|
Goblin Squad Member
|
Perhaps, for the long run, the answer might be a threat modifier on the target, rather than the threatened. So that causing damage (regardless of who it's too) increases my threat modifier.
Thus, say instead of base 50 threat, we make it base 25 threat, but give me a +25 for having launched an attack. Then, everyone in my party who launches an attack successfully also gains a +25 or +15 vs the mob. Now, instead of the mob being focused only on the initial attackers, they have an option, based on the threat level modifier of all around them. If the +15 remains for a round or two, and is refreshed with every attack, then If I attacked first, but Nihimon attacked more often, the mob would start chasing me, but gradually shift interest to Nihimon as my +15 vanished.
|
Goblin Squad Member
|
"More dangerous" is one thing. Not being able to take two steps without triggering aggro from multiple camps is something else altogether.
I'd like to see some room left for bandits to set up ambushes for caravans without being hip deep in monsters. Or for strike teams in the War of Towers to concentrate on fighting each other, rather than the wildlife. Or for caravans to carry gods between settlements without fighting every monster they see along the way. To me, there is definitely such a thing as overly dense monster placement.
40m might be okay. Maybe the threshold of "impassibility" would be 50m. After the patch comes out, GW can adjust the spreadsheet values, and we'll find out.
|
Goblin Squad Member
|
For me, the density of monsters and bandits just sort of loitering in small groups everywhere in the hex is immersion-breaking. I would prefer:
1) Fewer PvE baddies. My minimap should not look like a pepperoni pizza.
2) Baddies not at campfires should be doing something. Preferably roaming around.
3) Rather than having threat "splash" limited to 15m, it should be unlimited, but with the amount of threat diminishing with distance. If I'm clobbering a couple bandits and their wolves, it shouldn't be ignored by those five archers in plain view who just happen to be 25m away. Combat should attract attention, and enough combat should be able to pull enemies in from a long, long way away.
|
Goblin Squad Member
|
| 1 person marked this as a favorite. |
Is there a plan to go back to 200 max stamina, or are we staying at 100? That is linked into the issue. Many dps rotations are stamina starved after a couple of rounds. After those first few rounds they are using stamina as it recovers (1 stamina/0.1s iirc).
As you said, to maintain threat a player must do at least 30 damage each round (on average). He gets 60 stamina/round. So to do that 30 damage he needs to do 0.5 damage per point of stamina in a longer fight (eg. against ogres).
Assuming roughly equal matchup in terms of attack bonus vs. ref/fort/will that's ~85% of max damage. Base damage minus resistance is 45-r at the moment. So damage done is ~0.85*f*(45-r)~(38-0.85r)*f where f is the damage factor. Damage per stamina (what we're concerned with for a stamina starved character) is ~(38-0.85r)f/s where s is the stamina cost.
