Putting the 'role' back in role-playing games since 2002.
Donate to Codex
Good Old Games
  • Welcome to rpgcodex.net, a site dedicated to discussing computer based role-playing games in a free and open fashion. We're less strict than other forums, but please refer to the rules.

    "This message is awaiting moderator approval": All new users must pass through our moderation queue before they will be able to post normally. Until your account has "passed" your posts will only be visible to yourself (and moderators) until they are approved. Give us a week to get around to approving / deleting / ignoring your mundane opinion on crap before hassling us about it. Once you have passed the moderation period (think of it as a test), you will be able to post normally, just like all the other retards.

A Critique of Diablo 2-Style Skills & Trees

Joined
May 31, 2018
Messages
2,863
Location
The Present
I've been contemplating playing Diablo 2 Resurrected lately. I don't want to revisit it after 20+ years unless I try a unique build. Luckily there are a wealth of skill planners online, which has caused me to remember why D2's skill system, and those like it, irk me. There are a wealth of problems and they show up most prominently in caster characters will large amounts of active skills.

o1632848729085487.jpeg

  1. Costs: Sunk & Taxed
    The player has to commit to a skill in order for it to be worthwhile. The player either has to suffer near term by delaying character developing and hoarding points for the skills they intend to develop, or by permanently hamstringing their character by developing skills they don't want while they level up to unlock the tree. These sunk costs are also a major mismatch with skill obsolescence that occurs. Furthermore, patching rebalances can even destroy certain builds.
  2. Skill Obsolescence
    Early selections in many trees are not viable long-term and have no practical use even in normal difficulty once the player reaches Act 3. This is far more pronounced for caster classes, but can encumber martial classes too. For example, the Barbarian Masteries which allows the player to specialize in a class of weapon. For a game build around random loot drops, this is terrible design.
  3. Discourages Experimentation
    These lead to the player needing significant meta-knowledge or "burner characters" where they discover what works and what they enjoy. Any skill which is does not precede a skill they do want becomes strongly discouraged against. That many skills don't advance on a percentage basis will also punish players for valuing versatility, as they will be eclectic weaklings. While each class does have a few "one point wonders", the necessity to optimize for greater difficulties will shoe-horn players into fully investing into only a few skills.
Mastery Skills and Synergies

Originally, mastery skills were something that provided passive bonuses to other skills. In forgoing an active ability, the player could enhance a whole class of abilities or styles on their character. In Patch v1.10, Diablo 2 attempted to solve the skill tax, sunk cost, obsolescence, and skill point hoarding problems by giving skills synergy bonuses. Every skill now attained niche mastery qualities. The results are uneven. Some classes benefit more than others. Some builds no longer have skill hoarding problems. New issues were created though.

By shifting the difficulty equilibrium to account for the optimization power synergies could bring, synergies became a mandatory part of character building. This further reduces build styles, as it commits the player to skills they may not have desired to use. The logical consistency of some bonuses also creates redundancy on characters. The Sorceress is a prime example. Whereas Firebolt was an speed-bump on the way to Fireball, it is now necessary to the Fireball's full development. It solves the skill point hoarding issue, but obligates the character to a skill that will never against get used after Level 11. Other examples of these abound. While skills similar to each other granting synergies is logical, it produce undesirable results.

Proposed Solutions

To fix skill trees like Diablo 2's, I have a few ideas:
  1. Fundamentally similar skills are consolidated, but evolve into the higher order skill after additional investment. For example, The Sorceress's Firebolt would become Fireball. The Amazon's Cold Arrow would become Freezing Arrow, then Ice Arrow. Etcetera.
  2. Mastery skills and individual synergies are eliminated. Instead, each tree gets a global passive bonus for each skill point invested in that tree. For example, every 5 skill points in the Paladin's Offensive Auras gives a +1 skill point equivalent bonus to all unlocked skills in that tree.
  3. Base potency is not based primarily on skill investment, but tied to the character's attributes. In Diablo 2, the best derivative for this is probably character level.
These changes would allow greater build diversity that doesn't obligate or pigeon-hole the player into narrow builds. It allows developers to offer a wider array of skills to each class also. Specialists in a given tree would still be rewarded with the extremes, while generalists would not become impotent for having versatility. It would, for the most part, eliminate sunk costs, skill taxes, and remove the need to hoard skill points. Player would also no longer be penalized for experimentation, as all skills would remain forever relevant and useful.
 
