|
just got round to applying ILSpy to the beta the CharacterStats class has 4,500 lines with 7 levels of indentation e: to be fair some of that will be due to compiler optimiziations (i hope). a lot isn't though, unless the C# compiler is way better than i think it is coffeetable fucked around with this message at 14:48 on Sep 14, 2014 |
#
?
Sep 14, 2014 14:34
|
|
|
|
| # ? Apr 25, 2024 04:58 |
|
|
For the people who never took higher mathematics at school (me) it'd be a lot easier if there were more diagrams and graphs posted in this thread. PoE Statistical Mechanics: Effect of Attributes per point
I can't really add much to this discussion on statistics but looking at Dexterity it needs to do more, as while gives a slight edge in accuracy, it feels really boring. The problem with choosing Perception and Resolve at character creation for the newcomer is that it's hard to gauge how effective or game-changing investing in this Attribute is going to be. You can see the effects of Might, Constitution and Dexterity in the character screen while you can grasp what Intellect does in the field, but it's hard to tell how important +21% Interrupt or Concentration is as there's no context. You don't see what stats it exists in relation to. Looking back, Perception wasn't terribly useful in New Vegas. EDE had a perk which gave you 10 Perception for all intents and purposes and most people only invested in the attribute to get the Better Criticals perk, which is probably bad design. Pillars has it better as Attributes, Skills and Talents don't have any interplay regarding requirements, lessening the need to meta-game. Inspector Gesicht fucked around with this message at 15:46 on Sep 14, 2014 |
|
#
?
Sep 14, 2014 15:14
|
|
|
Here's our first (not current) proposal Might: Damage and Healing Con: Endurance and Health Dex: IAS (Increased Action Speed) Per: Accuracy and Interrupt Int: AoEs and Durations Res: Deflection and Concentration That should be easier to understand as its two offensive: Might and Per, Two defensive: Con and Res and two Universal: Int and Dex My new proposal will keep the elegance and symmetry of that while removing interrupt, balancing Accuracy against Might, Dexterity against Intellect and Resolve against Concentration. My theory for the first set was on the money, so let's see if the math supports my theory again. FTR: Josh's revised version Might: Damage and Healing Con: Endurance and Health Dex: IAS (Increased Action Speed) Per: Accuracy Int: Deflection and AoE Res: Duration and Concentration
|
|
#
?
Sep 14, 2014 15:25
|
|
|
Sensuki posted:Here's our first (not current) proposal Oddly enough that does seem to fit the names of the statistics a bit better. This has probably already been discussed but IAS might be problematic to balance.
|
|
#
?
Sep 14, 2014 15:27
|
|
|
Nah it's not because of DT. Attack Speed is horrible against armor. As stated in the paper, against armored enemies, any increase in IAS contributes the same percentage to effective DT as you are opposing it every time you attack. Balancing IAS against Intellect shouldn't be too difficult if we use a similar method to our "buff seconds" exercise in the paper. This same problem (IAS vs DT) is affecting the weapon balance at the moment as well. My main problem with Josh's one is Deflection and AoE, I don't think that works, but Durations and Concentration are both good together. Sensuki fucked around with this message at 15:55 on Sep 14, 2014 |
|
#
?
Sep 14, 2014 15:34
|
|
|
You don't think AoE and Deflection are good because it makes it too powerful or they don't seem like a natural fit or...? To me, Int sounds fine, and even makes some sense if you played a character with Expertise in DnD 3e before or used your intelligence modifier for AC in 4th edition.
|
|
#
?
Sep 14, 2014 16:06
|
|
|
I think AoE is fairly trivial because it can be accounted for most of the time by positioning and proper placement and timing of spells. The area difference of a Fireball at 3 Int and 18 Int, while being about double the area isn't actually that big, most of the time that will not matter too much, even with a mouse wheel scaling AoE size, or foe only AoE increases. So on the majority of casters if I am going to be a backline caster, I don't care about Deflection really - I would if I wanted to play more forward and aggressive *maybe*. Durations and Concentration together is a superior choice IMO. It also only makes a "difficult choice" note the inverted commas, for support casters. DPS and Tank classes would have a clear best attribute - Might for DPS, Con for tanks. Our Resolve is actually pretty equal with Constitution but separating Deflection and Concentration makes the Deflection attribute weaker in terms of survivability IMO. This is magnified by the fact that PE is currently tuned so that unit Accuracy is higher than Deflection most of the time, or very close and this is the realm where Constitution is only slightly worse than Deflection for survivability or much, much better.
|
|
#
?