Assuming for simplicity that it's only a single attack being spammed to maintain threat once stamina starved requires:
(38-0.85r)f/s > 30/60=0.5
r=0, Minimum f/s=0.013157894736842105
r=1, Minimum f/s=0.013458950201884253
r=2, Minimum f/s=0.013774104683195593
r=3, Minimum f/s=0.014104372355430182
r=4, Minimum f/s=0.014450867052023121
r=5, Minimum f/s=0.014814814814814815
r=6, Minimum f/s=0.015197568389057751
r=7, Minimum f/s=0.015600624024960999
r=8, Minimum f/s=0.016025641025641028
r=9, Minimum f/s=0.016474464579901153
r=10, Minimum f/s=0.01694915254237288
r=11, Minimum f/s=0.017452006980802792
r=12, Minimum f/s=0.017985611510791366
r=13, Minimum f/s=0.018552875695732836
r=14, Minimum f/s=0.019157088122605363
r=15, Minimum f/s=0.019801980198019802
r=16, Minimum f/s=0.020491803278688527
r=17, Minimum f/s=0.021231422505307854
r=18, Minimum f/s=0.02202643171806167
r=19, Minimum f/s=0.02288329519450801
r=20, Minimum f/s=0.023809523809523808
r=21, Minimum f/s=0.024813895781637715
r=22, Minimum f/s=0.025906735751295335
r=23, Minimum f/s=0.02710027100271003
r=24, Minimum f/s=0.028409090909090908
r=25, Minimum f/s=0.029850746268656716
r=26, Minimum f/s=0.03144654088050314
r=27, Minimum f/s=0.03322259136212624
r=28, Minimum f/s=0.035211267605633804
r=29, Minimum f/s=0.037453183520599245
r=30, Minimum f/s=0.04
Quick Shot: 0.13666666666666666
Basic Shortbow Strike: 0.13499999999999998
Burning Ray: 0.13444444444444445
Freezing Ray: 0.13444444444444445
Shocking Ray: 0.13444444444444445
Flare: 0.10666666666666666
Frost Ray: 0.10666666666666666
Wrathful Strike: 0.1032258064516129
Basic Longbow Strike: 0.102
Acid Dart: 0.10111111111111111
Melting Ray: 0.10111111111111111
Mental Ray: 0.10111111111111111
Midnight Ray: 0.10111111111111111
Thundering Ray: 0.10111111111111111
Basic Greatclub Strike: 0.09580645161290323
Basic Greatsword Strike: 0.09580645161290323
Basic Spear Strike: 0.09576923076923077
Shard Spray: 0.09555555555555556
Burning Flume: 0.094
Freezing Torrent: 0.094
Shocking Strike: 0.094
Grasping Shadows: 0.09000000000000001
Boiling Blood: 0.09000000000000001
Basic Dagger Strike: 0.08583333333333333
Melting Stream: 0.08560000000000001
Mental Strike: 0.08560000000000001
Midnight Strike: 0.08560000000000001
Thundering Torrent: 0.08560000000000001
Basic Club Strike: 0.08533333333333333
Basic Rapier Strike: 0.08533333333333333
Basic Longsword Strike: 0.08399999999999999
Basic Battleaxe Strike: 0.08347826086956521
Puncture: 0.08269230769230769
Sorrow's Release: 0.0816
Wraith's Cry: 0.08084745762711863
Slam: 0.0803225806451613
Wrath Guard: 0.0803225806451613
Point Blank Shot: 0.08
Half Draw: 0.07928571428571429
Shatter: 0.07774193548387097
Invisible Torrent: 0.076
Wind-Up: 0.07444444444444445
Figment Foes: 0.07333333333333333
Sweep: 0.0729032258064516
Slash: 0.072
Spike: 0.072
Slug: 0.072
Swing: 0.07129032258064516
Stab: 0.07115384615384615
Stagger: 0.07080645161290322
Precise Shot: 0.06999999999999999
Running Shot: 0.06999999999999999
Exhausting Burst: 0.06999999999999999
Pulverize: 0.06948717948717949
Round: 0.06777777777777777
Invisible Ray: 0.06777777777777777
Cacophony: 0.06689655172413793
Deadly Proximity: 0.06689655172413793
Thwack: 0.066
Searing Armor: 0.0652
Impale: 0.06462686567164179
Ruin: 0.06387096774193549
Cut: 0.06347826086956522
Chop: 0.06235294117647059
Bright Rune: 0.06222222222222223
Beat: 0.062000000000000006
Thrust: 0.062000000000000006
Penetrate: 0.06166666666666667
Transfix: 0.0611764705882353
Overdraw: 0.06063829787234043
Hack: 0.059565217391304354
Chill Touch: 0.05928571428571428
Rattle: 0.059166666666666666
Thump: 0.059166666666666666
Downburst: 0.05875
Lens of the Sun: 0.058571428571428566