Last edited:

Gargaune

Arcane
Joined
Mar 12, 2020
Messages
3,623
Much as I'm enjoying the game again now, I agree that it's not a great skills system and I've run into a bunch of issues for precisely the reasons you're describing. I would pick a bone with one of your solutions, though:

Base potency is not based primarily on skill investment, but tied to the character's attributes. In Diablo 2, the best derivative for this is probably character level.
I get why you're suggesting this, but I'm wary of progression systems offloading too much competency to the XP bar alone, and Diablo 2 already has plenty of that going on via loot.
 

octavius

Arcane
Patron
Joined
Aug 4, 2007
Messages
19,685
Location
Bjørgvin
Many of the problems are not real problems when you can Respec and have Shared Stash.

As for Firebolt vs Fireball, as long as you don't have enough mana to spam Fireball at will, Firebolt has its use. Pumping up Firebolt instead of Fireball may be an idea in order to keep the cost of Fireball down. So the Synergies system have both good and bad points IMO.
 

luj1

You're all shills
Vatnik
Joined
Jan 2, 2016
Messages
15,167
Location
Eastern block
The player has to commit to a skill in order for it to be worthwhile.

Nope, there are many 1-pt wonders

Barb - Leap, Leap Attack, Find Item, Find Potion, Grim Ward, practically the whole Passive tree...
Paladin - Charge, Holy Shield (works well with partial investment), Redemption, Salvation...
Assassin - Cloak of Shadows, Burst of Speed, Fade, Shadow Master, Shadow Mastery, some traps, Masteries...
Necro - Attract, Confuse, Revives, Life Tap, IM, Clay Golem, a ton of other curses...
etc., etc.
 
Joined
May 31, 2018
Messages
2,863
Location
The Present
The player has to commit to a skill in order for it to be worthwhile.

Nope, there are many 1-pt wonders

Barb - Leap, Leap Attack, Find Item, Find Potion, Grim Ward, practically the whole Passive tree...
Paladin - Charge, Holy Shield (works well with partial investment), Redemption, Salvation...
Assassin - Cloak of Shadows, Burst of Speed, Fade, Shadow Master, Shadow Mastery, some traps, Masteries...
Necro - Attract, Confuse, Revives, Life Tap, IM, Clay Golem, a ton of other curses...
etc., etc.
I agree that one point wonders exist. I had acknowledged that. Those skills support my points. Their value is in their function, like Barbarian Leap or Necromancer Fear curse. That's very different from say, Holy Bolt or any damage skill where degree matters. Furthermore, those utilities are nice to have. They are not necessary. Some of them are prime for combination, like Leap and Leap Attack. The PC could leap if they select the ground, or leap attack if they select a monster. Other skills you mentioned are % based, which scale no matter the difficulty. Being able to put 1-point into just about every Necromancer curse is a major asset the class and demonstrates why my critiques are valid.

I wasn't aware of respecialization is a feature now. That helps, but I think that too supports my points. A well designed system doesn't need this--certainly not multiple.
 
Last edited:

luj1

You're all shills
Vatnik
Joined
Jan 2, 2016
Messages
15,167
Location
Eastern block
Well according to you, 1) all skills should be equally good, 2) all skills should be effective with low investment, and 3) the player should be able to experiment without metagaming.

That's pretty fucked up. Are you Sawyer?
 
Joined
May 31, 2018
Messages
2,863
Location
The Present
1) Never claimed or even insinuated that. 2) Never claimed that, merely that skills shouldn't be taxes or become obsolete. 3) Yes. It's a game (see: toy) not a contract.

What I'm stating is that D2's style of skill advancement is flawed. Reread my OP to understand why. This style of skill system leads to traps and waste that doesn't make it better as a game. Character building is about facilitating the way the player plays the game, not act as a mine field to be navigated. Diablo 2 has a lot of wheat, but also an abundance of chaff. With greater consideration, it could be much better while losing nothing. Even something as simple as advancing skills through drops of skill books (Atma in Act II) would mitigate most of its oversights. It would be more like Diablo 1 and better fit the franchise's slot-machine mechanics. It would be no different than grinding for gear.
 