Sep 14, 2014 16:37
|
|
|
Sensuki posted:Nah it's not because of DT. Attack Speed is horrible against armor. As stated in the paper, against armored enemies, any increase in IAS contributes the same percentage to effective DT as you are opposing it every time you attack. Balancing IAS against Intellect shouldn't be too difficult if we use a similar method to our "buff seconds" exercise in the paper. I haven't seen you talk about this, but IAS isn't just a simple % boost to DPS. High IAS means you're less likely to be interrupted by the enemy and more likely to interrupt them since your action resolves before theirs does & the window of opportunity to interrupt your action is shorter. There's also a lot to be said for getting the first strike in an encounter: if you do enough damage with it, the enemy may never get their chance to react. You used a Dota analogy earlier here, so here's one for you: which spell would you rather have for ganking weaver, Venge's magic missile or Rubick's lift?
|
|
#
?
Sep 14, 2014 16:39
|
|
|
Unlike DotA the animation speed is not increased with IAS until Recovery time is zero. So it would not help at getting the first strike in, but the second - sure. We didn't do any math relating IAS impact on Interrupts but sure it does theoretically give you more of a chance to score them and less chance to receive them. For raw DPS though, Might is always better. Interrupts don't give you DPS. However there may be situations where high IAS and high Concentration prevent more DPS loss than Might grants, but at a guess I would say that in the long term Might would always be better.
|
|
#
?
Sep 14, 2014 16:40
|
|
|
Oh, I had misunderstood your design then, I thought you intended for IAS to affect the front swing as well. That actually takes care of one the problems I had with the idea, which is that if such a system would be applied to enemies, it would make it exceedingly difficult to judge how fast an enemy can execute a powerful attack or a spell.
|
|
#
?
Sep 14, 2014 16:55
|
|
|
You can thank rope kid for that actually. As well as making the frontswing cancellable, but not the backswing  But RMB cancel is not working atm like it was in v257bb 
|
|
#
?
Sep 14, 2014 16:57
|
|
|
So uh... how about that fantasy adventure game, huh? I look forward to playing Man With A Sword, and perhaps hitting things with that sword! I know the details of the plot have been kept quiet to maximize how fresh the game is on release, but I'm starting to think maybe the game could use a little more marketing in the "meet the setting" kind of way, or a little more teasing about what the game's supposed to be about. So many names and so much text got thrown around in the demo even just in character creation that I found it hard to grapple on to just who was who. Is our character just some wandering adventurer? Is there some kind of big war or cataclysm going on? I saw some talk in the game about kids being born without souls, is that a local problem or is that The Big Problem? Basically what's the pitch about the game here to get excited about besides it being an Obsidian game so it's probably a good bet?
|
|
#
?
Sep 14, 2014 17:43
|
|
|
Sensuki posted:That's a good point. I don't think Monks have an AoE
|
|
#
?
Sep 14, 2014 18:12
|
|
|
Ah okay. The description doesn't mention that Torment's Reach is an AoE, just that it increases their range. What level is Rooting Pain supposed to be ?
|
|
#
?
Sep 14, 2014 18:15
|
|
|
Yeah Torment's Reach has a confusing description and had a weird implementation. It's a cone AoE now. I can't remember what level Rooting Pain is. Somewhere in the mid-levels. It goes off automatically when the monk takes a Wound and causes a very short duration foe-only Stun to hit everyone around.
|
|
#
?
Sep 14, 2014 18:27
|
|
|
Torment's Reach sounds much better. Have you swapped Rooting Pain in recently? I think it's higher than level 8 in the current build.
|
|
#
?
Sep 14, 2014 18:40
|
|
|
Dolash posted:So uh... how about that fantasy adventure game, huh? I look forward to playing Man With A Sword, and perhaps hitting things with that sword! As someone who backed the project just a few weeks ago, basically just on the developer's reputations, what I'm gathering from the last few pages of this thread is that it's like Eve Online, but in a fantasy kingdom. Spreadsheets in Shannara! The Dark Lord, on his Dark Graph . . . I kid, I kid.
|
|
#
?
Sep 14, 2014 18:55
|
|
|
Currently, my personal opinion about rope kid's newly devised one is that, incentive-wise, I have mixed feeling when I'm taking INT (Deflection + AoE). As Sensuki posted, for something like Sea Otter fucked around with this message at 21:34 on Sep 14, 2014 |
|
#
?
Sep 14, 2014 20:39
|
|
|
Sea Otter posted:Currently, my personal opinion about rope kid's newly devised one is that, incentive-wise, I have mixed feeling when I'm taking INT (Deflection + AoE). As Sensuki posted, for something like Barbarian, probably, it's quite attractive but, for Fighter or non-front-liners, I'd rather like either bonus than the both. Sensuki+Matt version has this "elegant" thematic division of Offense/Defense/Utility bonuses, which makes the players feel easier to pick abilities depending their preferences although, as rope kid pointed out, that makes INT too easy choice for caster classes. I'm having a hard time seeing a serious problem with putting either deflection with AoE on INT. I can't really think of a class (or a build) that can't benefit from either AoE OR Deflection. Whereas, with AoE and durations together, I can think of a few builds (e.g. a passive fighter build) where INT could be completely dumpable. The battle in my head right now is between the symmetric simplicity of Sensuki's version and the interesting complexity of Josh's modification.
|
|
#
?