Skewer: 0.05761194029850746
Cleave: 0.05709677419354839
Cross Blow: 0.05615384615384615
Sunder: 0.05565217391304348
Prime: 0.054000000000000006
Press: 0.054000000000000006
Winter's Caress: 0.053200000000000004
Patient Anchor: 0.05285714285714286
Carve: 0.05258064516129032
Hoist: 0.05161290322580645
Pinpoint Targeting: 0.0516
Distant Shot: 0.0516
Remove Energy: 0.0516
Compound: 0.05153846153846154
Followthrough Foresight: 0.051071428571428566
Whirlwind: 0.05064516129032258
Ghostly Grasp: 0.05
Mangle: 0.049782608695652174
Basic Spear Exploit: 0.04882352941176471
Hellflume: 0.048799999999999996
Impact Critical Shot: 0.04839285714285714
Understrike: 0.048
Fearsome Visage: 0.04785714285714286
Shillelagh: 0.04782608695652174
Basic Longbow Exploit: 0.04732142857142857
Inertia: 0.04666666666666666
Basic Greatclub Exploit: 0.04640625
Basic Greatsword Exploit: 0.04640625
Shank: 0.046
Deception: 0.045454545454545456
Basic Shortbow Exploit: 0.045294117647058825
Clustered Shots: 0.045294117647058825
Torporous Rune: 0.044444444444444446
Daze: 0.044444444444444446
Break: 0.044444444444444446
Eel's Touch: 0.04428571428571428
Shiv: 0.044
Killing Joke: 0.043750000000000004
Bullseye Shot: 0.0432
Parting Shot: 0.04303571428571429
Basic Dagger Exploit: 0.042916666666666665
Explosive Rune: 0.04290322580645162
Jarring Mesmerism: 0.04290322580645162
Mace Conk: 0.042666666666666665
Basic Club Exploit: 0.042666666666666665
Parry: 0.042666666666666665
Basic Rapier Exploit: 0.042666666666666665
Iron Gate Ward: 0.042666666666666665
Deadly Aim: 0.04258064516129032
Basic Battleaxe Exploit: 0.041739130434782605
Shadow Blast: 0.04161290322580645
Electric Brand: 0.04125
Arctic Winds: 0.0412280701754386
Deafening Roar: 0.0412280701754386
Basic Longsword Exploit: 0.04097560975609756
Touch of Darkness: 0.04097560975609756
Shove: 0.04066666666666666
Hew: 0.04048387096774193
Thousand Cuts: 0.04048387096774193
Hamstring: 0.04
Blinding Pattern: 0.03935483870967742
Shard Storm: 0.0392
Force Hail: 0.0392
Wilting Surge: 0.03916666666666667
Shadow Ray: 0.03888888888888889
Hammer Conk: 0.03866666666666666
Instant Inferno: 0.0368
Distracting Shot: 0.035833333333333335
Club Conk: 0.034666666666666665
Frightening Blast: 0.032903225806451615
Windrider: 0.032400000000000005
Fan: 0.03166666666666667
Blinding Ray: 0.03166666666666667
Snap Shot: 0.03166666666666666
Pommel: 0.03125
Shield Slam: 0.029583333333333333
Telekinetic Fist: 0.029583333333333333
Shear: 0.027903225806451614
Passing Step Thrust: 0.027903225806451614
Shield Bash: 0.0275
Lightning Arc: 0.02666666666666667
Smash: 0.02661290322580645
Destructive Smite: 0.025925925925925925
Binding Web: 0.025416666666666667
Multi-Shot: 0.024705882352941175
Nightmarish Foe: 0.023333333333333334
False Edge: 0.02333333333333333
Haunting Mists: 0.022799999999999997
Gore: 0.022745098039215685
Holy Lance: 0.021666666666666667
Impaling Shot: 0.020806451612903225
Forehand: 0.02
Bull Rush: 0.01888888888888889
Lunge: 0.01875
Balestra: 0.018666666666666668
Shield Charge: 0.018
Fire Bolt: 0.017777777777777778
Footwork: 0.015833333333333335
Stop Cut: 0.015000000000000001
Carrion Strike: 0.015
Quick Frost: 0.0148
Plague Storm: 0.014642857142857141
Escape: 0.013666666666666666
Icicle: 0.013333333333333332
Corrosive Rain: 0.0124
Wyrmling Breath: 0.012
Sonic Thrust: 0.010476190476190476
Storm's End: 0.01
Razing Smite: 0.01
Steal Speed: 0.0062499999999999995
Shield Block: 0.0
Protective Ward: 0.0
Mage's Resistance: 0.0
Slippery Ground: 0.0
Concealing Mist: 0.0
Frighten: 0.0
Energetic Field: 0.0
Strength Surge: 0.0
Agile Feet: 0.0
Dispelling Touch: 0.0
Bleeding Touch: 0.0
Battle Rage: 0.0
Resistance: 0.0