Silvanus

Novice
Joined
Aug 15, 2024
Messages
68
OP, what are your thoughts on the progression of something like Path of Exile, in which passive bonuses are provided by the skill tree while activity abilities are gained elsewhere?
 

Gargaune

Arcane
Joined
Mar 12, 2020
Messages
3,623
luj1, I think the simple way to look at it is that in Diablo 2, absent Synergies you kinda have to either respec or "endure" until your top skills open up. For instance, once you open up the Paladin's Fanaticism, there's really no point in having any extra levels in Might, so if you can respec those points into the former, you should. Compare that to something like 3E - when I'm playing NWN I never have to think about respeccing, it's not like I've got five levels in Cleave and find myself thinking "boy, if only I could reallocate those to Great Cleave." The closest you get would be spells like Sleep, whose utility peters out at higher levels due to the HD cap, and specifically for Sorcerers, who can only unlearn so many spells.

Synergies themselves come across as a bit of an aftermarket bodge to address the issue - Blizzard weren't about to redesign the whole system, so here's a thing that means your early, expressly suboptimal point investments can still pay dividends as you keep leveling and focus on your top skills. A shortcut with mixed results, if I understand correctly, as some players were driven to min-max that aspect as well, whether for PvP or the pure personal fuck of it.

Ultimately, it boils down to Diablo 2 being designed for that Normal to Nightmare to Hell progression, each with a free respec. But the skill tree's built for a "single" playthrough, covering levels 1-30, and if you're not into this original "New Game Plus" approach it's a bit of a downer.
 

ferratilis

Arcane
Joined
Oct 23, 2019
Messages
2,904
The problems OP outlines were "fixed" in Diablo 3 by making every skill/passive available whenever, and we all know how that turned out. D2 skill trees are fine as they are imo. Maybe the presentation is not that good because everything looks drab, but when it comes to progression, it's done well.

As for presentation, GD has done it better imo, and it also fixed OP's "skill obsolescense" issue by making you invest into mastery tab to unlock skills.

03VIFAW.png
 

luj1

You're all shills
Vatnik
Joined
Jan 2, 2016
Messages
15,167
Location
Eastern block
For instance, once you open up the Paladin's Fanaticism, there's really no point in having any extra levels in Might

I dig what you're saying, but I don't have a problem with Fanaticism being stronger than Might

Fanaticism grants less raw damage than Might btw, so for builds that don't require IAS - Charge Paladin (Abbot), Blade Fury Sin (uses FCR not IAS) and old WW - it's still an interesting choice
 

Silvanus

Novice
Joined
Aug 15, 2024
Messages
68
The problems OP outlines were "fixed" in Diablo 3 by making every skill/passive available whenever, and we all know how that turned out. D2 skill trees are fine as they are imo. Maybe the presentation is not that good because everything looks drab, but when it comes to progression, it's done well.

As for presentation, GD has done it better imo, and it also fixed OP's "skill obsolescense" issue by making you invest into mastery tab to unlock skills.
Titan Quest had essentially the exact same system. It's also worth mentioning that both games have access to repeated respecs, which is invaluable for experimenting with different builds.
 

Gargaune

Arcane
Joined
Mar 12, 2020
Messages
3,623
Fanaticism grants less raw damage than Might btw
Wait, am I missing something? Fanaticism should have a higher damage bonus than Might for the player, but lower for the party. Going off on a tangent, I know, but I'm playing Diablo 2 right now so it's pertinent to my interests.
 

luj1

You're all shills
Vatnik
Joined
Jan 2, 2016
Messages
15,167
Location
Eastern block
Fanaticism grants less raw damage than Might btw
Wait, am I missing something? Fanaticism should have a higher damage bonus than Might for the player, but lower for the party. Going off on a tangent, I know, but I'm playing Diablo 2 right now so it's pertinent to my interests.

Nevermind I just checked Arreat Summit, it's something like +370% ED compared to +230% ED on Might which is definitely insanely too much

I got it mixed up with PD which successfully addressed that
 

Butter

Arcane
Patron
Joined
Oct 1, 2018
Messages
8,616
I don't presume to have the perfect solution. It should be noted that skill trees were pretty new when Diablo 2 shipped, so it's forgivable if such an early implementation wasn't quite there.