Sep 14, 2014 20:58
|
|
|
Ladolcevita posted:Whereas, with AoE and durations together, I can think of a few builds (e.g. a passive fighter build) where INT could be completely dumpable. This really isn't a problem though? There's a difference between an attribute not being very attractive to a very specific build/playstyle and an attribute not being attractive to a class.
|
|
#
?
Sep 14, 2014 21:01
|
|
|
Fair Bear Maiden posted:This really isn't a problem though? There's a difference between an attribute not being very attractive to a very specific build/playstyle and an attribute not being attractive to a class. Yeah, not really. That's why I'm pretty evenly split on the two proposals. I think either would work.
|
|
#
?
Sep 14, 2014 21:11
|
|
|
Ladolcevita posted:I'm having a hard time seeing a serious problem with putting either deflection with AoE on INT. I can't really think of a class (or a build) that can't benefit from either AoE OR Deflection. Whereas, with AoE and durations together, I can think of a few builds (e.g. a passive fighter build) where INT could be completely dumpable. The battle in my head right now is between the symmetric simplicity of Sensuki's version and the interesting complexity of Josh's modification. There's always going to be builds where you will be able to dump at least one attribute. Here's how I saw both suggestions affecting each class (Just IMO/Quickfire thoughts that I PM'd to Matt): quote:
Not everyone will agree on that though.
|
|
#
?
Sep 14, 2014 21:13
|
|
|
Hieronymous Alloy posted:what I'm gathering from the last few pages of this thread is that it's like Eve Online, but in a fantasy kingdom. Spreadsheets in Shannara! The Dark Lord, on his Dark Graph
|
|
#
?
Sep 14, 2014 21:53
|
|
|
Ladolcevita posted:Yeah, not really. That's why I'm pretty evenly split on the two proposals. I think either would work. As a side note, following some recent comments of rope kid in the official boards, which seem not to be currently accessible by rope kid, I've gotten an impression that he may be thinking of changing how AoE works. If this is the case, this may change the attractiveness of AoEs for offensive casters. Also, Rangers seem to still require lot of works before the players get the hang of how actually this class plays out. Sea Otter fucked around with this message at 22:17 on Sep 14, 2014 |
|
#
?
Sep 14, 2014 21:57
|
|
|
I understand all of this theorycrafting, but from everything I've read and followed from Josh up until this point stat distribution has been focused on not creating 'ideal builds' but rather on making most stats viable for all classes. Sure, you can make INT powerful so that it improves spellcasting builds, but then why would you choose anything except INT? That feels kinda contradictory to what they're trying to do with stats in PoE. I want to be able to make a Wizard that is the most accurate sonaofabitch in the world, so he's going to get an 18 in PER. Or maybe I want a fighter that's all about speed, landing a hundred blows in a few seconds, so he'll have a high DEX. I wouldn't want to have interesting or unique stat placements (ugh, hate the term 'builds') end up useless 5 levels in because what I should have done was put everything into INT.
|
|
#
?
Sep 14, 2014 22:02
|
|
|
Hieronymous Alloy posted:As someone who backed the project just a few weeks ago, basically just on the developer's reputations, what I'm gathering from the last few pages of this thread is that it's like Eve Online, but in a fantasy kingdom. Spreadsheets in Shannara! The Dark Lord, on his Dark Graph . . . I kid, I kid. No, you're right; some of this talk is getting really weird and masturbatory.
|
|
#
?
Sep 14, 2014 22:53
|
|
|
Ropekid made a new entry in Tumblr about Ability Scores. However, he managed to confuse me form the beginning. quote:It’s more important to me that players be able to make a wide variety of characters of a given class with different Attribute spreads than for those Attribute spreads to be equally viable in all circumstances/all different class styles (e.g. close-up gish wizard vs. back row wizard throwing out big AoEs). In any case, as far as my understanding goes, he seems to like to let the players to build various characters, not relying too much on Feats (Talents in PoE) or other special/back-end rules. Of course, if he can do that, I don't have any problem with it but...well, could you kindly do something about INT first, rope kid? :P
|
|
#
?
Sep 14, 2014 23:19
|
|
|
There is a whole lot of armchair game design going on in this thread nowadays, and I guess that's fine since it's a beta, but people need to remember what the stated goals of this project are vs. their own preconceived notions of what things should be.
|
|
#
?