Guidance: 0.0
Virtue: 0.0
Bless: 0.0
Minor Cure: 0.0
Augur: 0.0
Defending Weapon: 0.0
Dazing Touch: 0.0
Doom: 0.0
Inspiring Word: 0.0
Liberation: 0.0
Bane: 0.0
Charge: 0.0
Trip: 0.0
Lookout: 0.0
Combat Expertise: 0.0
Bulwark: 0.0
Evasion: 0.0
Resiliency: 0.0
Feint: 0.0
Tumble: 0.0
True Strike: 0.0
Shield of Faith: 0.0
Divine Favor: 0.0
(Data for attacks grabbed from your google docs Official Wiki data last night.)
Admittedly this is a bit simplified in that it ignores secondary effects and is only single attack spamming, but it should show that many attacks should have no problem holding aggro against single targets while others will lose it. Obviously focusing on a single target should make the other creatures drop aggro in the current system.
In my experience I've lost aggro aggro against ogres a few times and I'm still trying to reconcile that with the above numbers.
|
Goblin Squad Member
|
| 1 person marked this as a favorite. |
...set up ambushes for caravans without being hip deep in monsters.
Imagine if monster-density becomes part of detecting ambushes. "Look out, folks, they're pretty thin here, maybe someone's cleared them out!"
Merchants could keep their hired guards from becoming bored by charting their paths from Point A to Point B through the areas most-populated by monsters :-).
|
Goblinworks Game Designer
|
| 1 person marked this as a favorite. |
Summaries of what I've seen so far and answers to questions at the end.
Suggestions for stuff that requires programming, so I'll ask them how hard it is:
- Creatures continuing to respond to stimuli when in "return home" mode after reaching 0 threat (I'm not sure if they're actually completely ignoring stimuli right now when returning home).
- Sharing fractions of threat generated on hit with creatures in radius (rather than sharing the whole threat list periodically).
- Disable threat decay for a set period after gaining new threat.
- Reduce threat decay rate based on situation (e.g., current HP or whether any enemies are still in perception range).
- Constant threat generation just for being in perception range rather than just an initial spike.
- Increase damage to threat conversion ratio the longer the creature has been in combat or by a fixed amount for each discrete attack taken while in the current combat.
- Share a big burst of threat (with allies in proximity) to the killer when killed.
- Increase variables like perception range and initial threat amount for a short period after returning home.
- Set threat share radius to drop off over distance (so it may be set big enough to pull multiple groups, but only briefly at the edges).
Suggestions for stuff I can change:
- Threat gain rate on damage: Increase slightly
- Threat decay rate: Slow down slightly, Slow down significantly
- Initial on-perceived threat: Increase slightly, Decrease to keep similar time on non-damaged follow on threat decay changes, Set with decay rate to make sure they'll at least go 60m or so before giving up since that's how far people usually kite back
- Perception range: Double, Definitely don't double we'd accidentally pull everything! ;)
- Ally notification range: It's fine, Increase if it can fall off in magnitude over distance
Answers:
@Shadowbeast:
Yes, the intention is that soloing should not be possible for large groups of approximately your level or above. You should be able to take small groups that are your level or slightly weaker, and larger groups as you progressively outlevel them, but we did not intend to create a situation where you can bite off winnable chunks from groups that would be deadly to a solo character that engaged them in melee.