My "solution" would be to get rid of skill trees entirely:

* Every level the player gets a specialization point. There are 3 categories for each class to specialize in (roughly corresponding to the existing skill trees), and the more you specialize in one category, the more expensive it becomes to continue doing so.
* Abilities are items that you have to discover in the world (pretend runes don't exist and call them that). You must bind a rune to your character in order to get its ability, and removing it from the character destroys it.
* The power of a rune scales with how many points you've invested into its specialization category.
* The number of runes you can have bound to a character at one time scales with character level. Perhaps rare equipment could allow you to bind additional runes.
 

ciox

Liturgist
Joined
Feb 9, 2016
Messages
1,387
What's weird is Diablo 2 already has support for buying skills without investing in a previous ability. There can be skills that aren't connected to anything else. This property just wasn't used too often, though it could have, making skill development more freeform.

EOJRRgB.png


Diablo 2 has a good amount of band-aids to its design that make the skill trees work well enough. It's just that it's possible to imagine a way to maybe do it with a system that needs less band-aids. I think there's some merit to the ability evolution concept in the OP but I'm not sure I like solutions where the system holds your hand and invests points or otherwise allocates power for you though, it's pretty fundamental for RPGs that you should be allowed to make mistakes when making a character.

Also, for Might aura it could be cool if it added some fixed damage, letting you kill weak enemies faster than the more advanced damage auras, if it scaled properly with difficulty somehow then it would always be useful.
You could also tweak the auras a little more so it's obvious that Might is about scoring 1-hit kills, Concentration is about consistent damage output and Blessed Hammers, and Fanaticism is less about damage and more about triggering on-hit procs and elemental damage more often.
 

Faarbaute

Arbiter
Joined
Mar 2, 2017
Messages
826
A large part of the fun of playing Diablo 2 way back then, was that journey of discovery, of trying out different skills, figuring out what works and what doesn't, what's weak and what's strong, what's a trap and what's not, attempting to game the system, all that good stuff.

None of that would have been meaningful, none of it would have mattered, without these elements that, on their face, only seem to cause frustration, but are in fact vital to making character building matter at all.

You have to be able to fail. You have to be able to make mistakes. This is not a rollercoaster ride, and gameplay is not merely a formality.

The only critique I agree with is that the introduction of skill synergies was a mistake, because they were an attempt to fix something that wasn't a problem in the first place. (In response to a critique much in the spirit of this thread.)
 

Castozor

Augur
Joined
Nov 12, 2014
Messages
202
Being able to fail would be a lot more noble a design goal if you'd be able to respec somewhat more frequently though. In D2 it often came down to just rerolling your char if you weren't in the knows about the one respec a run. Also a bit unsure about what you mean exactly, "trap"/poor builds I agree with, finding out what works well and doesn't is part of the fun. Trap skills however should just not excist.
 

Faarbaute

Arbiter
Joined
Mar 2, 2017
Messages
826
Being able to fail would be a lot more noble a design goal if you'd be able to respec somewhat more frequently though. In D2 it often came down to just rerolling your char if you weren't in the knows about the one respec a run. Also a bit unsure about what you mean exactly, "trap"/poor builds I agree with, finding out what works well and doesn't is part of the fun. Trap skills however should just not excist.
Success shouldn't be guaranteed. Even less so, an optimal outcome. You could always just tough it out with your non optimal build. And if not, I'm fine with having to start over.
 
Joined
May 31, 2018
Messages
2,863
Location
The Present
The game is when you're adventuring and combating monsters. Builds are your chosen method to do that. They're tools for the fun. A means, not an ends. You guys act like a hammer isn't worth using unless the grip is covered in sharp spikes that you have to suffer through.
 
Joined
May 31, 2018
Messages
2,863
Location
The Present
The part you're referring to is the application of the character to slay the monsters and resolve the quests. Combat, exploration, and questing. That's the gameplay. The character is the means to do that. The game isn't determining what tools presented to you are false or trying to minimize traps built into the system.

Lord forgive them. They know not what they do.
 

As an Amazon Associate, rpgcodex.net earns from qualifying purchases.
Back
Top Bottom