Sep 14, 2014 23:28
|
|
|
ceaselessfuture posted:There is a whole lot of armchair game design going on in this thread nowadays, and I guess that's fine since it's a beta, but people need to remember what the stated goals of this project are vs. their own preconceived notions of what things should be.
|
|
#
?
Sep 14, 2014 23:37
|
|
|
ceaselessfuture posted:There is a whole lot of armchair game design going on in this thread nowadays, and I guess that's fine since it's a beta, but people need to remember what the stated goals of this project are vs. their own preconceived notions of what things should be. I'm a bit jealous in that I wish I had the time/energy for deep dives into PoE's design. Sort of wondering if Sensuki has a day job.  But I think it's cool. Nothing wrong with more discussion. I imagine Rope Kid would happily admit perfect design did not spring fully formed and unchangeable from his brain.
|
|
#
?
Sep 14, 2014 23:47
|
|
|
Sea Otter posted:Ropekid made a new entry in Tumblr about Ability Scores. I think he's saying that he wants all classes to viable on some form no matter which attributes you chose as long as you play to the strength of the character your attributes give you. Basically you should be able to make a character where the reason you spend points on attributes is because you want the character to be a certain way from a roleplaying/story point of view and not be screwed because of it. Int being AoE and Deflection fits that perfectly. No matter what class you pick, your smart character is going to be viable.
|
|
#
?
Sep 14, 2014 23:56
|
|
|
@DatonKallandor Thanks for the explanation and I also think the basic abilities should be balanced like that. Then again, if I image traditional and relatively weak back-row Wizard, the added Deflection goes against that image. Or should I start with the image of smart character in general? Personally, I'd do that in more simulationist games but that's not what I think when I'm creating a character for gamist one, especially, knowing the combat is a large part of this game. Reading the entry through my simulationist glasses, I can see the reasoning more clearly, though. Thinking about that now, I've gotten an impression that the character creation progress also intentionally avoiding the stats behind decisions (such as weapon choices). The devs may like to let the players choose their character backgrounds more based on the role-playing context.
|
|
#
?
Sep 15, 2014 00:20
|
|
|
DatonKallandor posted:I think he's saying that he wants all classes to viable on some form no matter which attributes you chose as long as you play to the strength of the character your attributes give you.
|
|
#
?
Sep 15, 2014 00:34
|
|
|
ceaselessfuture posted:There is a whole lot of armchair game design going on in this thread nowadays, and I guess that's fine since it's a beta, but people need to remember what the stated goals of this project are vs. their own preconceived notions of what things should be. But yeah, it's not as if this wasn't going to happen, and if a stink rises up past this point then it too was always going to happen. I wonder if the data is malleable enough that you couldn't just mod your way into your ideal system.
|
|
#
?
Sep 15, 2014 00:36
|
|
|
@rope kid Sorry if my post get in your nerves but, yeah, while I'm interested in different builds in terms of party chemistry in combat gameplay, the other part of me also wants to see PC-NPC interactions supported by Obsidian writers (Now it feels odd to see that I have totally forgotten about my other side, concentrating on just the gamist element). In any case, I appreciate what you are doing to satisfy various needs, related with this project.
|
|
#
?
Sep 15, 2014 01:47
|
|
|
I know I've asked this before, but can Chanters and Ciphers learn new abilities outside of leveling like Wizards? Somehow the whole notion of a Ciphers will always be foreign to me, as they're the only class with no real analogue in the IE games, but given the in-universe reason that's probably intentional.
|
|
#
?
Sep 15, 2014 02:39
|
|
|
Basic Chunnel posted:I wonder if the data is malleable enough that you couldn't just mod your way into your ideal system. It should be, I've actually already taken a look it - recompiling will be the tricky part.
|
|
#
?
Sep 15, 2014 08:00
|
|
|
How deep is the trap mechanic? I mean the sort of traps that you find throughout the dungeon. The trap mechanic in the IE and NWN games is easily one of the weakest aspects of the games. They just slow down the gameplay and dungeon exploration unnecessarily and don't really add any challenge, fun or strategy. In a game with clunky party control (like NWN2), they are a source of frustration.
|
|
#
?
Sep 15, 2014 08:23
|
|
|
FRINGE posted:You were joking, but that is something that I definitely dont want. I'm fairly sure absolutely none of this will be required knowledge for leveling a character in PoE, despite this thread becoming theorycrafting.txt
|
|
#
?
Sep 15, 2014 08:31
|
|
|
|
| # ? Apr 25, 2024 04:58 |
|
|
I don't think this can be so easily handwaived as theorycrafting since the head honcho of PoE is directly engaged in its discussion.
|
|
#
?
Sep 15, 2014 08:36
|
|