@Nihimon:
Threat decay is to prevent them from following you indefinitely, while allowing them to follow you if the fight is still ongoing. It's in place of a leashing mechanic that sets a hard "X meters from home, keep fighting, X+1 meters, give up and go away even if one side was about to win" (which is why we're particular about not calling it leashing ;) ).
@Nightdrifter:
We're happy to keep tweaking Stamina numbers. 200 seemed like it was too much, as it seemed to be allowing a lot of people to kill all their targets with burst damage such that they didn't even notice Stamina was a thing. 100 might not be enough. I'd love to see another thread comparing and contrasting peoples' experiences at the two different Stamina maxes. The intent is to leave Stamina regen at where it is now, so raising the total mostly just affects how many secondaries you can use before being fixed to your regen rate.
|
Goblin Squad Member
|
| 4 people marked this as a favorite. |
1.) There are several things much more important to be working on now (like companies and banks).
2.) There are several things much more important to be working on now (we were only posting about this while we were waiting for companies and banks).
3.) Reprogramming AI will not make monster farming a completely different and fulfilling experience.
4.) Changes to AI now could have wild implications when more important systems go in later (stamina, ammunition, armor speeds and encumbrance, etc.)
If we need to do something right now to satisfy a large bloc of the community on the issue, I favor Karlbob's idea of tweaking the decay lower and conversion rate a little higher, which doesn't distract programmers and can easily be undone later when more systems go in.
|
Goblin Squad Member
|
| 2 people marked this as a favorite. |
Personally, I don't want to see yall putting any more attention on this issue right now than maybe screwing with the variables in the table. There is a world of other things the programmers have to get done that are far more vital to the success of this game and its status as MVP.
This.
The general theory of the AI is pretty neat and it's in the game and working. Obviously there are many more complexities and tweaks to put in over time, but this is one phase of the game that is ready for EE.
A bunch of other stuff, isn't.
1.) There are several things much more important to be working on now (like companies and banks).
2.) There are several things much more important to be working on now (we were only posting about this while we were waiting for companies and banks).
3.) Reprogramming AI will not make monster farming a completely different and fulfilling experience.
4.) Changes to AI now could have wild implications when more important systems go in later (stamina, ammunition, armor speeds and encumbrance, etc.)If we need to do something right now to satisfy a large bloc of the community on the issue, I favor Karlbob's idea of tweaking the decay lower and conversion rate a little higher, which doesn't distract programmers and can easily be undone later when more systems go in.
+1 (except for the 3rd item)
|
Goblin Squad Member
|
| 1 person marked this as a favorite. |
If you increase threat duration don't at the same time increase aggro radius.
Adjust the ratio of spawns so that there are more duos and trios. People are essentially forced to attempt disproportionate pulls because the number of appropriate-sized spawns are too few.
Please check with the programmer whether it will be do difficult to release mobs from their spawn points. If the concern is that the mobs will be pulled away from resource nodes it should be possible that when proximity to other groups is detected then they will move away from the detected group as well.
Ultimately I'd like to see escalation spawns migrate into other hexes, and even engage in NPC vs NPC combat when they meet an unallied escalation.
|
Goblin Squad Member
|
| 1 person marked this as a favorite. |
Adjust the ratio of spawns so that there are more duos and trios. People are essentially forced to attempt disproportionate pulls because the number of appropriate-sized spawns are too few.
I actually agree with that. There are not enough spawns appropriately sized for a fresh character if they aren't using kiting techniques. Not even around Thornkeep.
Maybe the spawns as they are now are appropriate some places but around starter towns there needs to be more small/easy spawns and fewer spawns you have to kite to solo.
At least until it because easier/more practical to group all the time.
|
Goblin Squad Member
|
| 1 person marked this as a favorite. |
Imagining there will always be someone close to group with is, I think, utopian. I doubt there will always be a way to avoid soloing and still play (unless you are only intending to craft that session).
If you are on at an odd hour and the people already in-game are already at max group you are gonna be SOL without a way to fend for yourself.
Goblinworks Executive Founder
|
I've got lots of suggestions for how to make large groups more interesting by altering AI. I'll leave those out.
How hard would it be to change the aggro gained from a percentage of damage done to a flat amount per attack?
How much work to make the decay happen only towards a character that hasn't attacked? To change the rate of decay based on distance from spawn?
|
Goblin Squad Member
|
Imagining there will always be someone close to group with is, I think, utopian. I doubt there will always be a way to avoid soloing and still play (unless you are only intending to craft that session).
If you are on at an odd hour and the people already in-game are already at max group you are gonna be SOL without a way to fend for yourself.
One hopes most companies, settlements, or settlement alliances will encourage people to split a group of 6 and run a group of 3 and a group of 4. It all depends on which is more beneficial, I imagine, 6 + 1 or 4 + 3.
|
Goblin Squad Member
|
Imagining there will always be someone close to group with is, I think, utopian. I doubt there will always be a way to avoid soloing and still play (unless you are only intending to craft that session).
If you are on at an odd hour and the people already in-game are already at max group you are gonna be SOL without a way to fend for yourself.
It's not impossible. GW2 pulled it off pretty well IMO with their quest linked to areas rather than individuals.
However given the size of the map I think you are right when refering specifically to PFO. Most players will be soloing at some point in time and a system that expects people to constantly be grouped just won't work.
My desire for an ordered universe over-rode me.
Group Play vs. Solo Play thread here:
|
Goblin Squad Member
|
There just needs to be more duos and trios type mobs, instead of these half dozen to dozen groups.
Each Duo/Trio should have at least 1 Ranged unit, so as to not make it easy mode.
A majority of your time will be with a group, or at least 1 other person, if you are going out by yourself and you get PKed/Ganked while Soloing mobs, you are going to think twice. I agree though that some people will be soloing, and that people will probably do some of their achievements solo.
|
Goblin Squad Member
|
There just needs to be more duos and trios type mobs, instead of these half dozen to dozen groups.
Each Duo/Trio should have at least 1 Ranged unit, so as to not make it easy mode.
A majority of your time will be with a group, or at least 1 other person, if you are going out by yourself and you get PKed/Ganked while Soloing mobs, you are going to think twice. I agree though that some people will be soloing, and that people will probably do some of their achievements solo.
I dunno. I wonder if what you're seeing in "these half dozen to dozen groups" is that the soloers and 2 man teams are removing the 2- and 3-mob groups, and leaving the big groups. The big groups don't despawn and eventually the hex is nothing but big stuff, which encourages us to stop soloing and group up.
We've run 6-character parties and killed everything in a hex we came across. The second and third time we ran around the circle we have a good healthy mix of small, medium, and large group mobs.
There's value in a few duo and trios w/o a ranged unit. They're nothing but a speed bump for a large group, but a reasonable target for soloers and beginners.
|
Goblin Squad Member
|
I can solo groups up to 4-5 as long as there is no yellow leader in the group if I am wearing heavy armor and have a decent weapon. In my sweat suit with broom handle I started with I was lucky to kill one mob before having to run away and sometimes didn't get away, and my sweat suit got dirty and my broom handle was degraded. What made me very sad.
But that does lead me to believe we were able to wipe the small groups while we were all running solo, leaving the large groups intact. Running in a party will make the large groups less of a problem.
|
Goblin Squad Member
|
EE day one: at the Korova Milkbar Thirsty Ogre, we'll sharpen our senses before we hit the streets countryside for some good laughs and lashings of the old ultra violence.
Tabletop miniatures of new PFO characters (black starter hats not yet implemented in game)
|
Goblin Squad Member
|
No idea how hard that would be to implement, but how about this:
-if mobs get attacked while returning to there camp a new aggro cycle is started.
-so mobs get 50 thread again, again shared by other mobs in 20m.
that way, if you just try to run away, that is still possible, but if you keep attacking the mobs you might have a problem.